Adolescents’ Emotion Regulation Strategies Questionnaire–Extended: Further Development and Associations With Mental Health Problems in Adolescence

Gustaf Rådman; Benjamin Claréus; Daiva Daukantaitė

doi:10.1177/10731911231164619

Adolescents’ Emotion Regulation Strategies Questionnaire–Extended: Further Development and Associations With Mental Health Problems in Adolescence

Rådman, Gustaf; Claréus, Benjamin; Daukantaitė, Daiva 2024-03-01 00:00:00 Emotion regulation (ER) is implicated in a range of psychopathologies and behavioral problems that are prevalent or have their initial onset in adolescence. In this study, we aim to evaluate the psychometric properties (factor structure, internal consistency, and construct validity) of the Adolescents’ Emotion Regulation Strategies Questionnaire–Extended (AERSQ-E), a modified and extended version of an ER instrument developed by Zhou et al. Across six sub-studies using data from different Swedish adolescent community samples (1,104 students in total), we generated and validated a 23-item version containing six subscales: rumination/negative thinking, positive reorientation, creative expression, aggressive outlet, social support, and distraction. Assessing test–retest reliability, internal consistency, measurement invariance as well as convergent and discrimi- nant validity, we could establish, with some limitations, the general reliability and validity of the AERSQ-E as a valid measure of ER strategies for use in adolescence. Keywords emotion regulation, adolescence, psychopathology, psychometrics, questionnaire Emotions are intimately bound to regulatory processes behaviors (Aldao et al., 2010). Some examples include that continuously monitor and adjust their expression non-suicidal self-injury (NSSI) (Andover & Morris, (Yih et al., 2018). Emotion regulation (ER) describes 2014; Wolff et al., 2019); eating disorders and disordered these extrinsic and intrinsic processes that monitor the eating (DE) (Brockmeyer et al., 2014; Dingemans et al., demands of the current environment, evaluate what con- 2017); mood and anxiety disorders (Hofmann et al., stitutes a contextually appropriate response, and modify 2012); and aggression (Roberton et al., 2012; Sullivan emotional reactions accordingly (Thompson, 1994). et al., 2010). Hence, ER is crucially important for adaptive function- Many of these psychological and behavioral prob- ing in that it helps modulate the intensity and temporal- lems have their initial onset in adolescence, a period of ity of an emotional response in proportion to current rapid change in, not least, the underlying structures sup- events and in accordance with personal goals (Aldao, porting ER (Ahmed et al., 2015). Thus, identifying func- 2013). tional and dysfunctional patterns of ER in this critical ER theoretically implies both the regulation of nega- period is of high clinical relevance, as is by consequence tive and positive affect. However, dysfunctions in the the need for valid and reliable instruments to measure regulation of negative affect are particularly linked to ER, particularly in adolescence. Although numerous psychopathology (Beauchaine, 2015; Cole et al., 2017; McLaughlin et al., 2011) and have been suggested as a Lund University, Sweden transdiagnostic factor in the development and mainte- nance of both internalizing and externalizing psycholo- Corresponding Author: gical problems (Aldao et al., 2016). Difficulties in Daiva Daukantaite, _ Department of Psychology, Lund University, Box 213, regulating negative affect have been implicated in a wide 22100 Lund, Sweden. range of specific psychopathologies and destructive Email: daiva.daukantaite@psy.lu.se 2 Assessment 00(0) self-report instruments have been developed to assess undoubtedly important. Nevertheless, regardless of age, ER (e.g., Garnefski et al., 2001; Gratz & Roemer, we often turn to others for help with regulating our emo- 2004; Grob & Smolenski, 2005; Gross & John, 2003; tions, and research on the use of such interpersonal ER Hofmann et al., 2016; Hofmann & Kashdan, 2010; strategies has seen increasing research attention in recent Phillips & Power, 2007; Zeman et al., 2001; Zhou et al., years (e.g., Dixon-Gordon et al., 2015; Marroquı´n, 2011; 2020), the majority of these have been developed for Nozaki & Mikolajczak, 2020; Zaki & Williams, 2013). In measuring ER in adults or children, with only a few spe- the AERSQ, at least one subscale falls within each of these cifically targeting an adolescent demographic. categorizations. In this study, we sought to evaluate the psychometric In addition, ER strategies can be categorized as adap- properties of a further developed instrument, the tive or maladaptive, in the sense that they are either Adolescents’ Emotion Regulation Strategies Questionnaire effective or ineffective in modifying emotions and/or (AERSQ), a scale tailored to measure ER strategies used they are associated with either long-term negative or to regulate negative affect in adolescents (Zhou et al., positive outcomes (Aldao et al., 2010). Some ER strate- 2020). We sought to improve its psychometric properties gies are broadly characterized as maladaptive (e.g., and to extend it by adding and validating an additional expressive suppression or experiential avoidance) and ER strategy measure—aggressive outlet—which may be others as adaptive (e.g., acceptance or reappraisal). especially important for adolescents who self-injure Although evidence supports such a distinction at least as (Daukantaite _ et al., 2019; Tang et al., 2013). As part of a useful heuristic (e.g., Scha¨ fer et al., 2016), it is impor- the validating procedure for this extended version tant to note that no strategy is inherently one or the (referred to as AERSQ-E), we also explore the associa- other. Instead, adaptive ER is arguably best understood tions between adolescents’ ER strategies and problematic as the appropriate and effective use of ER strategies con- behaviors such as NSSI and DE, as well as mental health sidering the context, which can include specific aspects problems across several community samples of Swedish of the culture one finds oneself in, as well as personal adolescents. goals both immediate and long-term (Aldao, 2013). Conversely, maladaptive ER or emotion dysregulation can either imply a lack of successful regulatory action Classification of Emotion Regulation (e.g., rumination) or the inappropriate and ineffective use of ER strategies in considering personal goals (e.g., Strategies aggressive behavior such as lashing out at someone, There are many mechanisms by which we regulate our despite valuing their friendship). For example, distrac- emotions and several dimensions by which we concep- tion has an ambiguous status in that when combined tually classify ER strategies. The relationship between with an attitude of acceptance rather than avoidance, ER strategies and psychopathology varies depending on distraction can be supportive of positive psychological the specific strategy (Aldao et al., 2010). Thus, a key tar- development (Wolgast & Lundh, 2017). The AERSQ get of ER research is the mapping of ER profiles and aims to cover dimensions that are typically viewed as how they relate to positive and negative indices of men- both maladaptive and adaptive. tal health. This, in turn, requires tools that assess a broad range of ER types. One way to differentiate between different ER strate- Emotion Regulation Development and gies is based on the processes they primarily engage in. Psychopathology in Adolescence For instance, some ER strategies are more oriented toward changes in cognition (e.g., cognitive reappraisal, Developmental research suggests that children initially rumination) and some more oriented toward behavioral rely on caregivers to regulate their emotions and that changes (e.g., eating, workout, substance use, expressive they gradually come to internalize their ER abilities suppression, or aggression). Some ER strategies primarily across early childhood (Eisenberg et al., 1998; Kopp & rely on intrinsic or intrapersonal processes (e.g., distrac- Neufeld, 2003). These abilities further develop as chil- tion, expressive suppression), and some rely primarily on dren reach adolescence (Zeman et al., 2006), suggesting extrinsic or interpersonal processes, that is, regulation that as children mature, they gradually become more through social engagement with others (e.g., seeking con- skillful at regulating their own emotions. Despite this, solation, social modeling; Hofmann et al., 2016; Phillips adolescence is a period characterized by greater emo- & Power, 2007). Historically, intrinsic ER have been over- tional instability and negative affectivity (Larson et al., emphasized in ER research (McRae & Gross, 2020), and 2002), and many psychopathologies implicating ER further research into understanding the development of tend to emerge during adolescence (Kessler et al., 2005; these intrinsic ER strategies across development is F. S. Lee et al., 2014; Paus et al., 2008). ˚ Radman et al. 3 Some studies have highlighted important differences adults. This is problematic because adolescence differs in ER strategy use that characterize adolescence as com- from both earlier childhood and adulthood across a range of cognitive and self-evaluative capabilities, which pared with other age groups. Cracco and colleagues suggest that the same set of questions may be interpreted (2017) looked at the frequency of use for seven typically differently by adolescents than the intended age group adaptive and five typically maladaptive ER strategies (Zeman et al., 2007). For instance, the Emotion and found evidence suggesting that in adolescence there Regulation Questionnaire (ERQ; Gross & John, 2003) is a shift toward more frequent use of maladaptive ER was originally developed for adults and adapted for use strategies. Zimmermann and Iwanski (2014) looked at with children and adolescents by Gullone and Taffe adolescents’ and adults’ subjective beliefs about their (2012). It includes items such as ‘‘When I’m worried regulation of negative emotions such as anger, fear, and about something, I make myself think about it in a way sadness. They found that individuals aged 13 to 15 that helps me feel better,’’ which require relatively elabo- report having a smaller repertoire of ER strategies at rate self-evaluative knowledge about the interrelation their disposal than both preceding and succeeding age between cognition, emotion, and behavior. Although groups. The combination of a general shift toward more difficult in practice, it is crucial that measures tailored maladaptive types of ER with a general reduction of the for adolescents reduce such complex meta-cognitive ER strategy repertoire available suggest a particular vul- requirements to a minimum, as such capabilities may nerability with regard to emotional symptoms during be insufficiently developed (Casey et al., 2008). At the adolescence that can in part be accounted for by changes same time, adolescents, compared to younger children, in ER during the period. tend to exhibit advancements in areas such as abstract The social reinforcement of ER is an additional, and hypothetical thinking, logical reasoning, and potentially interacting, factor, as the use of many ER information-processing efficiency (Steinberg, 2005). strategies is necessarily linked to a social context. Therefore, it is appropriate to expect them to handle at Cultural factors, for instance, are likely to have an influ- least moderately more complex inquiries. To achieve ence on the adaptiveness of different ER strategies. This the fine balance that meet adolescents at their own includes how expressions of emotion tend to be inter- level, we must design instruments specifically tailored preted differently comparing men and women (Barrett & to their unique set of cognitive and metacognitive abil- Bliss-Moreau, 2009), which in turn might reinforce the ities, which has been a key focus of the AERSQ. use of specific regulatory strategies. Correspondingly, In addition, most of the work on ER both in adults studies have documented gender differences in the use of and in young samples emphasize intrinsic ER strategies specific ER strategies (e.g., Johnson & Whisman, 2013; (McRae & Gross, 2020). This can be seen in that many Nolen-Hoeksema & Aldao, 2011; Tamres et al., 2002) of the most popular self-report instruments today only and linked differences to the prevalence of various psy- assess intrapersonal ER strategies, such as the ERQ, the chopathologies such as anxiety, depression, and alcohol Difficulties in Emotion Regulation Scale (DERS; Gratz abuse (e.g., Bender et al., 2012; Nolen-Hoeksema, 2012). & Roemer, 2004), and the Cognitive Emotion However, the exact role of gender in the link between ER Regulation Questionnaire (CERQ; Garnefski et al., and psychopathology is not entirely understood (for a 2001). An exception is the Interpersonal Emotion discussion on the subject, see Nolen-Hoeksema, 2012). Regulation Questionnaire (IERSQ; Hofmann et al., 2016), which explicitly target extrinsic ER. However, the IERSQ was not developed with adolescents specifically Measuring ER in Adolescence and the in mind. Moreover, it only measures extrinsic ER and Rationale Behind the AERSQ must therefore be combined with scales targeting intrin- Given the importance of ER in adolescent psycho- sic ER (e.g., the DERS and CERQ) for an encompass- pathology, it is important that we develop valid mea- ing assessment of ER. The Regulation of Emotion sures of ER specifically tailored toward adolescents. In a Questionnaire (REQ; Phillips & Power, 2007) and review by Adrian and colleagues (2011) that examines FEEL-KJ (Grob & Smolenski, 2005) are perhaps the the assessment of ER, it was estimated that 44% of ER most comprehensive alternatives that have been specifi- studies that focused on middle childhood (age 6–12) cally developed for adolescents. REQ encompasses four deployed self-report measures, whereas 92.6% of studies subscales dividing ER strategies into internal-functional, that focused on adolescents (age 13–18) deployed self- internal-dysfunctional, external-functional, and exter- report measures. Clearly, self-report has become the nal-dysfunctional. However, as previously discussed, the method of choice in measuring ER among adolescents. functionality of ER strategies is partly dependent on However, a large portion of these studies used self- contextual factors rather than the type of ER strategy report instruments originally developed for children or alone (Aldao, 2013; Bonanno & Burton, 2013), and the 4 Assessment 00(0) AERSQ, in contrast to REQ, aims to delineate between during adolescence in the study by Cracco and col- strategies based on their presumed underlying processes leagues (2017), implying that it is an especially relevant rather than their presumed functionality. In addition, measure to capture when studying adolescent ER. the REQ is yet to be extensively validated, with the origi- Finally, previous studies have suggested that aggressive nal sample only consisting of 225 adolescents (12-19 tendencies in adolescence—including experiencing or years) with exploratory and confirmatory factor analy- regulating emotions such as anger toward oneself, self- ses performed on the same sample (Phillips & Power, hatred, or anger toward others—is associated with enga- 2007). FEEL-KJ is a comprehensive measure compris- ging in NSSI (Boxer, 2010; Brunner et al., 2007; ing 12 subscales that assess 15 ER strategies. Although Daukantaite _ et al., 2019; Fliege et al., 2009; Sourander showing promising psychometric properties in a valida- et al., 2006; Tang et al., 2013), making it an especially important measure for those who self-injure, with NSSI tion study (Cracco et al., 2015), it is a lengthier instru- showing the highest rates of lifetime prevalence during ment containing 90 items. In contrast, the AERSQ adolescence (Swannell et al., 2014). encompasses a wide range of ER types while maintain- ing a succinct format. This can be beneficial to projects that encompass a wide array of different questionnaires, The Current Study where brevity is an important consideration. In this study, we aim to evaluate the psychometric Finally, although some ER measures show good psy- properties (factor structure, internal consistency, and chometric properties when applied to adolescent sam- construct validity) of a modified and extended ples, the number of strategies assessed across these version of the AERSQ, referred to as the Adolescents’ measures is limited. For instance, the ERQ only assesses Emotion Regulation Strategies Questionnaire—Extended positive reappraisal and expressive suppression (Gross (AERSQ-E). As part of the validation process, we also & John, 2003). The CERQ, although developed for ado- intended to establish the convergent and discriminant lescents, focuses on cognitive ER strategies such as validity of the AERSQ-E by exploring the associations acceptance, rumination, catastrophizing, and positive between adolescents’ ER strategies and NSSI, DE, as refocusing (Garnefski et al., 2001). The AERSQ encom- well as other aspects of mental health including interna- passes dimensions of ER captured by some of these lizing problems (e.g., emotional symptoms, depression, instruments (e.g., rumination, positive reorientation, and anxiety), externalizing problems (e.g., peer problems distraction) but also includes at least one dimension of and conduct problems), and positive functioning (life ER not assessed by other instruments that we are aware satisfaction). We predicted that the associations for sub- of, namely, the use of expressive/creative behaviors (e.g., scales existing in the previous version would align with writing down thoughts about felt emotions) to cope with those previously found (see Zhou et al., 2020). That is, we negative affect (assessed by the subscale previously expected rumination/negative thinking to be positively called ‘‘cultural activities,’’ here renamed ‘‘creative associated with constructs related to negative functioning expression’’). Across the different ER strategy measures, and negatively associated with indicators of positive func- it is more common to include inhibitions of emotional tioning (i.e., life satisfaction), while we expected the oppo- expression (e.g., Gross & John, 2003; Hofmann & site pattern for positive reorientation and social support. Kashdan, 2010; Zeman et al., 2001), and few instru- Furthermore, we expected the creative expression subscale ments focus on actualized expression of emotions. Art- and the distraction subscale to only show weak or no based activities have previously been linked to positive associations to other variables. Finally, for the new theo- mental health and to ER in particular (e.g., Geipel et al., retical subscale aiming to capture aggression as an ER 2018; Saarikallio, 2010; van Lith et al., 2013; Zhao et al., strategy, we expected to find positive associations to most 2016), making it an important factor to consider. negative constructs including NSSI, anxiety, and depres- In a further developed version of the AERSQ, we sion, as well as various psychological difficulties including aimed to add a second expressive subscale intending to conduct, emotion, and peer problems, which would corro- capture aggression as an ER strategy. This inclusion was borate the status of aggression as a maladaptive type of motivated in several ways. First, when reanalyzing data ER and the previously discussed evidence linking it to from earlier iterations of the AERSQ, such a factor engagement with NSSI. emerged in the larger item pool. Second, aggression as an ER strategy has been successfully included in some previous ER instruments targeting adolescents such as Method FEEL-KJ (Grob & Smolenski, 2005) and REQ (Phillips Short Overview & Power, 2007), suggesting its relevance to the target age group. Third, the use of aggression to regulate affect The AERSQ presents respondents with a list of possible was among the maladaptive ER strategies that peaked behaviors and ways of thinking and asks respondents to ˚ Radman et al. 5 judge on a 5-point scale ranging from 1 (never)to5(very In the remaining three samples (4, 5, and 6), the psy- often) how often they engage in each item whenever they chometric properties of the final version of the AERSQ- feel ‘‘sad, disappointed, nervous, afraid, or experience E were evaluated. This validation took place both other negative or distressing feelings.’’ Item development among adolescents enrolled in junior high school (age of the original measure described in a recent article by ranging from 13 to 17, M = 14.42; Samples 4 and 5) age Zhou et al. (2020) combined a theory-driven approach and high school (age ranging from 16 to 20, M = age with feedback given from a pilot sample allowing ado- 17.10; Sample 6), in total covering an age range from 13 lescents to provide their own examples of ER strategies to 20 years. For external validity analyses, data from used. This resulted in a final 25-item version identifying Sample 2 and Sample 3 were also included wherever five factors, including rumination/negative thinking, data were available. positive reorientation, communication, distraction, and cultural activities. Based on the original AERSQ scale and new theoreti- Respondents and Recruitment cal considerations, an extended and modified version Data came from samples spread across several schools consisting of 33 items was first generated. Specific to this and municipalities in the southern part of Sweden. extended version, we generated five items aiming to cap- According to data acquired from Statistics Sweden ture a theoretical construct termed aggressive outlet, dated December 31, 2021, the average income of each reflecting the tendency to regulate emotions using forms municipality’s adult population ranged from below to of aggression. We also modified the item list of each sub- above the Swedish average (239,664 SEK/year vs. sam- scale to improve the conceptual separation of factors ple range 218,553–286,105 SEK/year). Adult education and in some cases, the interpretability of the items. This level was similar to or higher than Sweden as a whole led to a relatively large change to the overall item list for (46% vs. sample range 44%–65% having a university the new version compared with the old one (a lengthier education). The municipalities in which data were col- discussion of this procedure including motivation for lected were slightly more urban (87.6% vs. sample range each item change can be found in the supplementary 90.9%–96.6% living in urban areas). material under ‘‘Revising the scale’’). During this pro- The data collection spanned two separate projects: cess, some factors were reinterpreted and given new one focused on self-harm in adolescence and one names so that the terminology would more closely focused on self-control during adolescence. Thus, data reflect the presumed underlying function. For instance, were collected across several samples of Swedish adoles- the factor previously called cultural activities was cents aged 12 to 20. All participants responded to at renamed creative expression, and the factor previously least one version of the AERSQ-E and provided infor- called communication was renamed social support (for a mation about their gender and age. As the data were col- full comparison between versions including changes to lected within two separate larger projects, additional the item pool, see Supplemental Table S1). psychological variables were available for some but not In total, six different samples (1,104 students in total, all samples. These include measures of psychological dif- aged 12–20 years; a more detailed description of the ficulties, depression and anxiety, life satisfaction, non- samples is provided in the next section) were used to suicidal self-injury, and DE (see ‘‘Measures’’ section develop and validate the extended version of the below for more details). AERSQ in the present study. We used Sample 1 and Sample 2 (age ranging from 12 to 16, M = 14.15) to age refine and narrow down the first version of the AERSQ- Sample 1—Private Junior High School—Initial AERSQ-E E to a set of 20 items distributed across five factors (four Version. The first sample comprised 254 adolescents (130 items per factor) based on interitem correlations as well girls, 121 boys, 18 undisclosed or not identifying as as each item’s theoretical meaningfulness and distinc- either a girl or boy) in seventh to ninth grade. Ages ran- tiveness as an indicator of a particular factor. These five ged from 12 to 16 (mean [SD] age = 14.17 [0.96], miss- factors included rumination/negative thinking, positive ing = 13). Data were collected during a separate lecture reorientation, creative expression, aggressive outlet, and hour and administered alone by teachers at the school social support. However, it became evident that the fac- provided with a link to the survey by the researchers. tor aiming to assess distraction still needed further revis- Since respondents of this sample were only asked to ing. Consequently, in Sample 3, improvements were respond to the AERSQ-E and not measures directly made specifically for this subscale, and we managed to associated with psychological distress (e.g., measures of narrow down a final distraction subscale consisting of self-injury and depression), clinical psychologists were three items. This resulted in a final version consisting of available by phone or e-mail rather than in person. 23 items in total. 6 Assessment 00(0) Sample 2—Public Junior High School—Initial AERSQ-E Sample 4 and 5 form a combined sample for the valida- Version—Self-Harm Project. The second sample comprised tion of AERSQ-E in Grade 7 to 9 but differ in terms of 232 adolescents (102 girls, 126 boys, 4 undisclosed or other measured variables. not identifying as either a girl or boy) in Grades 7 to 9. Ages ranged from 13 to 16 (mean [SD] age = 14.13 Sample 6—Two Private High Schools–Final AERSQ Version— [0.88]). The survey was administered by a clinically Self-Control Project. The sixth sample comprised 340 ado- trained researcher along with a research assistant, two lescents (170 girls, 166 boys, 6 undisclosed or not identi- master’s students, and teachers at the school. The collec- fying as either a girl or boy) enrolled in two private high tion was conducted during a separate lecture hour in a schools in southern Sweden. Ages ranged from 16 to 20 school classroom. (mean [SD] age 17.10 [0.91], missing = 25). The survey Test–retest data also became available for 178 was administered by a clinically trained researcher and a respondents (mean age [SD] = 14.20 [0.86], 76 girls, 99 research assistant. The collection was conducted either boys, 3 undisclosed or not identifying as either a girl or during a separate lecture hour in a school classroom or boy) from this sample at a delay ranging from 28 to 35 using a digital classroom in Microsoft Teams. days (response rate = 76.6%). Students were provided digital access by e-mail, and the scale was administered by teachers at the school. The retest comprised only the Procedure AERSQ-E and one other scale that assesses embodi- Respondents in all samples answered the survey digitally ment, which was not part of this study. using either personal or school-distributed laptops, tablets, or cell phones. All respondents were informed of Sample 3—Public Junior High School—Second AERSQ-E the purpose and contents of the study, that personal Version —Self-Harm Project. The third sample comprised information (which was kept as part of the longitudinal 280 adolescents (151 girls, 126 boys, 3 undisclosed or aspirations of the broader project) would not be part of not identifying as either a girl or boy) in Grades 7 to 9. the analysis, and that participation was voluntary. At Ages ranged from 13 to 16 (mean [SD] age 14.17 [0.85]). least one clinically trained psychologist was available The survey was administered by a clinically trained either on site or/and by phone or e-mail to handle any researcher along with a research assistant, two master’s iatrogenic effects or other problems and concerns that students, and teachers at the school. The collection was could arise during or after the administration of the conducted during a separate lecture hour in a school questionnaire (see specific samples above for more classroom. details). All respondents provided consent prior to par- ticipation, and for all respondents under the age of 15, parental consent was obtained as well in accordance Sample 4—Public Junior High Schools—Final AERSQ-E with Swedish law. Ethical approval was provided by the Version—Self-Harm Project. The fourth sample comprised Swedish national ethics review board (registration num- 107 adolescents (51 girls, 57 boys, 2 undisclosed or not bers 2020-05885; 2021-06695-01; 2022-02093-02). identifying as either a girl or boy) in Grades 7 to 9. Ages ranged from 13 to 17 (mean (SD age 14.37 [0.96]). The survey was administered by a clinically trained Measures researcher and teacher at the schools. The collection was Across the different samples, we collected several psy- conducted during a separate lecture hour in a school chological variables used to assess the convergent and classroom. Sample 4 and 5 form a combined sample for discriminant validity of the AERSQ-E. Reliability esti- the validation of the AERSQ-E in Grade 7 to 9 but dif- mates and an overview of which variables were collected fer in terms of other measured variables. in each sample of the current study are presented in Table 1. Sample 5—Public Junior High School–Final AERSQ-E Version— Psychological difficulties were assessed using the Self-Control Project. The fifth sample comprised 145 ado- Strengths and Difficulties Questionnaire–self-report ver- lescents (74 girls, 70 boys, 1 undisclosed or not identify- sion (SDQ-s; Goodman, 1997). We used four subscales ing as either a girl or boy) in Grades 7 to 9 across two consisting of five items each: hyperactivity/inattention junior high schools. Ages ranged from 13 to 16 (mean (e.g., ‘‘I am easily distracted, I find it difficult to concen- [SD] age 14.45 [0.97]). The survey was administered by a trate’’), emotional symptoms (e.g., ‘‘I worry a lot’’), con- clinically trained researcher, a research assistant, and duct problems (e.g., ‘‘I get very angry and often lose my teachers at the schools. The collection was conducted temper’’), and peer relationship problems (e.g., ‘‘Other during a separate lecture hour in a school classroom. children or young people pick on me or bully me’’). ˚ Radman et al. 7 Table 1. Internal Consistency Coefficients (Cronbach’s a) for Variables Used for Evaluation of Convergent and Discriminant Validity of the AERSQ-E Across Samples. Sample 2 Sample 3 Sample 4 + Sample 6 Combined Scale (n = 236–237) (n = 281–283) 5(n = 250–252) (n = 350–351) (n = 1,017–1,023) SDQ-s Hyperactivity .78 .78 .82 .75 .78 Conduct .63 .55 .55 .46 .56 Emotion .77 .75 .77 .72 .74 Peer .48 .63 .60 .57 .59 RCADS-25 Anxiety .87 .87 .88 .86 .86 Depression .86 .89 .88 .87 .88 SLSS .88 .92 .90 .87 .90 DSHI-9r .90 .88 .90 — .88 RIBED-8 .88 .88 .87 — .88 Note. SDQ-s = Strength and Difficulties Questionnaire–self-report version; RCADS-25 = Revised Child Anxiety and Depression Scale–shortened version; SLSS = Students’ Life Satisfaction Scale; DSHI-9r = Deliberate Self-Harm Inventory–9-item version; RIBED = Risk Behaviour related to Eating Disorders; AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended. Measured only in sample 5 (n = 107). Items are rated on a 3-point scale (0 = not true,1= previous research where it showed good test–retest relia- somewhat true,2 = certainly true) taking the past 6 bility and a values ranging from .66 to .90 (Bja¨ rehed & months into consideration. The Swedish version was Lundh, 2008; Lundh et al., 2007; Lundh et al., 2011). empirically validated by Lundh et al. (2008) where it was The scale asks respondents to use a Likert-type scale shown to have similar psychometric properties to other from 0 (never)to6(more than five times) to indicate how language versions. often they had deliberately injured themselves (e.g., by Depression and anxiety were assessed using a short cutting, carving, or severely scratching themselves, or preventing wounds from healing) in the past 6 months. version of the Revised Children’s Anxiety and All individual items were summed into a total score Depression Scale (RCADS; Chorpita et al., 2000). The (ranging 0–54) reflecting frequency of engagement. shortened 25-item version (RCADS-25) comprises two Disordered eating was assessed using the eight-item subscales aiming to measure anxiety and depression and scale Risk Behaviour related to Eating Disorders has been shown to have comparable psychometric prop- (RiBED-8; Waaddegaard et al., 2003). The scale was erties to the long version, with alpha values ranging developed as a screening instrument aimed at assessing from .86 to .91 for the anxiety subscale and .79 to .80 for the prevalence of risk behaviors associated with eating the depression subscale (Ebesutani et al., 2012). The disorders. Respondents are asked to rate on a 4-point anxiety subscale consists of 15 items (e.g., ‘‘I worry when scale ranging from 1 (almost never/never)to4(very I think I have done poorly at something’’) and the often) how often they engage in thoughts or behaviors depression subscale consists of 10 items (e.g., ‘‘Nothing associated with food consumption (e.g., ‘‘I vomit to rid is much fun anymore’’). Items are rated on a 4-point myself of food I have eaten’’). In their original paper, Likert-type scale ranging from 0 (never)to 3 (always) Waaddegaard et al. (2003) demonstrated that the where high scores represent high degrees of anxiety or RiBED had good test–retest reliability and ability to depression, respectively. identify persons with eating pathology. In previous stud- Life satisfaction was assessed using the Student’s Life ies targeting Swedish adolescents, alpha values have ran- Satisfaction Scale (SLSS; (Huebner, 1991a, 1991b). The ged from .74 to .78 (Bjarehed & Lundh, 2008; Lundh scale consists of six items (e.g., ‘‘My life is going well’’) et al., 2008). rated on a 6-point scale ranging from 1 (strongly dis- agree)to 6 (strongly agree). The original article (Huebner, 1991a) demonstrated good test–retest reliabil- Data Analyses ity and a values of .80 and .82 across two studies. Non-suicidal self-injury was assessed with a modified Missingness and Exclusion. The missing completely at ran- version of the Deliberate Self-Harm Inventory (DSHI; dom (MCAR) assumption was rejected by Little’s Gratz, 2001) that was shortened to a nine-item version MCAR test for Sample 1, x (912) = 1,017.47, p = .008, (DSHI-9r) and adapted to Swedish adolescents in Sample 3, x (4,580) = 4,962.04, p \ .001), and Sample 8 Assessment 00(0) 5, x (2,534) = 2,712.51, p = .007. However, in all three Measure Testing. Confirmatory factors analysis (CFA) samples, the ratio of x to df was below 2, suggesting was used to evaluate the suggested EFA structure of the that the deviance from the MCAR assumption was AERSQ-E in Samples 4 to 6. The CFA was run with a maximum likelihood (ML) estimator using the lavaan minor (Ullman & Bentler, 2013). In addition, all items package in R (version 0.6.9, Rosseel, 2012). Model fit had relatively low percentage totals missing (range was evaluated using the root mean square error of across samples: 1.4–5.9%). Finally, inspection of the approximation (RMSEA) with 95 % confidence inter- multivariate missing patterns using MICE package in R vals, the Comparative Fit Index (CFI), and the standar- (van Buuren & Groothuis-Oudshoorn, 2011) showed dized root mean residual (SRMR). An acceptable model that no instances of multiple missing data occurred more fit was defined as CFI . .90, RMSEA \ .06, and than once. These patterns of missingness were therefore presumed as not meaningfully different from MCAR in SRMR \ .09, and a good model fit was defined as CFI any sample. Accordingly, prior to conducting any analy- . .95, RMSEA\ .05 and SRMR\ .08 (Hooper et al., ses, we listwise deleted 12 systematic responders because 2007; Hu & Bentler, 1999). In case of a poor model fit, potential improvements to the models were also evalu- they had zero variance in the AERSQ-E, and 42 respon- ated using modification indices and the theoretical dents missing more than 10 % in the AERSQ-E, sug- soundness of the proposed additional covariances not gesting that imputing their data could significantly bias implied by the original EFA. the factor analyses. The best-fitting AERSQ-E models were subsequently The remaining missing data values were imputed in tested for measurement invariance (MI) across age and one of two ways depending on the measure. Missing gender (S. T. Lee, 2018; Milfont & Fischer, 2010). items in the RIBED-8 and DSHI-9r, both of which tend Testing for MI requires sufficiently large test groups that to be positively skewed, were imputed with 0 for all cases are of about equal size (Chen, 2007). When using with less than 3 missing values as is common practice RMSEA as a fit index, it has been suggested that n.100 (e.g., Lundh et al., 2008; Viborg et al., 2014). All other is recommended for correct rejection of models (Putnick missing values were imputed using the expectation– & Bornstein, 2016). As the sample size was below 100 maximization (EM) algorithm from the mvdalab pack- for most specific ages, we opted to assess age by compar- age in R (Afanador et al., 2021). Items were imputed at ing adolescents attending junior high school (Samples 4– item-level with other values as predictors, apart from 5, N= 252, age range 13–17, mean age = 14.41) to five respondents who did not respond to some scales in those attending high school (Sample 6, N = 340, age full and consequently had their data for those scales range 16–20, mean age = 17.10). For model compari- imputed at the scale-level using the other scales are pre- son, we relied on CFI, RMSEA, and SRMR as has dictors. All analyses reported in the current paper were become common practice (Putnick & Bornstein, 2016) run with both the imputed and the nonimputed data, due to the sensitivity of chi-square difference tests to and since imputations did not markedly change any of sample size and deviations from normality assumptions the results, all results presented below are based on the (Chen, 2007; Sass, 2016). Here, we considered values of imputed data. DCFI \ .01, DRMSEA \ .015, and DSRMR \ .03 as indicating sufficient model similarity to conclude metric Measure Development. For Samples 1 and 2 and later invariance, and DCFI \ .01, DRMSEA \ .015, and Sample 3, through which the final version was gener- DSRMR \ .01 scalar to conclude scalar invariance ated, exploratory factor analyses (EFA) on the AERSQ- (Chen, 2007; Sass, 2016). E utilized an ordinary least squares solution with obli- Next, the implied subscales of the AERSQ-E were min, an oblique rotation method which assumes that the tested for internal consistency, test–retest reliability, and latent variables can be correlated. The optimal number divergent or discriminant validity. Internal consistency of factors was determined using parallel analysis, which was assessed using Cronbach’s a values, where we compares the scree of eigenvalues of the observed data sought an a of .7 or above. The test–retest reliability with a Monte Carlo–simulated matrix of data of the coefficients were computed for all but the distraction same size. We also considered which factors had an subscale using data from Sample 2. We calculated corre- eigenvalue .1 and the point of inflection in the scree lations between the AERSQ-E subscales and measures plot. All analyses were conducted with the psych package of NSSI, DE, internalizing/externalizing problems, emo- in R (Revelle, 2017). We initially opted to retain a maxi- tional distress, and positive functioning to evaluate con- mum of four items in each subscale with factor loadings vergent and discriminant validity. Correlations are above .4. We also aimed to identify an item pool with presented using Pearson’s r for ease of interpretability as satisfactory coverage of each identified ER strategy and nonparametric alternatives yielded comparable results thus screened for items too highly correlated. in terms of statistical significance and effect size when ˚ Radman et al. 9 the assumptions were violated. Bonferroni-corrected sig- aggressive outlet (4 items), social support (4 items), and nificance values were used to avoid spurious correlations distraction (3 items). One item from the positive reorien- and applied across all parametric tests, where we used a tation subscale (i.e., ‘‘Try in a calm manner to solve corrected alpha of .05/69 = 0.0007. what made me feel bad’’) showed low factor loading (l = .32) as well as cross-loading with aggressive outlet (l = .28) in Sample 3. However, given the content of Results the item as well as the previous factor loading from Samples 1 and 2 (l = .61), we decided to retain this Exploratory Factor Analysis item in the final positive reorientation subscale. Parallel analysis with data from Sample 1 and Sample 2 on the initial version of the AERSQ-E suggested six fac- tors to best fit the data. The same was suggested by Confirmatory Factor Analysis looking at eigenvalues and the point of inflection in the The final version consisting of 23 items was adminis- scree plot. Based on EFA, four items were retained per tered in Samples 4 and 5 (representing respondents in subscale (for a summary of the results from this factor junior high school) and in Sample 6 (representing analysis, see Table S2 of the Supplementary). However, respondents in high school). When combining all sam- poor factor loadings for many items included in the dis- ples (4–6), the six-factor model showed mixed results on traction subscale suggested it required further revision. fit indices with RMSEA and SRMR showing acceptable After a close inspection of the items, we theorized values, but CFI was found to be somewhat low (N = that the poor psychometric properties of the distraction 592, RMSEA [95% confidence interval, CI] = .064 [.059, subscale were likely caused by the specificity of the .069], SRMR = .066, CFI = .884). However, one pair of items, in that, compared with other subscales, it con- items in the aggressive outlet subscale, ‘‘I argue or fight tained much more specific activities putatively reflecting with people around me’’ and ‘‘I want to hurt others (phy- a distraction-oriented ER. Examples include ‘‘Watch sically or mentally)’’ demonstrated high residual covar- something (e.g., a movie, television shows, streaming)’’ iance not explained by the latent variable (s =.62). A or ‘‘Play games (e.g., video or computer games).’’ These rerun of the combined sample adding this covariance to activities, although presumably reflecting distractive the model demonstrated acceptable model fit across all activities, might not be endorsed within-person consis- indices (N = 592, RMSEA [95% CI] = .053 [.048, .058], tently such that a person might prefer one distractive SRMR = .061, CFI = .920, modeling these samples sep- activity over all others. Although in line with the focus arately yielded comparable results, cf. Supplemental of the AERSQ-E on assessing specific ER behaviors, we Table S3). The loadings and covariances of these models decided to reformulate the scale to focus more on the are visualized in Figure 1. The modified model signifi- cognitive processes at work, hopefully allowing the sub- cantly improved model fit (Dx = 19.06, p\ .001). scale to identify distraction more consistently and inde- pendently as an ER strategy across all items. Examples of new items include ‘‘Try to think about something Internal Consistency else’’ and ‘‘Distract myself with something to do’’ (also Table 3 presents the internal consistency values see Table 2). (Cronbach’s a) of the AERSQ for all samples. Across A second version of the AERSQ-E, including five all samples and in combination, alpha values for five out new items intended to capture distraction and 20 items of six factors tended to be in acceptable or high ranges. retained from the first EFA for the other subscales, was The exception was a distraction, which showcased ade- subsequently tested in Sample 3. We limited the second quate values for Sample 3 (a = .71) but somewhat lower revised distraction subscale to five new items because of values for Samples 4 and 5 (a = .57), Sample 6 (a = a need to minimize the burden put on respondents con- .65), and in a pooled analysis (a = .65). sidering the data collection at large. Once again, parallel analysis again suggested six factors to best fit the data (as did the eigenvalues and point of inflection using a Test–Retest Reliability scree plot), and results from the subsequent EFA is pre- sented in Table 2. For the distraction subscale, we All test–retest correlations were medium to large (n = selected three items showing satisfactory factor loading 178): r [95% CI] rumination/negative thinking = .77 and construct coverage, resulting in a final 23-item ver- [.70, .82], positive reorientation = .71 [.62, .77], creative sion of the AERSQ with six different subscales labeled expression = .73 [.66, .79], aggressive outlet .71 [.63, as rumination/negative thinking (4 items), positive reor- .77], and social support = .77 [.70, .82]. No retest data ientation (4 items), creative expression (4 items), were available for the distraction subscale, as it was 10 Assessment 00(0) Table 2. Exploratory Factor Loadings for the Final Version of the AERSQ-E (Sample 3). Sample 3 Factor Item (in English translation) 1 2 3456 R1. Think about things I have said or done (again and again) .60 –.05 .17 –.09 .00 .06 R2. Think that I am bad or worthless .69 –.25 .05 –.02 –.01 .04 R3. Worry about what others might think of me .66 –.10 .03 –.10 .02 –.02 R4. Believe that others have it much better off than me .64 .02 –.05 .16 .02 .01 P1. Try to find something positive in what has happened –.01 .56 .04 .01 .06 .16 P2. Move on and try to do things better next time –.03 .73 .00 –.13 –.02 .05 P3. Try in a calm manner to solve what made me feel bad .09 .32 .12 –.15 .28 .21 P4. Stay calm and think that it will pass –.10 .81 .01 .00 .03 –.01 C1. Create something that expresses how I feel –.19 –.04 .71 .07 .02 .11 C2. Write texts (e.g., stories, song lyrics, poems) .15 .10 .64 .00 –.04 –.11 C3. Draw or paint –.02 –.01 .64 .02 .01 .02 C4. Write down thoughts about how I feel .27 .05 .58 –.04 .02 –.06 A1. Punch or kick on things –.13 –.09 .00 .71 –.08 –.02 A2. Try to find something to break –.06 –.10 .04 .74 .02 .04 A3. Argue or fight with people around me .40 .00 .02 .49 .07 –.05 A4. Want to hurt others (physically or mentally) .30 .10 .07 .49 .00 –.18 S1. Tell someone else how I feel –.06 –.02 –.10 .01 .88 .06 S2. Seek support and comfort in others .01 .04 .03 .02 .87 –.08 S3. Ask others for advice or help .05 .05 –.02 –.07 .71 .01 S4. Seek physical contact (e.g., a hug) .05 –.06 .28 .00 .60 .06 D1. Try to think about something else .01 .12 –.06 –.01 –.01 .77 D2. Distract myself with something to do –.01 –.11 .12 –.07 .07 .62 D3. Try to forget that which makes me feel bad .01 .07 .01 –.01 –.03 .57 Avoid things that remind me of my feelings .38 .09 –.03 .15 .14 .37 Pretend like the emotions I feel do not exist .28 .03 .03 .24 –.22 .32 Note. The Exploratory Factor Analysis was based on oblimin rotation and extraction by ordinary least squares. Bold represents final factor assignment. AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended. a b Item P3 was retained with a factor loading of .61 in Samples 1 and 2. Items removed from the final version. Table 3. Cronbach’s a Values for AERSQ-E Subscales Across Samples. Sample Subscale 1 2 3 4–5 6 Combined Rumination/negative thinking .80 .84 .79 .80 .77 .80 Positive reorientation .76 .80 .78 .78 .78 .78 Creative expression .70 .72 .75 .68 .78 .73 Aggressive outlet .75 .68 .74 .77 .68 .72 Social support .82 .88 .85 .79 .84 .84 Distraction NA NA .71 .57 .65 .65 Note. AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended. Sample 1 and 2 are not included. revised in a later sample compared with when test–retest invariance between age groups (i.e., junior high school stu- reliability assessments were performed. dents were compared to high school students) and between girls and boys across all fit indices of interest (DCFI gender = .005, DRMSEA \ .001, DSRMR = .003; gender gender Measurement Invariance DCFI =.007, DRMSEA =.001, DSRMR = age age age Table 4 summarizes the results for method invariance .005). Considering scalar invariance for both age and gen- across age and gender. We found evidence of metric der, DRMSEA and DSRMR suggested scalar invariance ˚ Radman et al. 11 Figure 1 Six Factor AERSQ-E Model With 23 Indicator Items Note. N = 592, RMSEA [95% CI] = .053 [.048, .058], SRMR = .061, CFI = .920. Loadings and covariances represent combined data from Samples 4 to 6. See Table S3 of Supplementary for estimated confirmatory factor loadings, covariances and model fit for each sample separately. AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire —Extended; RMSEA = root mean square error of approximation; CI = confidence intervals; CFI = Comparative Fit Index; SRMR = standardized root mean residual. Table 4. Comparing Configural, Metric and Scalar Invariance Across Gender and Age Groups. Group Model x p value CFI RMSEA SRMR Gender 1. Configural 806.04 \.001 .909 .055 .061 2. Metric 835.06 \.001 .906 .055 .064 |D| 2-1 29.02 .034 .003 \.001 .003 3. Scalar 944.30 \.001 .883 .060 .069 |D| 3-2 109.24 \.001 .023 .005 .005 Age 1. Configural 765.23 \.001 .923 .052 .064 2. Metric 808.69 \.001 .917 .053 .068 |D| 2-1 43.46 \.001 .007 .001 .005 3. Scalar 895.94 \.001 .901 .057 .071 |D| 3-2 87.25 \.001 .016 .003 .003 Note. RMSEA = root mean square error of approximation; CFI = comparative fit index; SRMR = standardized root mean residual. but DCFI did not (DCFI = .023, DRMSEA \ found a positive association between rumination/negative gender gender .005, DSRMR = .005; DCFI = .016, DRMSEA thinking and aggressive outlet (Samples 2–6, r= .28), and gender age age = .003, DSRMR = .003). Scalar invariance across all both these were negatively associated with positive reorien- age indices (DCFI = .008–0.010, DRMSEA = .002, tation (rumination/negative thinking: Samples 2–6, r=– DSRMR = .002) could only be established by releas- .28; aggressive outlet: Samples 2–6, r=–.31). Positive reor- ing the two items significantly associated with the ientation, conversely, was as expected positively associated highest model fit improvement for gender (i.e., ‘‘Argue with both social support (Samples 2–6, r = .27) and dis- or fight with people around me’’ and ‘‘Seek physical traction (Samples 3–6, r = .37). Creative expression was contact (e.g., a hug)’’; x = 16.44–34.43) and age, positively associated with social support (Samples 2–6, r= respectively (i.e., ‘‘Think about things I have said or .24), rumination/negative thinking (Samples 2–6, r= .27), done (again and again)’’ and ‘‘Distract myself with and aggressive outlet (Samples 2–6, r= .15), whereas we something to do’’; x = 19.24–20.28). expected to find a positive association only with social sup- port and distraction. Finally, in line with expectations, Subscale Intercorrelations social support was weakly but positively associated with distraction (Samples 3-6, r= .21). All correlations pre- Table 5 presents expected and observed intercorrelations sented here were significant at p\ .0007. between all subscales of the AERSQ-E. As expected, we 12 Assessment 00(0) Table 5. Expected and Observed Intercorrelations Between the AERSQ-E Subscales (Sample 2–6, N = 872–1,104). Expected pattern Observed pattern Subscale RP C A S D R P C A S D N Rumination/negative thinking 1 1,104 Positive reorientation — –.28 1 1,104 *** Creative expression 0/- + .27 .03 1 . 1,104 *** Aggressive outlet ++ — — .28 –.31 .15 1 1,104 *** *** *** Social support — ++ + — .11 .27 .24 –.04 1 1,104 *** *** *** Distraction 0 ++ + 0 + .06 .37 .08 –.12 .21 1 872 *** * *** *** Note. Significant correlations after Bonferroni correction are shown in bold. Expected patterns reflects the rating of two independent raters +++ = strong positive correlation (r . .5), ++ = moderate positive correlation (.3\ r . .5), + = weak positive correlation (.1\ r . .3), 0 = no correlation, – = weak negative correlation (r\ –.1), –– = moderate negative correlation (–.1\ r . –.3), ––– = strong negative correlation (r . –.5). AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended; Columns R = Rumination/negative thinking; P = Positive reorientation; C = Creative Expression; A = Aggressive outlet; S = Social support; D = Distraction. *p\ .05. **p\ .01. ***p\ .001. Convergent and Discriminant Validity constructs including hyperactivity/inattention (Samples 2–6, r = –.34), emotional symptoms (Samples 2–6, r = Table 6 presents a summary of results including an –.29), conduct problems (Samples 2–6, r = –.26), anxi- expected and observed pattern of association with other ety (Samples 2–6, r = –.27), depression (Samples 2–6, r psychological variables. Key findings (all significant at p \ 0007 in accordance with Bonferroni correction) are –.38), NSSI (Samples 2-4, r = –.38) and DE (Samples summarized below. 2–4, r = –.33). As expected, moderate to strong positive associations We predicted social support to be weakly but nega- were found between rumination/negative and negative tively associated with most negative constructs and posi- constructs including anxiety (Samples 2–6, r = .59), tively associated with life satisfaction. In line with these depression (Samples 2–6, r = .51), NSSI (Samples 2–4, r predictions, we found a weak but negative correlation = .33), DE (Samples 2–4, r = .48), and emotional between psychological difficulties and social support symptoms (Samples 2–6, r = .53). Conversely, rumina- including hyperactivity/inattention (Samples 2–6, r =– tion/negative thinking had a strong negative correlation .13) and conduct problems (Samples 2–6, r = –.14) and with life satisfaction (Samples 2–6, r = –.51). We also a weak positive correlation to life satisfaction (Samples expected a moderate positive association between rumi- 2–6, r = –.15). Although some other weak correlations nation/negative thinking and conduct problems, which were also found, these did not retain significance after was not supported by the data. Bonferroni correction. Similar patterns were observed for an aggressive out- For creative expression, we only made predictions for let, which was positively correlated with anxiety some of the associations, and when we did, we expected (Samples 2–6, r = .24), depression (Samples 2–6, r = weak or no associations. However, we found weak but .31), NSSI (Samples 2–4, r = .37), DE (Samples 2–4, positive associations to most negative constructs includ- r = .25), and emotional symptoms (Samples 2–6, r = ing peer problems (Samples 2–6, r = .14), emotional .17) and negatively correlated with life satisfaction symptoms (Samples 2–6, r = .23), anxiety (Samples 2–6, (Samples 2–6, r = –.29). In addition, aggressive outlet r = .28), depression (Samples 2–6, r = .20), NSSI had a moderate to strong relationship with hyperactiv- (Samples 2–4, r = .18), and DE (Samples 2–4, r = .19) ity/inattention (Samples 2–6, r = .36) and conduct and weak but negative association to life satisfaction problems (Samples 2–6, r = .41). However, we did not (Samples 2–6, r = –.18). find the expected positive association between aggressive Finally, given the ambiguous functionality of distrac- outlet and peer problems (Samples 2–6, r = .10). tion as an ER strategy, we expected its subscale to have Positive reorientation, in line with expectations, was few and weak associations with other measured con- positively correlated to life satisfaction (Samples 2–6, structs. This was confirmed in that all correlations were r = .42) and negatively correlated to most negative ˚ Radman et al. 13 Table 6. Expected and Observed Intercorrelations Between the AERSQ-E and Other Studied Variables in Combined Sample (Sample 2–6, N = 619–1,104). Expected pattern Observed pattern Variable R P C A S D R P C A S D Psychological difficulties (SDQ) Hyperactivity/inattention + – 0/– + 0/– 0 .12 –.34 .02 .36 –.13 –.08 1,104 *** *** *** *** Peer relationship problems ++ – 0/– + – 0 .17 –.12 .14 .10 –.04 –.11 1,104 *** *** *** *** ** Emotional symptoms +++ 0/– + + – 0/+ .53 –.29 .23 .17 .12 –.01 1,104 *** *** *** *** *** Conduct problems ++ – 0 ++ – 0 .06 –.26 .04 .41 –.14 –.13 1,104 * *** *** *** *** Life satisfaction –+ 0/+ –+ 0 –.51 .42 –.18 –.29 .15 .12 1,104 (SLSS) *** *** *** *** *** *** Anxiety and depression (RCADS-25) Anxiety ++ – NA + – 0/+ .59 –.27 .28 .24 .11 .00 1,104 *** *** *** *** *** Depression +++ – NA + – 0/+ .51 –.38 .20 .31 –.04 .05 1,104 *** *** *** *** Self-injury (DSHI) ++ – NA ++ 0/– 0 .33 –.38 .18 .37 –.08 –.15 619 *** *** *** *** * Disordered eating (RIBED-8) ++ – NA NA 0/– – .48 –.33 .19 .25 –.01 –.04 619 *** *** *** *** Note. Significant correlations after the Bonferroni correction are shown in bold. Expected patterns reflects the rating of two independent raters (+++ = strong positive correlation (r . .5), ++ = moderate positive correlation (.3\ r . .5), + = weak positive correlation (.1\ r . .3), 0 = no correlation, – = weak negative correlation (r\ –.1), –– = moderate negative correlation (–.1\ r . –.3), ––– = strong negative correlation (r . –.5), NA = no predictions made). SDQ = Strengths and Difficulties Questionnaire –self-report version; SLSS = Student’s Life Satisfaction Scale; RCADS = RCADS; DSHI = Deliberate Self-Harm Inventory; RIBED = Risk Behaviour related to Eating Disorders; Columns R = Rumination/negative thinking; P = Positive reorientation; C = Creative Expression; A = Aggressive outlet; S = Social support; D = Distraction. p\ .05. **p\ .01. ***p\ .001. weak, and none retained significance after the AERSQ-E. However, the CFI did not reach adequate Bonferroni correction. values unless the residual of two aggressive outlet- related items (i.e., ‘‘Punch or kick things’’ and ‘‘Try to find something to break’’) was covaried as suggested by Discussion the modification indices. Post hoc modifications of the factor structure of a measurement should only be pur- In this study, we presented the psychometric properties sued when empirically or conceptually justified of the AERSQ-E, a modified and extended version of (MacCallum et al., 1992). In this case, the two items for the original AERSQ developed by Zhou and colleagues which residuals were covaried reflected tendencies to (2020). Across six community samples of Swedish youth direct aggressive behaviors toward inanimate objects, (aged 12–20), we generated and narrowed down a final while the other two items of the aggressive outlet sub- 23 item version of the AERSQ-E and evaluated its inter- scale reflected aggression aimed at other people. We nal structure, reliability, and validity as a measure asses- deemed this modification to be justified although it was sing the use of different ER strategies in adolescence. applied post hoc, where we reasoned that the conceptual The factor analyses generally supported a six-factor delineation this modification reflected between aggres- structure for the AERSQ-E including rumination/nega- sion toward inanimate objects or toward other people tive thinking (4 items), positive reorientation (4 items), (i.e., socially) does not take away from the use of the creative expression (4 items), aggressive outlet (4 items), subscale as a measure of ER through aggression at a social support (4 items), and distraction (3 items). These more general level. factors showed low to moderate correlations to each Assessing measurement invariance, we found that other. Using confirmatory factor analysis, both SRMR metric invariance of the modified factor structure of the and RMSEA showed adequate model fit for the AERSQ-E was supported across girls and boys and in 14 Assessment 00(0) comparing junior high school students and standard lower-than-usual a value could reflect sufficient con- high school students. However, while the DRMSEA and struct coverage (Taber, 2018) rather than problems of DSRMR gave support to scalar invariance across these multidimensionality. gender and age groups, the DCFI did not. We note that With respect to external validity, our findings gener- there are currently no clear conventions on how to inter- ally aligned with our expectations. We found that rumi- pret inconsistent results across different fit indices, and nation/negative thinking had positive associations with there is large variability in cut-off levels as well as what anxiety, depression, DE, and NSSI, aligning with data fit indices should be considered most important in evalu- found using the previous version of the AERSQ (Zhou ating measurement invariance (see Putnick & Bornstein, et al., 2020). Rumination, one of two cognitive strategies 2016, for a discussion on conventions of measurement assessed by the AERSQ-E and one of the most studied invariance testing). Nevertheless, scalar invariance is ER strategies, has been extensively linked to several important because it ensures that statistical differences mental health issues, with many studies specifically tar- in group means reflect actual differences in ER strategy geting adolescents (e.g., Calvete et al., 2015; Olatunji preferences and not unintended, biasing properties of et al., 2013; Rood et al., 2009; Royuela-Colomer et al., the scale (S. T. Lee, 2018). We managed to establish par- 2021). Accordingly, we found that rumination/negative tial scalar invariance by releasing some items but have thinking had the strongest associations with both posi- no strong theoretical reasons behind these specific modi- tive and negative aspects of mental health and psycholo- fications. Future work could examine more closely why gical functioning. This aligns with a meta-analytic specifically these items are interpreted differently by girls review by Aldao et al. (2010) who, comparing different and boys and by different levels of high school students. ER strategies, suggested that rumination had the stron- Some potentially important factors are the role played gest effect size in predicting anxiety, depression, eating, by social reinforcement in emotion and ER (e.g., Barrett and substance-related disorders. Moreover, we found & Bliss-Moreau, 2009; Nolen-Hoeksema, 2012) as well that positive reorientation, the second cognitive strategy as levels of cognitive maturation (e.g., Ahmed et al., assessed by the AERSQ-E, was associated with various 2015; Casey et al., 2008). aspects of positive functioning, showing positive asso- Regarding internal consistency, we found that all ciations with life satisfaction and negative associations subscales except distraction had acceptable internal con- with internalizing and externalizing problems. This cor- sistency (i.e., a . .70) and good test–retest reliability roborates the common understanding of strategies (i.e., r . .70). Alpha values for the distraction subscale involving positive reappraisal as adaptive (e.g., Cracco varied between 0.57 and 0.71 across different samples. et al., 2017; Schafer et al., 2016) and having wide- In a recent review on the use of Cronbach’s a in research ranging benefits to mental health (e.g., Aldao et al., on instrument development, Taber (2018) provided illus- 2010; Nowlan et al., 2015). trative examples from the science education literature Social support, the interpersonal dimension assessed showing a wide range of alpha values being treated as with the AERSQ-E, was weakly but negatively associ- acceptable or satisfactory (e.g., as low as a = .45). The ated with internalizing and externalizing problems and article raised concerns with the arbitrary value of .70 as positively associated with life satisfaction. This corrobo- a sufficient measure of acceptable internal consistency, rates findings suggesting that the availability of interper- citing several influential statisticians. For instance, sonal resources can contribute to positive functioning although Cronbach (1951) himself suggested that a high (Dixon-Gordon et al., 2015). However, we urge that value of alpha was ‘‘desirable,’’ he also emphasized the future work interprets this scale within context, as who importance of instrument interpretability which, accord- is providing the support could have important implica- ing to him, was often possible without having high val- tions for the effectiveness and availability of utilizing ues of alpha. Similar conclusions have been made by this strategy to regulate emotion. Contextual interpreta- Schmitt (1996) claiming there is no general level (such as tion is also necessary for the distraction subscale, as it .70) at which a becomes acceptable and that instruments demonstrated only negligible associations (r \ .15) with with quite low alpha values can prove useful. In relation the other variables. Other studies have similarly sug- to coping and ER specifically, early researchers postu- gested that distraction is only weakly associated with lated that a low alpha value is sometimes expected if the concurrent and future levels of depression (Rood et al., use of one type of coping (ER) strategy obviates the use 2009). These findings reinforce the view that distraction of another (Billings & Moos, 1981). Given that the dis- is neither adaptive nor maladaptive outside its context; traction subscale comprised only three items, meaning rather, it can be a predictor of positive functioning when that each item intercorrelation has a considerable effect combined with an attitude of acceptance and a predictor on the a value (the average inter-item correlation for the of negative functioning when combined with an attitude distraction subscale was r = .36), we suggest that a of avoidance (Wolgast & Lundh, 2017). ˚ Radman et al. 15 Finally, the two expressive ER strategies (i.e., creative adolescence. It includes both ER strategies more focused expression and aggressive outlet) were in this study both on changes in cognition (i.e., rumination/negative think- positively related to internalizing and externalizing ing or positive reorientation) and changes in behavior problems and negatively related to life satisfaction. This (i.e., creative expression, aggressive outlet, or distrac- is unsurprising for aggressive outlet, as aggression has tion). The instrument also covers the interpersonal previously been linked to NSSI and deliberate self-harm dimension of ER (i.e., social support) contrasting with (Boxer, 2010; Brunner et al., 2007; Daukantaite _ et al., intrapersonal ER strategies. Moreover, it also captures 2019; Fliege et al., 2009; Sourander et al., 2006; Tang ER strategies commonly viewed as maladaptive (rumi- et al., 2013) and is largely considered a maladaptive ER nation/negative thinking, aggression outlet) and adap- strategy (e.g., Cracco et al., 2017; Grob & Smolenski, tive (e.g., positive reorientation or social support). 2005). Finding consistent, albeit weak, positive links to Finally, one key feature of the AERSQ is the behavioral mental health problems for creative expression was ER strategies focusing on expressions of emotions, more surprising. In the original AERSQ paper (Zhou including the subscales aggressive outlet and creative et al., 2020), the ‘‘cultural activities’’ subscale (compara- expression, which is a less frequently considered dimen- ble to the creative expression subscale of the current ver- sion of ER across measurements, with expressive sup- sion) produced mixed results, showing positive pression, i.e., the inhibition of emotional expression, associations only to NSSI and emotional symptoms, being much more common (e.g., Gross & John, 2003; and only at one out of two measured time-points. Hofmann & Kashdan, 2010; Zeman et al., 2001). Thus, Conversely, research regarding the influence of perform- the AERSQ-E holds promise to be applicable in a wide ing art-based activities tend to highlight the beneficial range of research projects emphasizing different distinc- effect with regard to mental health (e.g., Geipel et al., tions in ER. 2018; Saarikallio, 2010; van Lith et al., 2013; Zhao et al., Beyond the limitations discussed previously, there are 2016), whereas our findings suggest the opposite. It a few others that should be discussed. First, given the should be noted, however, that the current study is fact that this study is cross-sectional and correlational, cross-sectional, meaning the associations found here do we cannot ascertain whether making use of certain ER not necessarily translate to the longitudinal effects of strategies poses a risk of developing mental health issues using art-based activities as ER strategies. It is conceiva- or if it is the other way around, whereby a preexisting ble that there would be prospective beneficial effects of level of distress causes the favoring of certain ER strate- the creative expression subscale when assessed longitud- gies over others. For instance, the cross-sectional nature inally except that art-based creative activities are com- of this study might be the cause of the weak but overall mon among a portion of those with a tendency toward positive association found between creative expression issues of mental health (Cropley, 1990). Future longitu- and mental health issues, contrasting previous findings dinal studies could investigate this further. showing mental health benefits for these types of activi- An important strength of the present study is that the ties (e.g., Geipel et al., 2018; Zhao et al., 2016). sample encompassed 1,104 adolescents in total sourced Longitudinal assessments are needed to further investi- from schools located in Swedish municipalities that were gate the validity of each subscale in this regard. comparable to the Swedish average. In addition, we had Second, we did not manage to collect test–retest data roughly an equal number of both adolescent girls and for the distraction subscale given its late revision in the boys covering a wide range of ages. Together, this lends development process, meaning we could not evidence strong support to the representativity of the sample and the longitudinal reliability of this subscale. Given this generalizability of the findings, at least covering a and its relatively low internal consistency values, future Swedish setting and across countries with comparable studies are warranted to seek improvements to this sub- demographics. Future studies are needed to evaluate the scale in particular. structure and psychometric properties of the AERSQ-E Third, although our goal was to avoid items requiring cross-culturally. Future studies should also examine the complex metacognitive evaluations, the degree to which instrument among those not identifying as either men/ we managed this is uncertain. For instance, we reformu- boys or women/girls. Finally, we did not recruit from lated the distraction subscale to be more general, thereby exclusively clinical populations of adolescents who increasing its level of abstraction. This means the ques- might show unique difficulties regarding ER, implying tions (e.g., ‘‘Try to think about something else’’) require that its applicability in a clinical setting has not been an ability to not only draw connections between felt established, something we suggest for future studies. emotions and behavior, but also to some extent under- Another strength is that the AERSQ-E encompasses stand the cognitive processes that drove that behavior. a broad range of important ER dimensions, making it a Importantly, these items as well as the items included in holistic yet resource-efficient measurement of ER in the other subscale do not require insight about the 16 Assessment 00(0) consequences of these cognitive and behavioral pro- numbers SOL: 2020-05885; VIS: 2021-06695-01; 2022-02093- 02). cesses, which would have added metacognitive self- evaluative requirements we deem particularly proble- matic for assessment in adolescence. ORCID iDs To summarize, the AERSQ-E has several merits as it Gustaf Ra˚ dman https://orcid.org/0000-0002-7042-0196 addresses some limitations posed by the current instru- Benjamin Clare´ us https://orcid.org/0000-0003-0972-6660 ments used to measure adolescent ER, and it showed Daiva Daukantaite https://orcid.org/0000-0002-1994-041X some improvement in psychometric properties com- pared with its predecessor, the AERSQ. However, fur- Data Availability Statement ther improvements could be made, and specifically, The data that support the findings of this study are available regarding the distraction subscale the validity and relia- on request from the corresponding author. The data are not bility could be further investigated. In addition, the publicly available due to privacy or ethical restrictions. instrument should be validated using longitudinal data and for use with clinical samples. Nevertheless, we believe the AERSQ-E has the potential to contribute to Supplemental Material our knowledge of ER across adolescence. Supplemental material for this article is available online. Acknowledgment References The authors wish to extend their sincere gratitude to Lars- Adrian, M., Zeman, J., & Veits, G. (2011). Methodological Gunnar Lundh and Margit Wa˚ ngby Lundh for their invalu- implications of the affect revolution: A 35-year review of able contribution in developing the AERSQ, and for their emotion regulation assessment in children. Journal of assistance in creating the extended version of the AERSQ. Experimental Child Psychology, 110(2), 171–197. https:// doi.org/10.1016/J.JECP.2011.03.009 Afanador, N. L., Tran, T., Blanchet, L., & Baumgartner, R. Author Contributions (2021). Mvdalab: Multivariate data analysis laboratory. DD contributed to the design of the larger project. GR had https://CRAN.R-project.org/package=mvdalab primary responsibility for questionnaire revisions and exten- Ahmed, S. P., Bittencourt-Hewitt, A., & Sebastian, C. L. sions and collected most of the data. GR and BC both con- (2015). Neurocognitive bases of emotion regulation devel- ducted the statistical analyses and wrote the first full draft of opment in adolescence. Developmental Cognitive Neu- the manuscript. All authors worked on several edits of the roscience, 15, 11–25. https://doi.org/10.1016/J.DCN.2015. paper. All authors contributed to and have approved the final 07.006 manuscript. Aldao, A. (2013). The future of emotion regulation research: Capturing context. Perspectives on Psychological Science, 8(2), 155–172. https://doi.org/10.1177/1745691612459518 Declaration of Conflicting Interests Aldao, A., Gee, D. G., de Los Reyes, A., & Seager, I. (2016). The author(s) declared no potential conflicts of interest with Emotion regulation as a transdiagnostic factor in the devel- respect to the research, authorship, and/or publication of this opment of internalizing and externalizing psychopathology: article. Current and future directions. Development and Psycho- pathology, 28(4pt1), 927–946. https://doi.org/10.1017/S095 Funding Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emo- The author(s) disclosed receipt of the following financial sup- tion-regulation strategies across psychopathology: A meta- port for the research, authorship, and/or publication of this analytic review. Clinical Psychology Review, 30(2), 217–237. article: This study was funded by a grant from the Swedish https://doi.org/10.1016/J.CPR.2009.11.004 Research Council for Health, Working Life and Welfare Andover, M. S., & Morris, B. W. (2014). Expanding and clari- (2016-00248); by the Thora Ohlsson Foundation; and by the fying the role of emotion regulation in nonsuicidal self- Lundh Research Foundation. injury. The Canadian Journal of Psychiatry, 59(11), 569–575. https://doi.org/10.1177/070674371405901102 Barrett, L. F., & Bliss-Moreau, E. (2009). She’s emotional. Ethical Approval He’s having a bad day: Attributional explanations for emo- The development of this new version was done as part of two tion stereotypes. Emotion, 9(5), 649–658. https://doi.org/10. project, one on self-harm, emotion regulation and interperso- 1037/A0016821 nal relations in youth (self-harm project; SOL), and one on Beauchaine, T. P. (2015). Future directions in emotion dysre- executive functions and mental health among young adoles- gulation and youth psychopathology. Journal of Clinical cents (self-control project; VIS). Ethical approval was pro- Child & Adolescent Psychology, 44(5), 875–896. https://doi. vided by the Swedish national ethics review board (registration org/10.1080/15374416.2015.1038827 ˚ Radman et al. 17 Bender, P. K., Reinholdt-Dunne, M. L., Esbjørn, B. H., & Cracco, E., van Durme, K., & Braet, C. (2015). Validation of Pons, F. (2012). Emotion dysregulation and anxiety in chil- the FEEL-KJ: An instrument to measure emotion regula- dren and adolescents: Gender differences. Personality and tion strategies in children and adolescents. PLOS ONE, Individual Differences, 53(3), 284–288. https://doi.org/10. 10(9), Article e0137080. https://doi.org/10.1371/journal. 1016/J.PAID.2012.03.027 pone.0137080 Billings, A. G., & Moos, R. H. (1981). The role of coping Cronbach, L. J. (1951). Coefficient alpha and the internal responses and social resources in attenuating the stress of structure of tests. Psychometrika, 16(3), 297–334. https:// life events. Journal of Behavioral Medicine, 4(2), 139–157. doi.org/10.1007/BF02310555 https://doi.org/10.1007/BF00844267 Cropley, A. J. (1990). Creativity and mental health in everyday Bjarehed, J., & Lundh, L. (2008). Deliberate self-harm in 14- life. Creativity Research Journal, 3(3), 167–178. https://doi. year-old adolescents: How frequent is it, and how is it asso- org/10.1080/10400419009534351 Daukantaite, _ D., Lundh, L. G., & Wa˚ ngby-Lundh, M. (2019). ciated with psychopathology, relationship variables, and styles of emotional regulation? Cognitive Behaviour Ther- Association of direct and indirect aggression and victimiza- apy, 37(1), 26–37. https://doi.org/10.1080/1650607070177 tion with self-harm in young adolescents: A person-oriented 8951 approach. Development and Psychopathology, 31(2), Bonanno, G. A., & Burton, C. L. (2013). Regulatory flexibil- 727–739. https://doi.org/10.1017/S0954579418000433 ity: An individual differences perspective on coping and Dingemans, A., Danner, U., & Parks, M. (2017). Emotion reg- emotion regulation. Perspectives on Psychological Science, ulation in Binge eating disorder: A review. Nutrients 2017, 9(11), 1274. https://doi.org/10.3390/NU9111274 8(6), 591–612. https://doi.org/10.1177/1745691613504116 Dixon-Gordon, K. L., Bernecker, S. L., & Christensen, K. Boxer, P. (2010). Covariation of self- and other-directed aggression among inpatient youth: Continuity in the transi- (2015). Recent innovations in the field of interpersonal tion to treatment and shared risk factors. Aggressive Beha- emotion regulation. Current Opinion in Psychology, 3, vior, 36(3), 205–217. https://doi.org/10.1002/AB.20343 36–42. https://doi.org/10.1016/J.COPSYC.2015.02.001 Brockmeyer, T., Skunde, M., Wu, M., Bresslein, E., Rudofsky, Ebesutani, C., Reise, S. P., Chorpita, B. F., Ale, C., Regan, J., G., Herzog, W., & Friederich, H. C. (2014). Difficulties in Young, J., Charmaine, H. M. M., & Weisz, J. R. (2012). emotion regulation across the spectrum of eating disorders. The Revised Child Anxiety and Depression Scale-Short Comprehensive Psychiatry, 55(3), 565–571. https://doi.org/ Version: Scale reduction via exploratory bifactor modeling of the broad anxiety factor. Psychological Assessment, 10.1016/J.COMPPSYCH.2013.12.001 Brunner, R., Parzer, P., Haffner, J., Steen, R., Roos, J., Klett, 24(4), 833–845. https://doi.org/10.1037/A0027283 M., & Resch, F. (2007). Prevalence and psychological corre- Eisenberg, N., Cumberland, A., & Spinrad, T. L. (1998). Par- lates of occasional and repetitive deliberate self-harm in ado- ental socialization of emotion. Psychological Inquiry, 9(4), lescents. Archives of Pediatrics & Adolescent Medicine, 161(7), 241–273. https://doi.org/10.1207/s15327965pli0904_1 641–649. https://doi.org/10.1001/ARCHPEDI.161.7.641 Fliege, H., Lee, J. R., Grimm, A., & Klapp, B. F. (2009). Risk Calvete, E., Orue, I., & Hankin, B. L. (2015). Cross-lagged factors and correlates of deliberate self-harm behavior: A associations among ruminative response style, stressors, systematic review. Journal of Psychosomatic Research, and depressive symptoms in adolescents. Journal of Social 66(6), 477–493. https://doi.org/10.1016/J.JPSYCHORES. 2008.10.013 and Clinical Psychology, 34(3), 203–220. https://doi.org/10. Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative 1521/jscp.2015.34.3.203 Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent life events, cognitive emotion regulation and emotional brain. Developmental Review, 28(1), 62–77. https://doi.org/ problems. Personality and Individual Differences, 30(8), 10.1016/J.DR.2007.08.003 1311–1327. https://doi.org/10.1016/S0191-8869(00)00113-6 Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack Geipel, J., Koenig, J., Hillecke, T. K., Resch, F., & Kaess, M. of measurement invariance. Structural Equation Modeling: (2018). Music-based interventions to reduce internalizing A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/ symptoms in children and adolescents: A meta-analysis. 10.1080/10705510701301834 Journal of Affective Disorders, 225, 647–656. https://doi. Chorpita, B. F., Yim, L., Moffitt, C., Umemoto, L. A., & org/10.1016/J.JAD.2017.08.035 Goodman, R. (1997). The strengths and difficulties question- Francis, S. E. (2000). Assessment of symptoms of DSM-IV anxiety and depression in children: A revised child anxiety naire: A research note. Journal of Child Psychology and Psy- and depression scale. Behaviour Research and Therapy, 38(8), chiatry, 38(5), 581–586. https://doi.org/10.1111/j.1469- 835–855. https://doi.org/10.1016/S0005-7967(99)00130-8 7610.1997.tb01545.x Cole, P. M., Hall, S. E., & Hajal, N. (2017). Emotion dysregu- Gratz, K. L. (2001). Measurement of deliberate self-harm: Pre- lation as a risk factor for psychopathology. Child and Ado- liminary data on the deliberate self-harm inventory. Journal lescent Psychopathology, 2, 341–373. of Psychopathology and Behavioral Assessment, 23(4), Cracco, E., Goossens, L., & Braet, C. (2017). Emotion regula- 253–263. https://doi.org/10.1023/A:1012779403943 Gratz, K. L., & Roemer, L. (2004). Multidimensional assess- tion across childhood and adolescence: Evidence for a maladaptive shift in adolescence. European Child & Adoles- ment of emotion regulation and dysregulation: Develop- cent Psychiatry, 26(8), 909–921. https://doi.org/10.1007/ ment, factor structure, and initial validation of the s00787-017-0952-8 difficulties in emotion regulation scale. Journal of 18 Assessment 00(0) Psychopathology and Behavioral Assessment, 26(1), 41–54. Goldsmith (Eds.), Handbook of affective sciences (pp. 347– https://doi.org/10.1023/B:JOBA.0000007455.08539.94 374). Oxford University Press. Grob, A., & Smolenski, C. (2005). Fragebogen zur Erhebung Larson, R. W., Moneta, G., Richards, M. H., & Wilson, S. der Emotionsregulation bei Kindern und Jugendlichen (2002). Continuity, stability, and change in daily emotional (FEEL-KJ) [Questionnaire for surveying emotion regula- experience across adolescence. Child Development, 73(4), tion in children and adolescents (FEEL-KJ)]. Huber 1151–1165. https://doi.org/10.1111/1467-8624.00464 Verlag. Lee, F. S., Heimer, H., Giedd, J. N., Lein, E. S., Sestan, N., Gross, J. J., & John, O. P. (2003). Individual differences in two Weinberger, D. R., & Casey, B. J. (2014). Adolescent men- emotion regulation processes: Implications for affect, rela- tal health-opportunity and obligation: Emerging neu- tionships, and well-being. Journal of Personality and Social roscience offers hope for treatments. Science, 346(6209), Psychology, 85(2), 348–362. https://doi.org/10.1037/0022- 547–549. https://doi.org/10.1126/science.1260497 3514.85.2.348 Lee, S. T. (2018). Testing for measurement invariance: Does Gullone, E., & Taffe, J. (2012). The emotion regulation ques- your measure mean the same thing for different partici- tionnaire for children and adolescents (ERQ-CA): A psy- pants? APS Observer, 31(8). https://www.psychologi- chometric evaluation. Psychological Assessment, 24(2), calscience.org/observer/testing-for-measurement-invariance Lundh, L., Karim, J., Quilisch, E., & Lundh, L.-G. (2007). 409–417. https://doi.org/10.1037/A0025777 Hofmann, S. G., Carpenter, J. K., & Curtiss, J. (2016). Inter- Deliberate self-harm in 15-year-old adolescents: A pilot personal Emotion Regulation Questionnaire (IERQ): Scale study with a modified version of the Deliberate Self-Harm development and psychometric characteristics. Cognitive Inventory. Scandinavian Journal of Psychology, 48, 33–41. Therapy and Research, 40(3), 341–356. https://doi.org/10. https://doi.org/10.1111/j.1467-9450.2006.00567.x 1007/s10608-016-9756-2 Lundh, L. G., Wa˚ ngby-Lundh, M., & Bja¨ rehed, J. (2008). Self- Hofmann, S. G., & Kashdan, T. B. (2010). The affective style reported emotional and behavioral problems in Swedish 14 questionnaire: Development and psychometric properties. to 15-year-old adolescents: A study with the self-report ver- Journal of Psychopathology and Behavioral Assessment, sion of the Strengths and Difficulties Questionnaire. Scan- 32(2), 255–263. https://doi.org/10.1007/s10862-009-9142-4 dinavian Journal of Psychology, 49(6), 523–532. https://doi. Hofmann, S. G., Sawyer, A. T., Fang, A., & Asnaani, A. org/10.1111/j.1467-9450.2008.00668.x (2012). Emotion dysregulation model of mood and anxiety Lundh, L. G., Wa˚ ngby-Lundh, M., & Bjarehed, J. (2011). disorder. Depression and Anxiety, 29, 409–416. https://doi. Deliberate self-harm and psychological problems in young org/10.1002/da.21888 adolescents: Evidence of a bidirectional relationship in girls. Hooper, D., Coughlan, J., & Mullen, M. (2007). Structural Scandinavian Journal of Psychology, 52(5), 476–483. https:// equation modeling: Guidelines for determining model fit. doi.org/10.1111/J.1467-9450.2011.00894.X The Electronic Journal of Business Research Methods, 6(1), MacCallum, R. C., Roznowski, M., & Necowitz, L. B. (1992). 53–60. Model modifications in covariance structure analysis: The Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes problem of capitalization on chance. Psychological Bulletin, in covariance structure analysis: Conventional criteria ver- 111(3), 490–504. https://doi.org/10.1037/0033-2909.111.3. sus new alternatives. Structural Equation Modeling: A Mul- 490 tidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/ Marroquı´n, B. (2011). Interpersonal emotion regulation as a 10705519909540118 mechanism of social support in depression. Clinical Psy- Huebner, E. S. (1991a). Initial development of the student’s life chology Review, 31(8), 1276–1290. https://doi.org/10.1016/j. satisfaction scale. School Psychology International, 12(3), cpr.2011.09.005 231–240. https://doi.org/10.1177/0143034391123010 McLaughlin, K. A., Hatzenbuehler, M. L., Mennin, D. S., & Huebner, E. S. (1991b). Further validation of the students’ life Nolen-Hoeksema, S. (2011). Emotion dysregulation and satisfaction scale: The independence of satisfaction and adolescent psychopathology: A prospective study. Beha- affect ratings. Journal of Psychoeducational Assessment, viour Research and Therapy, 49(9), 544–554. https://doi.org/ 9(4), 363–368. https://doi.org/10.1177/073428299100900408 10.1016/J.BRAT.2011.06.003 Johnson, D. P., & Whisman, M. A. (2013). Gender differences McRae, K., & Gross, J. J. (2020). Emotion regulation. Emo- in rumination: A meta-analysis. Personality and Individual tion, 20(1), 1–9. https://doi.org/10.1037/EMO0000703 Differences, 55(4), 367–374. https://doi.org/10.1016/J. Milfont, T. L., & Fischer, R. (2010). Testing measurement PAID.2013.03.019 invariance across groups: Applications in cross-cultural Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, research. International Journal of Psychological Research, K. R., & Walters, E. E. (2005). Lifetime prevalence and 3(1), 111–130. https://doi.org/10.21500/20112084.857 age-of-onset distributions of DSM-IV disorders in the Nolen-Hoeksema, S. (2012). Emotion regulation and psycho- national comorbidity survey replication. Archives of Gen- pathology: The role of gender. Annual Review of Clinical eral Psychiatry, 62(6), 593–602. https://doi.org/10.1001/ Psychology, 8(1), 161–187. https://doi.org/10.1146/ ARCHPSYC.62.6.593 annurev-clinpsy-032511-143109 Kopp, C. B., & Neufeld, S. J. (2003). Emotional development Nolen-Hoeksema, S., & Aldao, A. (2011). Gender and age dif- during infancy. In R. J. Davidson, K. R. Scherer & H. H. ferences in emotion regulation strategies and their ˚ Radman et al. 19 relationship to depressive symptoms. Personality and Indi- review. Journal of Youth and Adolescence, 46(2), 61–76. vidual Differences, 51(6), 704–708. https://doi.org/10.1016/ https://doi.org/10.1007/S10964-016-0585-0 Schmitt, N. (1996). Uses and abuses of coefficient alpha. Psy- J.PAID .2011.06.012 Nowlan, J. S., Wuthrich, V. M., & Rapee, R. M. (2015). Posi- chological Assessment, 8(4), 350–353. https://doi.org/10. tive reappraisal in older adults: A systematic literature 1037/1040-3590.8.4.350 review. Aging & Mental Health, 19(6), 475–484. https://doi. Sourander, A., Aromaa, M., Pihlakoski, L., Haavisto, A., org/10.1080/13607863.2014.954528 Rautava, P., Helenius, H., & Sillanpa¨ a¨ , M. (2006). Early Nozaki, Y., & Mikolajczak, M. (2020). Extrinsic emotion regu- predictors of deliberate self-harm among adolescents. A lation. Emotion, 20(1), 10–15. https://doi.org/10.1037/ prospective follow-up study from age 3 to age 15. Journal of emo0000636 Affective Disorders, 93(1–3), 87–96. https://doi.org/10.1016/ Olatunji, B. O., Naragon-Gainey, K., & Wolitzky-Taylor, K. J.JAD.2006.02.015 B. (2013). Specificity of rumination in anxiety and depres- Steinberg, L. (2005). Cognitive and affective development in sion: A multimodal meta-analysis. Clinical Psychology: Sci- adolescence. Trends in Cognitive Sciences, 9(2), 69–74. ence and Practice, 20(3), 225–257. https://doi.org/10.1037/ https://doi.org/10.1016/J.TICS.2004.12.005 H0101719 Sullivan, T. N., Helms, S. W., Kliewer, W., & Goodman, K. L. Paus, T., Keshavan, M., & Giedd, J. N. (2008). Why do many (2010). Associations between sadness and anger regulation psychiatric disorders emerge during adolescence? Nature coping, emotional expression, and physical and relational Reviews Neuroscience, 9(12), 947–957. https://doi.org/10. aggression among urban adolescents. Social Development, 1038/nrn2513 19(1), 30–51. https://doi.org/10.1111/J.1467-9507.2008.00531.X Phillips, K. F. V., & Power, M. J. (2007). A new self-report Swannell, S. v., Martin, G. E., Page, A., Hasking, P.St, & measure of emotion regulation in adolescents: The Regula- John, N. J. (2014). Prevalence of nonsuicidal self-injury in tion of Emotions Questionnaire. Clinical Psychology & Psy- nonclinical samples: Systematic review, meta-analysis and chotherapy, 14(2), 145–156. https://doi.org/10.1002/CPP.523 meta-regression. Suicide and Life-Threatening Behavior, Putnick, D. L., & Bornstein, M. H. (2016). Measurement 44(3), 273–303. https://doi.org/10.1111/SLTB.12070 invariance conventions and reporting: The state of the art Taber, K. S. (2018). The use of Cronbach’s alpha when devel- and future directions for psychological research. Develop- oping and reporting research instruments in science educa- mental Review, 41, 71–90. https://doi.org/10.1016/J.DR. tion. Research in Science Education, 48(6), 1273–1296. 2016.06.004 https://doi.org/10.1007/s11165-016-9602-2 Revelle, W. (2017). Psych: Procedures for personality and psy- Tamres, L. K., Janicki, D., & Helgeson, V. S. (2002). Sex dif- chological research [Statistical software]. ferences in coping behavior: A meta-analytic review and an Roberton, T., Daffern, M., & Bucks, R. S. (2012). Emotion examination of relative coping. Personality and Social Psy- regulation and aggression. Aggression and Violent Behavior, chology Review, 6(1), 2–30. https://doi.org/10.1207/S15327 17(1), 72–82. https://doi.org/10.1016/J.AVB.2011.09.006 957PSPR0601_1 Rood, L., Roelofs, J., Bo¨ gels, S. M., Nolen-Hoeksema, S., & Tang, J., Ma, Y., Guo, Y., Ahmed, N. I., Yu, Y., & Wang, J. Schouten, E. (2009). The influence of emotion-focused (2013). Association of aggression and non-suicidal self rumination and distraction on depressive symptoms in non- injury: A school-based sample of adolescents. PLOS ONE, clinical youth: A meta-analytic review. Clinical Psychology 8(10), Article e78149. https://doi.org/10.1371/JOURNAL. Review, 29(7), 607–616. https://doi.org/10.1016/J.CPR. PONE.0078149 2009.07.001 Thompson, R. A. (1994). Emotion regulation: A theme in Rosseel, Y. (2012). Lavaan: An R package for structural equa- search of definition. Monographs of the Society for Research tion modeling. Journal of Statistical Software, 48(2), 1–36. in Child Development, 59(2/3), 25. https://doi.org/10.2307/ https://doi.org/10.18637/jss.v048.i02 1166137 Royuela-Colomer, E., Ferna´ ndez-Gonza´ lez, L., & Orue, I. Ullman, J. B., & Bentler, P. M. (2013). Structural equation (2021). Longitudinal associations between internalizing modeling. In J. A. Schinka, W. F. Velicer & I. B. Weiner symptoms, dispositional mindfulness, rumination and (Eds.), Handbook of Psychology (2nd ed., pp. 661–690). impulsivity in adolescents. Journal of Youth and Adoles- John Wiley & Sons. https://doi.org/10.1002/9781118133880. cence, 50(10), 2067–2078. https://doi.org/10.1007/s10964- HOP202023 021-01476-2 van Buuren, S., & Groothuis-Oudshoorn, K. (2011). Mice: Saarikallio, S. (2010). Music as emotional self-regulation Multivariate imputation by chained equations in R. Journal throughout adulthood. Psychology of Music, 39(3), of Statistical Software, 45(3), 1–67. https://doi.org/10. 307–327. https://doi.org/10.1177/0305735610374894 18637/jss.v045.i03 Sass, D. A. (2016). Testing measurement invariance and com- van Lith, T., Schofield, M. J., & Fenner, P. (2013). Identifying paring latent factor means within a confirmatory factor anal- the evidence-base for art-based practices and their potential ysis framework. Journal of Psychoeducational Assessment, benefit for mental health recovery: A critical review. Dis- 29(4), 347–363. https://doi.org/10.1177/0734282911406661 ability and Rehabilitation, 35(16), 1309–1323. https://doi. Schafer, J. O., Naumann, E., Holmes, E. A., Tuschen-Caffier, ¨ org/10.3109/09638288.2012.732188 B., & Samson, A. C. (2016). Emotion regulation strategies in Viborg, N., Wa˚ ngby-Lundh, M., & Lundh, L. G. (2014). Reci- depressive and anxiety symptoms in youth: A meta-analytic procal prospective associations between disordered eating 20 Assessment 00(0) and other psychological problems in a community sample 155–168. https://doi.org/10.1097/00004703-200604000- of Swedish adolescent girls. Eating Behaviors, 15(1), 00014 159–163. https://doi.org/10.1016/J.EATBEH.2013.11.002 Zeman, J., Klimes-Dougan, B., Cassano, M., & Adrian, M. Waaddegaard, M., Thoning, H., & Petersson, B. (2003). Vali- (2007). Measurement issues in emotion research with chil- dation of a screening instrument for identifying risk beha- dren and adolescents. Clinical Psychology: Science and viour related to eating disorders. European Eating Disorders Practice, 14(4), 377–401. https://doi.org/10.1111/j.1468- Review, 11(6), 433–455. https://doi.org/https://doi.org/10. 2850.2007.00098.x 1002/erv.537 Zeman, J., Shipman, K., & Penza-Clyve, S. (2001). Develop- Wolff, J. C., Thompson, E., Thomas, S. A., Nesi, J., Bettis, A. ment and initial validation of the children’s sadness man- H., Ransford, B., Scopelliti, K., Frazier, E. A., & Liu, R. T. agement scale. Journal of Nonverbal Behavior, 25(3), (2019). Emotion dysregulation and non-suicidal self-injury: 187–205. https://doi.org/10.1023/A:1010623226626 A systematic review and meta-analysis. European Psychia- Zhao, K., Bai, Z. G., Bo, A., & Chi, I. (2016). A systematic try, 59, 25–36. https://doi.org/10.1016/J.EURPSY.2019.03. review and meta-analysis of music therapy for the older adults with depression. International Journal of Geriatric Psy- Wolgast, M., & Lundh, L. G. (2017). Is distraction an adaptive chiatry, 31(11), 1188–1198. https://doi.org/10.1002/GPS.4494 or maladaptive strategy for emotion regulation? A person- Zhou, Y., Daukantaite, _ D., Lundh, L. G., Wa˚ ngby-Lundh, oriented approach. Journal of Psychopathology and Beha- M., & Ryde, A. (2020). Adolescents’ emotion regulation vioral Assessment, 39(1), 117–127. https://doi.org/10.1007/ strategies questionnaire: Initial validation and prospective s10862-016-9570-x associations with nonsuicidal self-injury and other mental Yih, J., Uusberg, A., Taxer, J. L., & Gross, J. J. (2018). Better health problems in adolescence and young adulthood in a together: A unified perspective on appraisal and emotion Swedish Youth Cohort. Frontiers in Psychiatry, 11, 462. regulation. Cognition and Emotion, 33(1), 41–47. https:// https://doi.org/10.3389/fpsyt.2020.00462 doi.org/10.1080/02699931.2018.1504749 Zimmermann, P., & Iwanski, A. (2014). Emotion regulation Zaki, J., & Williams, W. C. (2013). Interpersonal emotion reg- from early adolescence to emerging adulthood and middle ulation. Emotion, 13(5), 803–810. https://doi.org/10.1037/ adulthood: Age differences, gender differences, and a0033839 emotion-specific developmental variations. International Zeman, J., Cassano, M., Perry-Parrish, C., & Stegall, S. Journal of Behavioral Development, 38(2), 182–194. https:// (2006). Emotion regulation in children and adolescents. doi.org/10.1177/0165025413515405 Journal of Developmental & Behavioral Pediatrics, 27(2), http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Assessment SAGE http://www.deepdyve.com/lp/sage/adolescents-emotion-regulation-strategies-questionnaire-extended-dAtn07h8mC

Loading next page...

References (95)

L. Rood, J. Roelofs, S. Bögels, S. Nolen-Hoeksema, E. Schouten (2009)
The influence of emotion-focused rumination and distraction on depressive symptoms in non-clinical youth: a meta-analytic review.
Clinical psychology review, 29 7
E. Huebner (1991)
Further Validation of the Students' Life Satisfaction Scale: The Independence of Satisfaction and Affect Ratings
Journal of Psychoeducational Assessment, 9
D. Sass (2011)
Testing Measurement Invariance and Comparing Latent Factor Means Within a Confirmatory Factor Analysis Framework
Journal of Psychoeducational Assessment, 29
Estíbaliz Royuela-Colomer, L. Fernández-González, I. Orue (2021)
Longitudinal Associations between Internalizing Symptoms, Dispositional Mindfulness, Rumination and Impulsivity in Adolescents
Journal of Youth and Adolescence, 50
S. Swannell, G. Martin, A. Page, P. Hasking, Nathan John (2014)
Prevalence of nonsuicidal self-injury in nonclinical samples: systematic review, meta-analysis and meta-regression.
Suicide & life-threatening behavior, 44 3
Molly Adrian, J. Zeman, G. Veits (2011)
Methodological implications of the affect revolution: a 35-year review of emotion regulation assessment in children.
Journal of experimental child psychology, 110 2
Li-tze Hu, P. Bentler (1999)
Cutoff criteria for fit indexes in covariance structure analysis : Conventional criteria versus new alternatives
Structural Equation Modeling, 6
R. Brunner, P. Parzer, J. Haffner, R. Steen, J. Roos, M. Klett, F. Resch (2007)
Prevalence and psychological correlates of occasional and repetitive deliberate self-harm in adolescents.
Archives of pediatrics & adolescent medicine, 161 7
Daniel Johnson, M. Whisman (2013)
Gender differences in rumination: A meta-analysis.
Personality and individual differences, 55 4
J. Schäfer, E. Naumann, E. Holmes, B. Tuschen-Caffier, Andrea Samson (2016)
Emotion Regulation Strategies in Depressive and Anxiety Symptoms in Youth: A Meta-Analytic Review
Journal of Youth and Adolescence, 46
Ross Thompson (1994)
Emotion regulation: a theme in search of definition.
Monographs of the Society for Research in Child Development, 59 2-3
A. Sourander, M. Aromaa, L. Pihlakoski, A. Haavisto, P. Rautava, H. Helenius, M. Sillanpää (2006)
Early predictors of deliberate self-harm among adolescents. A prospective follow-up study from age 3 to age 15.
Journal of affective disorders, 93 1-3
A. Aldao, S. Nolen-Hoeksema, S. Schweizer (2010)
Emotion-regulation strategies across psychopathology: A meta-analytic review.
Clinical psychology review, 30 2
R. Kessler, P. Berglund, Olga Demler, R. Jin, K. Merikangas, E. Walters (2005)
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication.
Archives of general psychiatry, 62 6
S. Nolen-Hoeksema, A. Aldao (2011)
Gender and age differences in emotion regulation strategies and their relationship to depressive symptoms
Personality and Individual Differences, 51
L. Steinberg (2005)
Cognitive and affective development in adolescence
Trends in Cognitive Sciences, 9
R. Goodman (1997)
The Strengths and Difficulties Questionnaire: a research note.
Journal of child psychology and psychiatry, and allied disciplines, 38 5
S. Buuren, K. Groothuis-Oudshoorn (2011)
MICE: Multivariate Imputation by Chained Equations in R
Journal of Statistical Software, 45
N. Schmitt (1996)
Uses and abuses of coefficient alpha.
Psychological Assessment, 8
K. McLaughlin, M. Hatzenbuehler, D. Mennin, S. Nolen-Hoeksema (2011)
Emotion dysregulation and adolescent psychopathology: a prospective study.
Behaviour research and therapy, 49 9
T. Sullivan, Sarah Helms, W. Kliewer, Kimberly Goodman (2010)
Associations between Sadness and Anger Regulation Coping, Emotional Expression, and Physical and Relational Aggression among Urban Adolescents.
Social development, 19 1
N. Viborg, Margit Wångby-Lundh, L. Lundh (2014)
Reciprocal prospective associations between disordered eating and other psychological problems in a community sample of Swedish adolescent girls.
Eating behaviors, 15 1
Terri Roberton, M. Daffern, R. Bucks (2012)
Emotion regulation and aggression
Aggression and Violent Behavior, 17
Y. Nozaki, Moïra Mikolajczak (2019)
Extrinsic emotion regulation.
Emotion, 20 1
E. Huebner (1991)
Initial Development of the Student's Life Satisfaction Scale
School Psychology International, 12
Jie Tang, Ying Ma, Yongjie Guo, N. Ahmed, Yizhen Yu, Jiaji Wang (2013)
Association of Aggression and Non-Suicidal Self Injury: A School-Based Sample of Adolescents
PLoS ONE, 8
K. Phillips, M. Power (2007)
A new self‐report measure of emotion regulation in adolescents: The Regulation of Emotions Questionnaire
Clinical Psychology & Psychotherapy, 14
T. Milfont, R. Fischer (2010)
Testing measurement invariance across groups: applications in cross-cultural research.
International journal of psychological research, 3
Lisa Barrett, E. Bliss-Moreau, L. Barrett, Feldman Barrett
She's Emotional. He's Having a Bad Day: Attributional Explanations for Emotion Stereotypes Are Women Really More Emotional?
J. Zeman, B. Klimes-Dougan, M. Cassano, Molly Adrian (2007)
Measurement Issues in Emotion Research With Children and Adolescents
Clinical Psychology-science and Practice, 14
Margaret Andover, Blair Morris (2014)
Expanding and Clarifying the Role of Emotion Regulation in Nonsuicidal Self-Injury
The Canadian Journal of Psychiatry, 59
Suvi Saarikallio (2011)
Music as emotional self-regulation throughout adulthood
Psychology of Music, 39
P. Zimmermann, Alexandra Iwanski (2014)
Emotion regulation from early adolescence to emerging adulthood and middle adulthood
International Journal of Behavioral Development, 38
E. Gullone, J. Taffe (2012)
The Emotion Regulation Questionnaire for Children and Adolescents (ERQ-CA): a psychometric evaluation.
Psychological assessment, 24 2
Y. Rosseel (2012)
lavaan: An R Package for Structural Equation Modeling
Journal of Statistical Software, 48
K. L. Gratz, L. Roemer (2004)
Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale
Journal of Psychopathology and Behavioral Assessment, 26
F. Lee, H. Heimer, J. Giedd, E. Lein, N. Šestan, D. Weinberger, B. Casey (2014)
Adolescent mental health—Opportunity and obligation
Science, 346
J. Geipel, J. Koenig, T. Hillecke, F. Resch, M. Kaess (2018)
Music-based interventions to reduce internalizing symptoms in children and adolescents: A meta-analysis.
Journal of affective disorders, 225
M. Waaddegaard, H. Thoning, B. Petersson (2003)
Validation of a screening instrument for identifying risk behaviour related to eating disorders
European Eating Disorders Review, 11
M. Wolgast, L. Lundh (2016)
Is Distraction an Adaptive or Maladaptive Strategy for Emotion Regulation? A Person-Oriented Approach
Journal of Psychopathology and Behavioral Assessment, 39
P. Bender, M. Reinholdt-Dunne, B. Esbjørn, F. Pons (2012)
Emotion dysregulation and anxiety in children and adolescents: Gender differences
Personality and Individual Differences, 53
Emiel Cracco, L. Goossens, C. Braet (2017)
Emotion regulation across childhood and adolescence: evidence for a maladaptive shift in adolescence
European Child & Adolescent Psychiatry, 26
R. Maccallum, M. Roznowski, Lawrence Necowitz (1992)
Model modifications in covariance structure analysis: the problem of capitalization on chance.
Psychological bulletin, 111 3
D. Daukantaitė, L. G. Lundh, M. Wångby-Lundh (2019)
Association of direct and indirect aggression and victimization with self-harm in young adolescents: A person-oriented approach
Development and Psychopathology, 31
N. L. Afanador, T. Tran, L. Blanchet, R. Baumgartner (2021)
Mvdalab: Multivariate data analysis laboratory
P. M. Cole, S. E. Hall, N. Hajal (2017)
Emotion dysregulation as a risk factor for psychopathology
Child and Adolescent Psychopathology, 2
L. Lundh, J. Karim, Eva Quilisch (2007)
Deliberate self-harm in 15-year-old adolescents: a pilot study with a modified version of the Deliberate Self-Harm Inventory.
Scandinavian journal of psychology, 48 1
L. J. Cronbach (1951)
Coefficient alpha and the internal structure of tests
Psychometrika, 16
Ya Zhou, D. Daukantaitė, L. Lundh, Margit Wångby-Lundh, Adam Ryde (2020)
Adolescents’ Emotion Regulation Strategies Questionnaire: Initial Validation and Prospective Associations With Nonsuicidal Self-Injury and Other Mental Health Problems in Adolescence and Young Adulthood in a Swedish Youth Cohort
Frontiers in Psychiatry, 11
L. Lundh, Margit Wångby-Lundh, Jonas Bjärehed (2011)
Deliberate self-harm and psychological problems in young adolescents: evidence of a bidirectional relationship in girls.
Scandinavian journal of psychology, 52 5
N. Eisenberg, A. Cumberland, Tracy Spinrad (1998)
Parental Socialization of Emotion.
Psychological inquiry, 9 4
A. Aldao, D. G. Gee, A. de Los Reyes, I. Seager (2016)
Emotion regulation as a transdiagnostic factor in the development of internalizing and externalizing psychopathology: Current and future directions
Development and Psychopathology, 28
B. F. Chorpita, L. Yim, C. Moffitt, L. A. Umemoto, S. E. Francis (2000)
Assessment of symptoms of DSM-IV anxiety and depression in children: A revised child anxiety and depression scale
Behaviour Research and Therapy, 38
Saz Ahmed, Amanda Bittencourt-Hewitt, C. Sebastian (2015)
Neurocognitive bases of emotion regulation development in adolescence
Developmental Cognitive Neuroscience, 15
A. G. Billings, R. H. Moos (1981)
The role of coping responses and social resources in attenuating the stress of life events
Journal of Behavioral Medicine, 4
S. Hofmann, Joseph Carpenter, Joshua Curtiss (2016)
Interpersonal Emotion Regulation Questionnaire (IERQ): Scale Development and Psychometric Characteristics
Cognitive Therapy and Research, 40
T. Brockmeyer, Mandy Skunde, Mudan Wu, Esther Bresslein, G. Rudofsky, W. Herzog, H. Friederich (2014)
Difficulties in emotion regulation across the spectrum of eating disorders.
Comprehensive psychiatry, 55 3
K. Dixon-Gordon, Samantha Bernecker, K. Christensen (2015)
Recent innovations in the field of interpersonal emotion regulation
Current opinion in psychology, 3
J. Wolff, Elizabeth Thompson, Sarah Thomas, J. Nesi, Alexandra Bettis, Briana Ransford, Katie Scopelliti, Elisabeth Frazier, Richard Liu (2019)
Emotion dysregulation and non-suicidal self-injury: A systematic review and meta-analysis
European Psychiatry, 59
A. Dingemans, U. Danner, M. Parks (2017)
Emotion Regulation in Binge Eating Disorder: A Review
Nutrients, 9
L. Lundh, Margit Wångby-Lundh, Jonas Bjärehed (2008)
Self-reported emotional and behavioral problems in Swedish 14 to 15-year-old adolescents: a study with the self-report version of the Strengths and Difficulties Questionnaire.
Scandinavian journal of psychology, 49 6
Kateri McRae, J. Gross (2020)
Emotion regulation.
Emotion, 20 1
Jonas Bjärehed, L. Lundh (2008)
Deliberate Self‐Harm in 14‐Year‐Old Adolescents: How Frequent Is It, and How Is It Associated with Psychopathology, Relationship Variables, and Styles of Emotional Regulation?
Cognitive Behaviour Therapy, 37
B. Marroquín (2011)
Interpersonal emotion regulation as a mechanism of social support in depression.
Clinical psychology review, 31 8
G. A. Bonanno, C. L. Burton (2013)
Regulatory flexibility: An individual differences perspective on coping and emotion regulation
Perspectives on Psychological Science, 8
Jamil Zaki, W. Williams (2013)
Interpersonal emotion regulation.
Emotion, 13 5
Theresa Lith, M. Schofield, Patricia Fenner (2013)
Identifying the evidence-base for art-based practices and their potential benefit for mental health recovery: A critical review
Disability and Rehabilitation, 35
Jennifer Yih, A. Uusberg, J. Taxer, J. Gross (2018)
Better together: a unified perspective on appraisal and emotion regulation
Cognition and Emotion, 33
J. Zeman, M. Cassano, C. Perry-Parrish, Sheri Stegall (2006)
Emotion Regulation in Children and Adolescents
Journal of Developmental & Behavioral Pediatrics, 27
A. Aldao (2013)
The Future of Emotion Regulation Research
Perspectives on Psychological Science, 8
S. Hofmann, Alice Sawyer, A. Fang, Anu Asnaani (2012)
EMOTION DYSREGULATION MODEL OF MOOD AND ANXIETY DISORDERS
Depression and Anxiety, 29
K. L. Gratz (2001)
Measurement of deliberate self-harm: Preliminary data on the deliberate self-harm inventory
Journal of Psychopathology and Behavioral Assessment, 23
J. Gross, O. John (2003)
Individual differences in two emotion regulation processes: implications for affect, relationships, and well-being.
Journal of personality and social psychology, 85 2
N. Garnefski, V. Kraaij, P. Spinhoven (2001)
Negative life events, cognitive emotion regulation and emotional problems
Personality and Individual Differences, 30
G. Bonanno, C. Burton (2013)
Regulatory Flexibility
Perspectives on Psychological Science, 8
A. Aldao (2013)
The future of emotion regulation research: Capturing context
Perspectives on Psychological Science, 8
K. Taber (2017)
The Use of Cronbach’s Alpha When Developing and Reporting Research Instruments in Science Education
Research in Science Education, 48
S. Hofmann, T. Kashdan (2010)
The Affective Style Questionnaire: Development and Psychometric Properties
Journal of Psychopathology and Behavioral Assessment, 32
T. Paus, M. Keshavan, J. Giedd (2008)
Why do many psychiatric disorders emerge during adolescence?
Nature Reviews Neuroscience, 9
P. Boxer (2010)
Covariation of self- and other-directed aggression among inpatient youth: continuity in the transition to treatment and shared risk factors.
Aggressive behavior, 36 3
L. F. Barrett, E. Bliss-Moreau (2009)
She’s emotional. He’s having a bad day: Attributional explanations for emotion stereotypes
Emotion, 9
A. Cropley (1990)
Creativity and mental health in everyday life.
Creativity Research Journal, 3
R. Larson, G. Moneta, M. Richards, Suzanne Wilson (2002)
Continuity, stability, and change in daily emotional experience across adolescence.
Child development, 73 4
H. Fliege, Jeong-Ran Lee, A. Grimm, B. Klapp (2009)
Risk factors and correlates of deliberate self-harm behavior: a systematic review.
Journal of psychosomatic research, 66 6
D. Putnick, M. Bornstein (2016)
Measurement Invariance Conventions and Reporting: The State of the Art and Future Directions for Psychological Research.
Developmental review : DR, 41
(2015)
Validation of the FEEL-KJ: An Instrument to Measure Emotion Regulation Strategies in Children and Adolescents
PLoS ONE, 10
S. Nolen-Hoeksema (2012)
Emotion regulation and psychopathology: the role of gender.
Annual review of clinical psychology, 8
Jamie Nowlan, V. Wuthrich, R. Rapee (2015)
Positive reappraisal in older adults: a systematic literature review
Aging & Mental Health, 19
B. Olatunji, K. Naragon-Gainey, K. Wolitzky-Taylor (2013)
Specificity of Rumination in Anxiety and Depression: A Multimodal Meta‐Analysis
Clinical Psychology-science and Practice, 20
K. Zhao, Z. Bai, Ai Bo, I. Chi (2016)
A systematic review and meta‐analysis of music therapy for the older adults with depression
International Journal of Geriatric Psychiatry, 31
B. Casey, S. Getz, A. Galván (2008)
The adolescent brain.
Developmental review : DR, 28 1
Theodore Beauchaine (2015)
Future Directions in Emotion Dysregulation and Youth Psychopathology
Journal of Clinical Child & Adolescent Psychology, 44
F. Chen (2007)
Sensitivity of Goodness of Fit Indexes to Lack of Measurement Invariance
Structural Equation Modeling: A Multidisciplinary Journal, 14
E. Calvete, I. Orue, B. Hankin (2015)
Cross-lagged associations among ruminative response style, stressors, and depressive symptoms in adolescents
Journal of Social and Clinical Psychology, 34
Chad Ebesutani, S. Reise, B. Chorpita, C. Ale, Jennifer Regan, John Young, C. Higa-McMillan, J. Weisz (2012)
The Revised Child Anxiety and Depression Scale-Short Version: scale reduction via exploratory bifactor modeling of the broad anxiety factor.
Psychological assessment, 24 4

Publisher: SAGE
Copyright: © The Author(s) 2023
ISSN: 1073-1911
eISSN: 1552-3489
DOI: 10.1177/10731911231164619
Publisher site: See Article on Publisher Site

Abstract

Emotion regulation (ER) is implicated in a range of psychopathologies and behavioral problems that are prevalent or have their initial onset in adolescence. In this study, we aim to evaluate the psychometric properties (factor structure, internal consistency, and construct validity) of the Adolescents’ Emotion Regulation Strategies Questionnaire–Extended (AERSQ-E), a modified and extended version of an ER instrument developed by Zhou et al. Across six sub-studies using data from different Swedish adolescent community samples (1,104 students in total), we generated and validated a 23-item version containing six subscales: rumination/negative thinking, positive reorientation, creative expression, aggressive outlet, social support, and distraction. Assessing test–retest reliability, internal consistency, measurement invariance as well as convergent and discrimi- nant validity, we could establish, with some limitations, the general reliability and validity of the AERSQ-E as a valid measure of ER strategies for use in adolescence. Keywords emotion regulation, adolescence, psychopathology, psychometrics, questionnaire Emotions are intimately bound to regulatory processes behaviors (Aldao et al., 2010). Some examples include that continuously monitor and adjust their expression non-suicidal self-injury (NSSI) (Andover & Morris, (Yih et al., 2018). Emotion regulation (ER) describes 2014; Wolff et al., 2019); eating disorders and disordered these extrinsic and intrinsic processes that monitor the eating (DE) (Brockmeyer et al., 2014; Dingemans et al., demands of the current environment, evaluate what con- 2017); mood and anxiety disorders (Hofmann et al., stitutes a contextually appropriate response, and modify 2012); and aggression (Roberton et al., 2012; Sullivan emotional reactions accordingly (Thompson, 1994). et al., 2010). Hence, ER is crucially important for adaptive function- Many of these psychological and behavioral prob- ing in that it helps modulate the intensity and temporal- lems have their initial onset in adolescence, a period of ity of an emotional response in proportion to current rapid change in, not least, the underlying structures sup- events and in accordance with personal goals (Aldao, porting ER (Ahmed et al., 2015). Thus, identifying func- 2013). tional and dysfunctional patterns of ER in this critical ER theoretically implies both the regulation of nega- period is of high clinical relevance, as is by consequence tive and positive affect. However, dysfunctions in the the need for valid and reliable instruments to measure regulation of negative affect are particularly linked to ER, particularly in adolescence. Although numerous psychopathology (Beauchaine, 2015; Cole et al., 2017; McLaughlin et al., 2011) and have been suggested as a Lund University, Sweden transdiagnostic factor in the development and mainte- nance of both internalizing and externalizing psycholo- Corresponding Author: gical problems (Aldao et al., 2016). Difficulties in Daiva Daukantaite, _ Department of Psychology, Lund University, Box 213, regulating negative affect have been implicated in a wide 22100 Lund, Sweden. range of specific psychopathologies and destructive Email: daiva.daukantaite@psy.lu.se 2 Assessment 00(0) self-report instruments have been developed to assess undoubtedly important. Nevertheless, regardless of age, ER (e.g., Garnefski et al., 2001; Gratz & Roemer, we often turn to others for help with regulating our emo- 2004; Grob & Smolenski, 2005; Gross & John, 2003; tions, and research on the use of such interpersonal ER Hofmann et al., 2016; Hofmann & Kashdan, 2010; strategies has seen increasing research attention in recent Phillips & Power, 2007; Zeman et al., 2001; Zhou et al., years (e.g., Dixon-Gordon et al., 2015; Marroquı´n, 2011; 2020), the majority of these have been developed for Nozaki & Mikolajczak, 2020; Zaki & Williams, 2013). In measuring ER in adults or children, with only a few spe- the AERSQ, at least one subscale falls within each of these cifically targeting an adolescent demographic. categorizations. In this study, we sought to evaluate the psychometric In addition, ER strategies can be categorized as adap- properties of a further developed instrument, the tive or maladaptive, in the sense that they are either Adolescents’ Emotion Regulation Strategies Questionnaire effective or ineffective in modifying emotions and/or (AERSQ), a scale tailored to measure ER strategies used they are associated with either long-term negative or to regulate negative affect in adolescents (Zhou et al., positive outcomes (Aldao et al., 2010). Some ER strate- 2020). We sought to improve its psychometric properties gies are broadly characterized as maladaptive (e.g., and to extend it by adding and validating an additional expressive suppression or experiential avoidance) and ER strategy measure—aggressive outlet—which may be others as adaptive (e.g., acceptance or reappraisal). especially important for adolescents who self-injure Although evidence supports such a distinction at least as (Daukantaite _ et al., 2019; Tang et al., 2013). As part of a useful heuristic (e.g., Scha¨ fer et al., 2016), it is impor- the validating procedure for this extended version tant to note that no strategy is inherently one or the (referred to as AERSQ-E), we also explore the associa- other. Instead, adaptive ER is arguably best understood tions between adolescents’ ER strategies and problematic as the appropriate and effective use of ER strategies con- behaviors such as NSSI and DE, as well as mental health sidering the context, which can include specific aspects problems across several community samples of Swedish of the culture one finds oneself in, as well as personal adolescents. goals both immediate and long-term (Aldao, 2013). Conversely, maladaptive ER or emotion dysregulation can either imply a lack of successful regulatory action Classification of Emotion Regulation (e.g., rumination) or the inappropriate and ineffective use of ER strategies in considering personal goals (e.g., Strategies aggressive behavior such as lashing out at someone, There are many mechanisms by which we regulate our despite valuing their friendship). For example, distrac- emotions and several dimensions by which we concep- tion has an ambiguous status in that when combined tually classify ER strategies. The relationship between with an attitude of acceptance rather than avoidance, ER strategies and psychopathology varies depending on distraction can be supportive of positive psychological the specific strategy (Aldao et al., 2010). Thus, a key tar- development (Wolgast & Lundh, 2017). The AERSQ get of ER research is the mapping of ER profiles and aims to cover dimensions that are typically viewed as how they relate to positive and negative indices of men- both maladaptive and adaptive. tal health. This, in turn, requires tools that assess a broad range of ER types. One way to differentiate between different ER strate- Emotion Regulation Development and gies is based on the processes they primarily engage in. Psychopathology in Adolescence For instance, some ER strategies are more oriented toward changes in cognition (e.g., cognitive reappraisal, Developmental research suggests that children initially rumination) and some more oriented toward behavioral rely on caregivers to regulate their emotions and that changes (e.g., eating, workout, substance use, expressive they gradually come to internalize their ER abilities suppression, or aggression). Some ER strategies primarily across early childhood (Eisenberg et al., 1998; Kopp & rely on intrinsic or intrapersonal processes (e.g., distrac- Neufeld, 2003). These abilities further develop as chil- tion, expressive suppression), and some rely primarily on dren reach adolescence (Zeman et al., 2006), suggesting extrinsic or interpersonal processes, that is, regulation that as children mature, they gradually become more through social engagement with others (e.g., seeking con- skillful at regulating their own emotions. Despite this, solation, social modeling; Hofmann et al., 2016; Phillips adolescence is a period characterized by greater emo- & Power, 2007). Historically, intrinsic ER have been over- tional instability and negative affectivity (Larson et al., emphasized in ER research (McRae & Gross, 2020), and 2002), and many psychopathologies implicating ER further research into understanding the development of tend to emerge during adolescence (Kessler et al., 2005; these intrinsic ER strategies across development is F. S. Lee et al., 2014; Paus et al., 2008). ˚ Radman et al. 3 Some studies have highlighted important differences adults. This is problematic because adolescence differs in ER strategy use that characterize adolescence as com- from both earlier childhood and adulthood across a range of cognitive and self-evaluative capabilities, which pared with other age groups. Cracco and colleagues suggest that the same set of questions may be interpreted (2017) looked at the frequency of use for seven typically differently by adolescents than the intended age group adaptive and five typically maladaptive ER strategies (Zeman et al., 2007). For instance, the Emotion and found evidence suggesting that in adolescence there Regulation Questionnaire (ERQ; Gross & John, 2003) is a shift toward more frequent use of maladaptive ER was originally developed for adults and adapted for use strategies. Zimmermann and Iwanski (2014) looked at with children and adolescents by Gullone and Taffe adolescents’ and adults’ subjective beliefs about their (2012). It includes items such as ‘‘When I’m worried regulation of negative emotions such as anger, fear, and about something, I make myself think about it in a way sadness. They found that individuals aged 13 to 15 that helps me feel better,’’ which require relatively elabo- report having a smaller repertoire of ER strategies at rate self-evaluative knowledge about the interrelation their disposal than both preceding and succeeding age between cognition, emotion, and behavior. Although groups. The combination of a general shift toward more difficult in practice, it is crucial that measures tailored maladaptive types of ER with a general reduction of the for adolescents reduce such complex meta-cognitive ER strategy repertoire available suggest a particular vul- requirements to a minimum, as such capabilities may nerability with regard to emotional symptoms during be insufficiently developed (Casey et al., 2008). At the adolescence that can in part be accounted for by changes same time, adolescents, compared to younger children, in ER during the period. tend to exhibit advancements in areas such as abstract The social reinforcement of ER is an additional, and hypothetical thinking, logical reasoning, and potentially interacting, factor, as the use of many ER information-processing efficiency (Steinberg, 2005). strategies is necessarily linked to a social context. Therefore, it is appropriate to expect them to handle at Cultural factors, for instance, are likely to have an influ- least moderately more complex inquiries. To achieve ence on the adaptiveness of different ER strategies. This the fine balance that meet adolescents at their own includes how expressions of emotion tend to be inter- level, we must design instruments specifically tailored preted differently comparing men and women (Barrett & to their unique set of cognitive and metacognitive abil- Bliss-Moreau, 2009), which in turn might reinforce the ities, which has been a key focus of the AERSQ. use of specific regulatory strategies. Correspondingly, In addition, most of the work on ER both in adults studies have documented gender differences in the use of and in young samples emphasize intrinsic ER strategies specific ER strategies (e.g., Johnson & Whisman, 2013; (McRae & Gross, 2020). This can be seen in that many Nolen-Hoeksema & Aldao, 2011; Tamres et al., 2002) of the most popular self-report instruments today only and linked differences to the prevalence of various psy- assess intrapersonal ER strategies, such as the ERQ, the chopathologies such as anxiety, depression, and alcohol Difficulties in Emotion Regulation Scale (DERS; Gratz abuse (e.g., Bender et al., 2012; Nolen-Hoeksema, 2012). & Roemer, 2004), and the Cognitive Emotion However, the exact role of gender in the link between ER Regulation Questionnaire (CERQ; Garnefski et al., and psychopathology is not entirely understood (for a 2001). An exception is the Interpersonal Emotion discussion on the subject, see Nolen-Hoeksema, 2012). Regulation Questionnaire (IERSQ; Hofmann et al., 2016), which explicitly target extrinsic ER. However, the IERSQ was not developed with adolescents specifically Measuring ER in Adolescence and the in mind. Moreover, it only measures extrinsic ER and Rationale Behind the AERSQ must therefore be combined with scales targeting intrin- Given the importance of ER in adolescent psycho- sic ER (e.g., the DERS and CERQ) for an encompass- pathology, it is important that we develop valid mea- ing assessment of ER. The Regulation of Emotion sures of ER specifically tailored toward adolescents. In a Questionnaire (REQ; Phillips & Power, 2007) and review by Adrian and colleagues (2011) that examines FEEL-KJ (Grob & Smolenski, 2005) are perhaps the the assessment of ER, it was estimated that 44% of ER most comprehensive alternatives that have been specifi- studies that focused on middle childhood (age 6–12) cally developed for adolescents. REQ encompasses four deployed self-report measures, whereas 92.6% of studies subscales dividing ER strategies into internal-functional, that focused on adolescents (age 13–18) deployed self- internal-dysfunctional, external-functional, and exter- report measures. Clearly, self-report has become the nal-dysfunctional. However, as previously discussed, the method of choice in measuring ER among adolescents. functionality of ER strategies is partly dependent on However, a large portion of these studies used self- contextual factors rather than the type of ER strategy report instruments originally developed for children or alone (Aldao, 2013; Bonanno & Burton, 2013), and the 4 Assessment 00(0) AERSQ, in contrast to REQ, aims to delineate between during adolescence in the study by Cracco and col- strategies based on their presumed underlying processes leagues (2017), implying that it is an especially relevant rather than their presumed functionality. In addition, measure to capture when studying adolescent ER. the REQ is yet to be extensively validated, with the origi- Finally, previous studies have suggested that aggressive nal sample only consisting of 225 adolescents (12-19 tendencies in adolescence—including experiencing or years) with exploratory and confirmatory factor analy- regulating emotions such as anger toward oneself, self- ses performed on the same sample (Phillips & Power, hatred, or anger toward others—is associated with enga- 2007). FEEL-KJ is a comprehensive measure compris- ging in NSSI (Boxer, 2010; Brunner et al., 2007; ing 12 subscales that assess 15 ER strategies. Although Daukantaite _ et al., 2019; Fliege et al., 2009; Sourander showing promising psychometric properties in a valida- et al., 2006; Tang et al., 2013), making it an especially important measure for those who self-injure, with NSSI tion study (Cracco et al., 2015), it is a lengthier instru- showing the highest rates of lifetime prevalence during ment containing 90 items. In contrast, the AERSQ adolescence (Swannell et al., 2014). encompasses a wide range of ER types while maintain- ing a succinct format. This can be beneficial to projects that encompass a wide array of different questionnaires, The Current Study where brevity is an important consideration. In this study, we aim to evaluate the psychometric Finally, although some ER measures show good psy- properties (factor structure, internal consistency, and chometric properties when applied to adolescent sam- construct validity) of a modified and extended ples, the number of strategies assessed across these version of the AERSQ, referred to as the Adolescents’ measures is limited. For instance, the ERQ only assesses Emotion Regulation Strategies Questionnaire—Extended positive reappraisal and expressive suppression (Gross (AERSQ-E). As part of the validation process, we also & John, 2003). The CERQ, although developed for ado- intended to establish the convergent and discriminant lescents, focuses on cognitive ER strategies such as validity of the AERSQ-E by exploring the associations acceptance, rumination, catastrophizing, and positive between adolescents’ ER strategies and NSSI, DE, as refocusing (Garnefski et al., 2001). The AERSQ encom- well as other aspects of mental health including interna- passes dimensions of ER captured by some of these lizing problems (e.g., emotional symptoms, depression, instruments (e.g., rumination, positive reorientation, and anxiety), externalizing problems (e.g., peer problems distraction) but also includes at least one dimension of and conduct problems), and positive functioning (life ER not assessed by other instruments that we are aware satisfaction). We predicted that the associations for sub- of, namely, the use of expressive/creative behaviors (e.g., scales existing in the previous version would align with writing down thoughts about felt emotions) to cope with those previously found (see Zhou et al., 2020). That is, we negative affect (assessed by the subscale previously expected rumination/negative thinking to be positively called ‘‘cultural activities,’’ here renamed ‘‘creative associated with constructs related to negative functioning expression’’). Across the different ER strategy measures, and negatively associated with indicators of positive func- it is more common to include inhibitions of emotional tioning (i.e., life satisfaction), while we expected the oppo- expression (e.g., Gross & John, 2003; Hofmann & site pattern for positive reorientation and social support. Kashdan, 2010; Zeman et al., 2001), and few instru- Furthermore, we expected the creative expression subscale ments focus on actualized expression of emotions. Art- and the distraction subscale to only show weak or no based activities have previously been linked to positive associations to other variables. Finally, for the new theo- mental health and to ER in particular (e.g., Geipel et al., retical subscale aiming to capture aggression as an ER 2018; Saarikallio, 2010; van Lith et al., 2013; Zhao et al., strategy, we expected to find positive associations to most 2016), making it an important factor to consider. negative constructs including NSSI, anxiety, and depres- In a further developed version of the AERSQ, we sion, as well as various psychological difficulties including aimed to add a second expressive subscale intending to conduct, emotion, and peer problems, which would corro- capture aggression as an ER strategy. This inclusion was borate the status of aggression as a maladaptive type of motivated in several ways. First, when reanalyzing data ER and the previously discussed evidence linking it to from earlier iterations of the AERSQ, such a factor engagement with NSSI. emerged in the larger item pool. Second, aggression as an ER strategy has been successfully included in some previous ER instruments targeting adolescents such as Method FEEL-KJ (Grob & Smolenski, 2005) and REQ (Phillips Short Overview & Power, 2007), suggesting its relevance to the target age group. Third, the use of aggression to regulate affect The AERSQ presents respondents with a list of possible was among the maladaptive ER strategies that peaked behaviors and ways of thinking and asks respondents to ˚ Radman et al. 5 judge on a 5-point scale ranging from 1 (never)to5(very In the remaining three samples (4, 5, and 6), the psy- often) how often they engage in each item whenever they chometric properties of the final version of the AERSQ- feel ‘‘sad, disappointed, nervous, afraid, or experience E were evaluated. This validation took place both other negative or distressing feelings.’’ Item development among adolescents enrolled in junior high school (age of the original measure described in a recent article by ranging from 13 to 17, M = 14.42; Samples 4 and 5) age Zhou et al. (2020) combined a theory-driven approach and high school (age ranging from 16 to 20, M = age with feedback given from a pilot sample allowing ado- 17.10; Sample 6), in total covering an age range from 13 lescents to provide their own examples of ER strategies to 20 years. For external validity analyses, data from used. This resulted in a final 25-item version identifying Sample 2 and Sample 3 were also included wherever five factors, including rumination/negative thinking, data were available. positive reorientation, communication, distraction, and cultural activities. Based on the original AERSQ scale and new theoreti- Respondents and Recruitment cal considerations, an extended and modified version Data came from samples spread across several schools consisting of 33 items was first generated. Specific to this and municipalities in the southern part of Sweden. extended version, we generated five items aiming to cap- According to data acquired from Statistics Sweden ture a theoretical construct termed aggressive outlet, dated December 31, 2021, the average income of each reflecting the tendency to regulate emotions using forms municipality’s adult population ranged from below to of aggression. We also modified the item list of each sub- above the Swedish average (239,664 SEK/year vs. sam- scale to improve the conceptual separation of factors ple range 218,553–286,105 SEK/year). Adult education and in some cases, the interpretability of the items. This level was similar to or higher than Sweden as a whole led to a relatively large change to the overall item list for (46% vs. sample range 44%–65% having a university the new version compared with the old one (a lengthier education). The municipalities in which data were col- discussion of this procedure including motivation for lected were slightly more urban (87.6% vs. sample range each item change can be found in the supplementary 90.9%–96.6% living in urban areas). material under ‘‘Revising the scale’’). During this pro- The data collection spanned two separate projects: cess, some factors were reinterpreted and given new one focused on self-harm in adolescence and one names so that the terminology would more closely focused on self-control during adolescence. Thus, data reflect the presumed underlying function. For instance, were collected across several samples of Swedish adoles- the factor previously called cultural activities was cents aged 12 to 20. All participants responded to at renamed creative expression, and the factor previously least one version of the AERSQ-E and provided infor- called communication was renamed social support (for a mation about their gender and age. As the data were col- full comparison between versions including changes to lected within two separate larger projects, additional the item pool, see Supplemental Table S1). psychological variables were available for some but not In total, six different samples (1,104 students in total, all samples. These include measures of psychological dif- aged 12–20 years; a more detailed description of the ficulties, depression and anxiety, life satisfaction, non- samples is provided in the next section) were used to suicidal self-injury, and DE (see ‘‘Measures’’ section develop and validate the extended version of the below for more details). AERSQ in the present study. We used Sample 1 and Sample 2 (age ranging from 12 to 16, M = 14.15) to age refine and narrow down the first version of the AERSQ- Sample 1—Private Junior High School—Initial AERSQ-E E to a set of 20 items distributed across five factors (four Version. The first sample comprised 254 adolescents (130 items per factor) based on interitem correlations as well girls, 121 boys, 18 undisclosed or not identifying as as each item’s theoretical meaningfulness and distinc- either a girl or boy) in seventh to ninth grade. Ages ran- tiveness as an indicator of a particular factor. These five ged from 12 to 16 (mean [SD] age = 14.17 [0.96], miss- factors included rumination/negative thinking, positive ing = 13). Data were collected during a separate lecture reorientation, creative expression, aggressive outlet, and hour and administered alone by teachers at the school social support. However, it became evident that the fac- provided with a link to the survey by the researchers. tor aiming to assess distraction still needed further revis- Since respondents of this sample were only asked to ing. Consequently, in Sample 3, improvements were respond to the AERSQ-E and not measures directly made specifically for this subscale, and we managed to associated with psychological distress (e.g., measures of narrow down a final distraction subscale consisting of self-injury and depression), clinical psychologists were three items. This resulted in a final version consisting of available by phone or e-mail rather than in person. 23 items in total. 6 Assessment 00(0) Sample 2—Public Junior High School—Initial AERSQ-E Sample 4 and 5 form a combined sample for the valida- Version—Self-Harm Project. The second sample comprised tion of AERSQ-E in Grade 7 to 9 but differ in terms of 232 adolescents (102 girls, 126 boys, 4 undisclosed or other measured variables. not identifying as either a girl or boy) in Grades 7 to 9. Ages ranged from 13 to 16 (mean [SD] age = 14.13 Sample 6—Two Private High Schools–Final AERSQ Version— [0.88]). The survey was administered by a clinically Self-Control Project. The sixth sample comprised 340 ado- trained researcher along with a research assistant, two lescents (170 girls, 166 boys, 6 undisclosed or not identi- master’s students, and teachers at the school. The collec- fying as either a girl or boy) enrolled in two private high tion was conducted during a separate lecture hour in a schools in southern Sweden. Ages ranged from 16 to 20 school classroom. (mean [SD] age 17.10 [0.91], missing = 25). The survey Test–retest data also became available for 178 was administered by a clinically trained researcher and a respondents (mean age [SD] = 14.20 [0.86], 76 girls, 99 research assistant. The collection was conducted either boys, 3 undisclosed or not identifying as either a girl or during a separate lecture hour in a school classroom or boy) from this sample at a delay ranging from 28 to 35 using a digital classroom in Microsoft Teams. days (response rate = 76.6%). Students were provided digital access by e-mail, and the scale was administered by teachers at the school. The retest comprised only the Procedure AERSQ-E and one other scale that assesses embodi- Respondents in all samples answered the survey digitally ment, which was not part of this study. using either personal or school-distributed laptops, tablets, or cell phones. All respondents were informed of Sample 3—Public Junior High School—Second AERSQ-E the purpose and contents of the study, that personal Version —Self-Harm Project. The third sample comprised information (which was kept as part of the longitudinal 280 adolescents (151 girls, 126 boys, 3 undisclosed or aspirations of the broader project) would not be part of not identifying as either a girl or boy) in Grades 7 to 9. the analysis, and that participation was voluntary. At Ages ranged from 13 to 16 (mean [SD] age 14.17 [0.85]). least one clinically trained psychologist was available The survey was administered by a clinically trained either on site or/and by phone or e-mail to handle any researcher along with a research assistant, two master’s iatrogenic effects or other problems and concerns that students, and teachers at the school. The collection was could arise during or after the administration of the conducted during a separate lecture hour in a school questionnaire (see specific samples above for more classroom. details). All respondents provided consent prior to par- ticipation, and for all respondents under the age of 15, parental consent was obtained as well in accordance Sample 4—Public Junior High Schools—Final AERSQ-E with Swedish law. Ethical approval was provided by the Version—Self-Harm Project. The fourth sample comprised Swedish national ethics review board (registration num- 107 adolescents (51 girls, 57 boys, 2 undisclosed or not bers 2020-05885; 2021-06695-01; 2022-02093-02). identifying as either a girl or boy) in Grades 7 to 9. Ages ranged from 13 to 17 (mean (SD age 14.37 [0.96]). The survey was administered by a clinically trained Measures researcher and teacher at the schools. The collection was Across the different samples, we collected several psy- conducted during a separate lecture hour in a school chological variables used to assess the convergent and classroom. Sample 4 and 5 form a combined sample for discriminant validity of the AERSQ-E. Reliability esti- the validation of the AERSQ-E in Grade 7 to 9 but dif- mates and an overview of which variables were collected fer in terms of other measured variables. in each sample of the current study are presented in Table 1. Sample 5—Public Junior High School–Final AERSQ-E Version— Psychological difficulties were assessed using the Self-Control Project. The fifth sample comprised 145 ado- Strengths and Difficulties Questionnaire–self-report ver- lescents (74 girls, 70 boys, 1 undisclosed or not identify- sion (SDQ-s; Goodman, 1997). We used four subscales ing as either a girl or boy) in Grades 7 to 9 across two consisting of five items each: hyperactivity/inattention junior high schools. Ages ranged from 13 to 16 (mean (e.g., ‘‘I am easily distracted, I find it difficult to concen- [SD] age 14.45 [0.97]). The survey was administered by a trate’’), emotional symptoms (e.g., ‘‘I worry a lot’’), con- clinically trained researcher, a research assistant, and duct problems (e.g., ‘‘I get very angry and often lose my teachers at the schools. The collection was conducted temper’’), and peer relationship problems (e.g., ‘‘Other during a separate lecture hour in a school classroom. children or young people pick on me or bully me’’). ˚ Radman et al. 7 Table 1. Internal Consistency Coefficients (Cronbach’s a) for Variables Used for Evaluation of Convergent and Discriminant Validity of the AERSQ-E Across Samples. Sample 2 Sample 3 Sample 4 + Sample 6 Combined Scale (n = 236–237) (n = 281–283) 5(n = 250–252) (n = 350–351) (n = 1,017–1,023) SDQ-s Hyperactivity .78 .78 .82 .75 .78 Conduct .63 .55 .55 .46 .56 Emotion .77 .75 .77 .72 .74 Peer .48 .63 .60 .57 .59 RCADS-25 Anxiety .87 .87 .88 .86 .86 Depression .86 .89 .88 .87 .88 SLSS .88 .92 .90 .87 .90 DSHI-9r .90 .88 .90 — .88 RIBED-8 .88 .88 .87 — .88 Note. SDQ-s = Strength and Difficulties Questionnaire–self-report version; RCADS-25 = Revised Child Anxiety and Depression Scale–shortened version; SLSS = Students’ Life Satisfaction Scale; DSHI-9r = Deliberate Self-Harm Inventory–9-item version; RIBED = Risk Behaviour related to Eating Disorders; AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended. Measured only in sample 5 (n = 107). Items are rated on a 3-point scale (0 = not true,1= previous research where it showed good test–retest relia- somewhat true,2 = certainly true) taking the past 6 bility and a values ranging from .66 to .90 (Bja¨ rehed & months into consideration. The Swedish version was Lundh, 2008; Lundh et al., 2007; Lundh et al., 2011). empirically validated by Lundh et al. (2008) where it was The scale asks respondents to use a Likert-type scale shown to have similar psychometric properties to other from 0 (never)to6(more than five times) to indicate how language versions. often they had deliberately injured themselves (e.g., by Depression and anxiety were assessed using a short cutting, carving, or severely scratching themselves, or preventing wounds from healing) in the past 6 months. version of the Revised Children’s Anxiety and All individual items were summed into a total score Depression Scale (RCADS; Chorpita et al., 2000). The (ranging 0–54) reflecting frequency of engagement. shortened 25-item version (RCADS-25) comprises two Disordered eating was assessed using the eight-item subscales aiming to measure anxiety and depression and scale Risk Behaviour related to Eating Disorders has been shown to have comparable psychometric prop- (RiBED-8; Waaddegaard et al., 2003). The scale was erties to the long version, with alpha values ranging developed as a screening instrument aimed at assessing from .86 to .91 for the anxiety subscale and .79 to .80 for the prevalence of risk behaviors associated with eating the depression subscale (Ebesutani et al., 2012). The disorders. Respondents are asked to rate on a 4-point anxiety subscale consists of 15 items (e.g., ‘‘I worry when scale ranging from 1 (almost never/never)to4(very I think I have done poorly at something’’) and the often) how often they engage in thoughts or behaviors depression subscale consists of 10 items (e.g., ‘‘Nothing associated with food consumption (e.g., ‘‘I vomit to rid is much fun anymore’’). Items are rated on a 4-point myself of food I have eaten’’). In their original paper, Likert-type scale ranging from 0 (never)to 3 (always) Waaddegaard et al. (2003) demonstrated that the where high scores represent high degrees of anxiety or RiBED had good test–retest reliability and ability to depression, respectively. identify persons with eating pathology. In previous stud- Life satisfaction was assessed using the Student’s Life ies targeting Swedish adolescents, alpha values have ran- Satisfaction Scale (SLSS; (Huebner, 1991a, 1991b). The ged from .74 to .78 (Bjarehed & Lundh, 2008; Lundh scale consists of six items (e.g., ‘‘My life is going well’’) et al., 2008). rated on a 6-point scale ranging from 1 (strongly dis- agree)to 6 (strongly agree). The original article (Huebner, 1991a) demonstrated good test–retest reliabil- Data Analyses ity and a values of .80 and .82 across two studies. Non-suicidal self-injury was assessed with a modified Missingness and Exclusion. The missing completely at ran- version of the Deliberate Self-Harm Inventory (DSHI; dom (MCAR) assumption was rejected by Little’s Gratz, 2001) that was shortened to a nine-item version MCAR test for Sample 1, x (912) = 1,017.47, p = .008, (DSHI-9r) and adapted to Swedish adolescents in Sample 3, x (4,580) = 4,962.04, p \ .001), and Sample 8 Assessment 00(0) 5, x (2,534) = 2,712.51, p = .007. However, in all three Measure Testing. Confirmatory factors analysis (CFA) samples, the ratio of x to df was below 2, suggesting was used to evaluate the suggested EFA structure of the that the deviance from the MCAR assumption was AERSQ-E in Samples 4 to 6. The CFA was run with a maximum likelihood (ML) estimator using the lavaan minor (Ullman & Bentler, 2013). In addition, all items package in R (version 0.6.9, Rosseel, 2012). Model fit had relatively low percentage totals missing (range was evaluated using the root mean square error of across samples: 1.4–5.9%). Finally, inspection of the approximation (RMSEA) with 95 % confidence inter- multivariate missing patterns using MICE package in R vals, the Comparative Fit Index (CFI), and the standar- (van Buuren & Groothuis-Oudshoorn, 2011) showed dized root mean residual (SRMR). An acceptable model that no instances of multiple missing data occurred more fit was defined as CFI . .90, RMSEA \ .06, and than once. These patterns of missingness were therefore presumed as not meaningfully different from MCAR in SRMR \ .09, and a good model fit was defined as CFI any sample. Accordingly, prior to conducting any analy- . .95, RMSEA\ .05 and SRMR\ .08 (Hooper et al., ses, we listwise deleted 12 systematic responders because 2007; Hu & Bentler, 1999). In case of a poor model fit, potential improvements to the models were also evalu- they had zero variance in the AERSQ-E, and 42 respon- ated using modification indices and the theoretical dents missing more than 10 % in the AERSQ-E, sug- soundness of the proposed additional covariances not gesting that imputing their data could significantly bias implied by the original EFA. the factor analyses. The best-fitting AERSQ-E models were subsequently The remaining missing data values were imputed in tested for measurement invariance (MI) across age and one of two ways depending on the measure. Missing gender (S. T. Lee, 2018; Milfont & Fischer, 2010). items in the RIBED-8 and DSHI-9r, both of which tend Testing for MI requires sufficiently large test groups that to be positively skewed, were imputed with 0 for all cases are of about equal size (Chen, 2007). When using with less than 3 missing values as is common practice RMSEA as a fit index, it has been suggested that n.100 (e.g., Lundh et al., 2008; Viborg et al., 2014). All other is recommended for correct rejection of models (Putnick missing values were imputed using the expectation– & Bornstein, 2016). As the sample size was below 100 maximization (EM) algorithm from the mvdalab pack- for most specific ages, we opted to assess age by compar- age in R (Afanador et al., 2021). Items were imputed at ing adolescents attending junior high school (Samples 4– item-level with other values as predictors, apart from 5, N= 252, age range 13–17, mean age = 14.41) to five respondents who did not respond to some scales in those attending high school (Sample 6, N = 340, age full and consequently had their data for those scales range 16–20, mean age = 17.10). For model compari- imputed at the scale-level using the other scales are pre- son, we relied on CFI, RMSEA, and SRMR as has dictors. All analyses reported in the current paper were become common practice (Putnick & Bornstein, 2016) run with both the imputed and the nonimputed data, due to the sensitivity of chi-square difference tests to and since imputations did not markedly change any of sample size and deviations from normality assumptions the results, all results presented below are based on the (Chen, 2007; Sass, 2016). Here, we considered values of imputed data. DCFI \ .01, DRMSEA \ .015, and DSRMR \ .03 as indicating sufficient model similarity to conclude metric Measure Development. For Samples 1 and 2 and later invariance, and DCFI \ .01, DRMSEA \ .015, and Sample 3, through which the final version was gener- DSRMR \ .01 scalar to conclude scalar invariance ated, exploratory factor analyses (EFA) on the AERSQ- (Chen, 2007; Sass, 2016). E utilized an ordinary least squares solution with obli- Next, the implied subscales of the AERSQ-E were min, an oblique rotation method which assumes that the tested for internal consistency, test–retest reliability, and latent variables can be correlated. The optimal number divergent or discriminant validity. Internal consistency of factors was determined using parallel analysis, which was assessed using Cronbach’s a values, where we compares the scree of eigenvalues of the observed data sought an a of .7 or above. The test–retest reliability with a Monte Carlo–simulated matrix of data of the coefficients were computed for all but the distraction same size. We also considered which factors had an subscale using data from Sample 2. We calculated corre- eigenvalue .1 and the point of inflection in the scree lations between the AERSQ-E subscales and measures plot. All analyses were conducted with the psych package of NSSI, DE, internalizing/externalizing problems, emo- in R (Revelle, 2017). We initially opted to retain a maxi- tional distress, and positive functioning to evaluate con- mum of four items in each subscale with factor loadings vergent and discriminant validity. Correlations are above .4. We also aimed to identify an item pool with presented using Pearson’s r for ease of interpretability as satisfactory coverage of each identified ER strategy and nonparametric alternatives yielded comparable results thus screened for items too highly correlated. in terms of statistical significance and effect size when ˚ Radman et al. 9 the assumptions were violated. Bonferroni-corrected sig- aggressive outlet (4 items), social support (4 items), and nificance values were used to avoid spurious correlations distraction (3 items). One item from the positive reorien- and applied across all parametric tests, where we used a tation subscale (i.e., ‘‘Try in a calm manner to solve corrected alpha of .05/69 = 0.0007. what made me feel bad’’) showed low factor loading (l = .32) as well as cross-loading with aggressive outlet (l = .28) in Sample 3. However, given the content of Results the item as well as the previous factor loading from Samples 1 and 2 (l = .61), we decided to retain this Exploratory Factor Analysis item in the final positive reorientation subscale. Parallel analysis with data from Sample 1 and Sample 2 on the initial version of the AERSQ-E suggested six fac- tors to best fit the data. The same was suggested by Confirmatory Factor Analysis looking at eigenvalues and the point of inflection in the The final version consisting of 23 items was adminis- scree plot. Based on EFA, four items were retained per tered in Samples 4 and 5 (representing respondents in subscale (for a summary of the results from this factor junior high school) and in Sample 6 (representing analysis, see Table S2 of the Supplementary). However, respondents in high school). When combining all sam- poor factor loadings for many items included in the dis- ples (4–6), the six-factor model showed mixed results on traction subscale suggested it required further revision. fit indices with RMSEA and SRMR showing acceptable After a close inspection of the items, we theorized values, but CFI was found to be somewhat low (N = that the poor psychometric properties of the distraction 592, RMSEA [95% confidence interval, CI] = .064 [.059, subscale were likely caused by the specificity of the .069], SRMR = .066, CFI = .884). However, one pair of items, in that, compared with other subscales, it con- items in the aggressive outlet subscale, ‘‘I argue or fight tained much more specific activities putatively reflecting with people around me’’ and ‘‘I want to hurt others (phy- a distraction-oriented ER. Examples include ‘‘Watch sically or mentally)’’ demonstrated high residual covar- something (e.g., a movie, television shows, streaming)’’ iance not explained by the latent variable (s =.62). A or ‘‘Play games (e.g., video or computer games).’’ These rerun of the combined sample adding this covariance to activities, although presumably reflecting distractive the model demonstrated acceptable model fit across all activities, might not be endorsed within-person consis- indices (N = 592, RMSEA [95% CI] = .053 [.048, .058], tently such that a person might prefer one distractive SRMR = .061, CFI = .920, modeling these samples sep- activity over all others. Although in line with the focus arately yielded comparable results, cf. Supplemental of the AERSQ-E on assessing specific ER behaviors, we Table S3). The loadings and covariances of these models decided to reformulate the scale to focus more on the are visualized in Figure 1. The modified model signifi- cognitive processes at work, hopefully allowing the sub- cantly improved model fit (Dx = 19.06, p\ .001). scale to identify distraction more consistently and inde- pendently as an ER strategy across all items. Examples of new items include ‘‘Try to think about something Internal Consistency else’’ and ‘‘Distract myself with something to do’’ (also Table 3 presents the internal consistency values see Table 2). (Cronbach’s a) of the AERSQ for all samples. Across A second version of the AERSQ-E, including five all samples and in combination, alpha values for five out new items intended to capture distraction and 20 items of six factors tended to be in acceptable or high ranges. retained from the first EFA for the other subscales, was The exception was a distraction, which showcased ade- subsequently tested in Sample 3. We limited the second quate values for Sample 3 (a = .71) but somewhat lower revised distraction subscale to five new items because of values for Samples 4 and 5 (a = .57), Sample 6 (a = a need to minimize the burden put on respondents con- .65), and in a pooled analysis (a = .65). sidering the data collection at large. Once again, parallel analysis again suggested six factors to best fit the data (as did the eigenvalues and point of inflection using a Test–Retest Reliability scree plot), and results from the subsequent EFA is pre- sented in Table 2. For the distraction subscale, we All test–retest correlations were medium to large (n = selected three items showing satisfactory factor loading 178): r [95% CI] rumination/negative thinking = .77 and construct coverage, resulting in a final 23-item ver- [.70, .82], positive reorientation = .71 [.62, .77], creative sion of the AERSQ with six different subscales labeled expression = .73 [.66, .79], aggressive outlet .71 [.63, as rumination/negative thinking (4 items), positive reor- .77], and social support = .77 [.70, .82]. No retest data ientation (4 items), creative expression (4 items), were available for the distraction subscale, as it was 10 Assessment 00(0) Table 2. Exploratory Factor Loadings for the Final Version of the AERSQ-E (Sample 3). Sample 3 Factor Item (in English translation) 1 2 3456 R1. Think about things I have said or done (again and again) .60 –.05 .17 –.09 .00 .06 R2. Think that I am bad or worthless .69 –.25 .05 –.02 –.01 .04 R3. Worry about what others might think of me .66 –.10 .03 –.10 .02 –.02 R4. Believe that others have it much better off than me .64 .02 –.05 .16 .02 .01 P1. Try to find something positive in what has happened –.01 .56 .04 .01 .06 .16 P2. Move on and try to do things better next time –.03 .73 .00 –.13 –.02 .05 P3. Try in a calm manner to solve what made me feel bad .09 .32 .12 –.15 .28 .21 P4. Stay calm and think that it will pass –.10 .81 .01 .00 .03 –.01 C1. Create something that expresses how I feel –.19 –.04 .71 .07 .02 .11 C2. Write texts (e.g., stories, song lyrics, poems) .15 .10 .64 .00 –.04 –.11 C3. Draw or paint –.02 –.01 .64 .02 .01 .02 C4. Write down thoughts about how I feel .27 .05 .58 –.04 .02 –.06 A1. Punch or kick on things –.13 –.09 .00 .71 –.08 –.02 A2. Try to find something to break –.06 –.10 .04 .74 .02 .04 A3. Argue or fight with people around me .40 .00 .02 .49 .07 –.05 A4. Want to hurt others (physically or mentally) .30 .10 .07 .49 .00 –.18 S1. Tell someone else how I feel –.06 –.02 –.10 .01 .88 .06 S2. Seek support and comfort in others .01 .04 .03 .02 .87 –.08 S3. Ask others for advice or help .05 .05 –.02 –.07 .71 .01 S4. Seek physical contact (e.g., a hug) .05 –.06 .28 .00 .60 .06 D1. Try to think about something else .01 .12 –.06 –.01 –.01 .77 D2. Distract myself with something to do –.01 –.11 .12 –.07 .07 .62 D3. Try to forget that which makes me feel bad .01 .07 .01 –.01 –.03 .57 Avoid things that remind me of my feelings .38 .09 –.03 .15 .14 .37 Pretend like the emotions I feel do not exist .28 .03 .03 .24 –.22 .32 Note. The Exploratory Factor Analysis was based on oblimin rotation and extraction by ordinary least squares. Bold represents final factor assignment. AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended. a b Item P3 was retained with a factor loading of .61 in Samples 1 and 2. Items removed from the final version. Table 3. Cronbach’s a Values for AERSQ-E Subscales Across Samples. Sample Subscale 1 2 3 4–5 6 Combined Rumination/negative thinking .80 .84 .79 .80 .77 .80 Positive reorientation .76 .80 .78 .78 .78 .78 Creative expression .70 .72 .75 .68 .78 .73 Aggressive outlet .75 .68 .74 .77 .68 .72 Social support .82 .88 .85 .79 .84 .84 Distraction NA NA .71 .57 .65 .65 Note. AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended. Sample 1 and 2 are not included. revised in a later sample compared with when test–retest invariance between age groups (i.e., junior high school stu- reliability assessments were performed. dents were compared to high school students) and between girls and boys across all fit indices of interest (DCFI gender = .005, DRMSEA \ .001, DSRMR = .003; gender gender Measurement Invariance DCFI =.007, DRMSEA =.001, DSRMR = age age age Table 4 summarizes the results for method invariance .005). Considering scalar invariance for both age and gen- across age and gender. We found evidence of metric der, DRMSEA and DSRMR suggested scalar invariance ˚ Radman et al. 11 Figure 1 Six Factor AERSQ-E Model With 23 Indicator Items Note. N = 592, RMSEA [95% CI] = .053 [.048, .058], SRMR = .061, CFI = .920. Loadings and covariances represent combined data from Samples 4 to 6. See Table S3 of Supplementary for estimated confirmatory factor loadings, covariances and model fit for each sample separately. AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire —Extended; RMSEA = root mean square error of approximation; CI = confidence intervals; CFI = Comparative Fit Index; SRMR = standardized root mean residual. Table 4. Comparing Configural, Metric and Scalar Invariance Across Gender and Age Groups. Group Model x p value CFI RMSEA SRMR Gender 1. Configural 806.04 \.001 .909 .055 .061 2. Metric 835.06 \.001 .906 .055 .064 |D| 2-1 29.02 .034 .003 \.001 .003 3. Scalar 944.30 \.001 .883 .060 .069 |D| 3-2 109.24 \.001 .023 .005 .005 Age 1. Configural 765.23 \.001 .923 .052 .064 2. Metric 808.69 \.001 .917 .053 .068 |D| 2-1 43.46 \.001 .007 .001 .005 3. Scalar 895.94 \.001 .901 .057 .071 |D| 3-2 87.25 \.001 .016 .003 .003 Note. RMSEA = root mean square error of approximation; CFI = comparative fit index; SRMR = standardized root mean residual. but DCFI did not (DCFI = .023, DRMSEA \ found a positive association between rumination/negative gender gender .005, DSRMR = .005; DCFI = .016, DRMSEA thinking and aggressive outlet (Samples 2–6, r= .28), and gender age age = .003, DSRMR = .003). Scalar invariance across all both these were negatively associated with positive reorien- age indices (DCFI = .008–0.010, DRMSEA = .002, tation (rumination/negative thinking: Samples 2–6, r=– DSRMR = .002) could only be established by releas- .28; aggressive outlet: Samples 2–6, r=–.31). Positive reor- ing the two items significantly associated with the ientation, conversely, was as expected positively associated highest model fit improvement for gender (i.e., ‘‘Argue with both social support (Samples 2–6, r = .27) and dis- or fight with people around me’’ and ‘‘Seek physical traction (Samples 3–6, r = .37). Creative expression was contact (e.g., a hug)’’; x = 16.44–34.43) and age, positively associated with social support (Samples 2–6, r= respectively (i.e., ‘‘Think about things I have said or .24), rumination/negative thinking (Samples 2–6, r= .27), done (again and again)’’ and ‘‘Distract myself with and aggressive outlet (Samples 2–6, r= .15), whereas we something to do’’; x = 19.24–20.28). expected to find a positive association only with social sup- port and distraction. Finally, in line with expectations, Subscale Intercorrelations social support was weakly but positively associated with distraction (Samples 3-6, r= .21). All correlations pre- Table 5 presents expected and observed intercorrelations sented here were significant at p\ .0007. between all subscales of the AERSQ-E. As expected, we 12 Assessment 00(0) Table 5. Expected and Observed Intercorrelations Between the AERSQ-E Subscales (Sample 2–6, N = 872–1,104). Expected pattern Observed pattern Subscale RP C A S D R P C A S D N Rumination/negative thinking 1 1,104 Positive reorientation — –.28 1 1,104 *** Creative expression 0/- + .27 .03 1 . 1,104 *** Aggressive outlet ++ — — .28 –.31 .15 1 1,104 *** *** *** Social support — ++ + — .11 .27 .24 –.04 1 1,104 *** *** *** Distraction 0 ++ + 0 + .06 .37 .08 –.12 .21 1 872 *** * *** *** Note. Significant correlations after Bonferroni correction are shown in bold. Expected patterns reflects the rating of two independent raters +++ = strong positive correlation (r . .5), ++ = moderate positive correlation (.3\ r . .5), + = weak positive correlation (.1\ r . .3), 0 = no correlation, – = weak negative correlation (r\ –.1), –– = moderate negative correlation (–.1\ r . –.3), ––– = strong negative correlation (r . –.5). AERSQ-E = Adolescents’ Emotion Regulation Strategies Questionnaire–Extended; Columns R = Rumination/negative thinking; P = Positive reorientation; C = Creative Expression; A = Aggressive outlet; S = Social support; D = Distraction. *p\ .05. **p\ .01. ***p\ .001. Convergent and Discriminant Validity constructs including hyperactivity/inattention (Samples 2–6, r = –.34), emotional symptoms (Samples 2–6, r = Table 6 presents a summary of results including an –.29), conduct problems (Samples 2–6, r = –.26), anxi- expected and observed pattern of association with other ety (Samples 2–6, r = –.27), depression (Samples 2–6, r psychological variables. Key findings (all significant at p \ 0007 in accordance with Bonferroni correction) are –.38), NSSI (Samples 2-4, r = –.38) and DE (Samples summarized below. 2–4, r = –.33). As expected, moderate to strong positive associations We predicted social support to be weakly but nega- were found between rumination/negative and negative tively associated with most negative constructs and posi- constructs including anxiety (Samples 2–6, r = .59), tively associated with life satisfaction. In line with these depression (Samples 2–6, r = .51), NSSI (Samples 2–4, r predictions, we found a weak but negative correlation = .33), DE (Samples 2–4, r = .48), and emotional between psychological difficulties and social support symptoms (Samples 2–6, r = .53). Conversely, rumina- including hyperactivity/inattention (Samples 2–6, r =– tion/negative thinking had a strong negative correlation .13) and conduct problems (Samples 2–6, r = –.14) and with life satisfaction (Samples 2–6, r = –.51). We also a weak positive correlation to life satisfaction (Samples expected a moderate positive association between rumi- 2–6, r = –.15). Although some other weak correlations nation/negative thinking and conduct problems, which were also found, these did not retain significance after was not supported by the data. Bonferroni correction. Similar patterns were observed for an aggressive out- For creative expression, we only made predictions for let, which was positively correlated with anxiety some of the associations, and when we did, we expected (Samples 2–6, r = .24), depression (Samples 2–6, r = weak or no associations. However, we found weak but .31), NSSI (Samples 2–4, r = .37), DE (Samples 2–4, positive associations to most negative constructs includ- r = .25), and emotional symptoms (Samples 2–6, r = ing peer problems (Samples 2–6, r = .14), emotional .17) and negatively correlated with life satisfaction symptoms (Samples 2–6, r = .23), anxiety (Samples 2–6, (Samples 2–6, r = –.29). In addition, aggressive outlet r = .28), depression (Samples 2–6, r = .20), NSSI had a moderate to strong relationship with hyperactiv- (Samples 2–4, r = .18), and DE (Samples 2–4, r = .19) ity/inattention (Samples 2–6, r = .36) and conduct and weak but negative association to life satisfaction problems (Samples 2–6, r = .41). However, we did not (Samples 2–6, r = –.18). find the expected positive association between aggressive Finally, given the ambiguous functionality of distrac- outlet and peer problems (Samples 2–6, r = .10). tion as an ER strategy, we expected its subscale to have Positive reorientation, in line with expectations, was few and weak associations with other measured con- positively correlated to life satisfaction (Samples 2–6, structs. This was confirmed in that all correlations were r = .42) and negatively correlated to most negative ˚ Radman et al. 13 Table 6. Expected and Observed Intercorrelations Between the AERSQ-E and Other Studied Variables in Combined Sample (Sample 2–6, N = 619–1,104). Expected pattern Observed pattern Variable R P C A S D R P C A S D Psychological difficulties (SDQ) Hyperactivity/inattention + – 0/– + 0/– 0 .12 –.34 .02 .36 –.13 –.08 1,104 *** *** *** *** Peer relationship problems ++ – 0/– + – 0 .17 –.12 .14 .10 –.04 –.11 1,104 *** *** *** *** ** Emotional symptoms +++ 0/– + + – 0/+ .53 –.29 .23 .17 .12 –.01 1,104 *** *** *** *** *** Conduct problems ++ – 0 ++ – 0 .06 –.26 .04 .41 –.14 –.13 1,104 * *** *** *** *** Life satisfaction –+ 0/+ –+ 0 –.51 .42 –.18 –.29 .15 .12 1,104 (SLSS) *** *** *** *** *** *** Anxiety and depression (RCADS-25) Anxiety ++ – NA + – 0/+ .59 –.27 .28 .24 .11 .00 1,104 *** *** *** *** *** Depression +++ – NA + – 0/+ .51 –.38 .20 .31 –.04 .05 1,104 *** *** *** *** Self-injury (DSHI) ++ – NA ++ 0/– 0 .33 –.38 .18 .37 –.08 –.15 619 *** *** *** *** * Disordered eating (RIBED-8) ++ – NA NA 0/– – .48 –.33 .19 .25 –.01 –.04 619 *** *** *** *** Note. Significant correlations after the Bonferroni correction are shown in bold. Expected patterns reflects the rating of two independent raters (+++ = strong positive correlation (r . .5), ++ = moderate positive correlation (.3\ r . .5), + = weak positive correlation (.1\ r . .3), 0 = no correlation, – = weak negative correlation (r\ –.1), –– = moderate negative correlation (–.1\ r . –.3), ––– = strong negative correlation (r . –.5), NA = no predictions made). SDQ = Strengths and Difficulties Questionnaire –self-report version; SLSS = Student’s Life Satisfaction Scale; RCADS = RCADS; DSHI = Deliberate Self-Harm Inventory; RIBED = Risk Behaviour related to Eating Disorders; Columns R = Rumination/negative thinking; P = Positive reorientation; C = Creative Expression; A = Aggressive outlet; S = Social support; D = Distraction. p\ .05. **p\ .01. ***p\ .001. weak, and none retained significance after the AERSQ-E. However, the CFI did not reach adequate Bonferroni correction. values unless the residual of two aggressive outlet- related items (i.e., ‘‘Punch or kick things’’ and ‘‘Try to find something to break’’) was covaried as suggested by Discussion the modification indices. Post hoc modifications of the factor structure of a measurement should only be pur- In this study, we presented the psychometric properties sued when empirically or conceptually justified of the AERSQ-E, a modified and extended version of (MacCallum et al., 1992). In this case, the two items for the original AERSQ developed by Zhou and colleagues which residuals were covaried reflected tendencies to (2020). Across six community samples of Swedish youth direct aggressive behaviors toward inanimate objects, (aged 12–20), we generated and narrowed down a final while the other two items of the aggressive outlet sub- 23 item version of the AERSQ-E and evaluated its inter- scale reflected aggression aimed at other people. We nal structure, reliability, and validity as a measure asses- deemed this modification to be justified although it was sing the use of different ER strategies in adolescence. applied post hoc, where we reasoned that the conceptual The factor analyses generally supported a six-factor delineation this modification reflected between aggres- structure for the AERSQ-E including rumination/nega- sion toward inanimate objects or toward other people tive thinking (4 items), positive reorientation (4 items), (i.e., socially) does not take away from the use of the creative expression (4 items), aggressive outlet (4 items), subscale as a measure of ER through aggression at a social support (4 items), and distraction (3 items). These more general level. factors showed low to moderate correlations to each Assessing measurement invariance, we found that other. Using confirmatory factor analysis, both SRMR metric invariance of the modified factor structure of the and RMSEA showed adequate model fit for the AERSQ-E was supported across girls and boys and in 14 Assessment 00(0) comparing junior high school students and standard lower-than-usual a value could reflect sufficient con- high school students. However, while the DRMSEA and struct coverage (Taber, 2018) rather than problems of DSRMR gave support to scalar invariance across these multidimensionality. gender and age groups, the DCFI did not. We note that With respect to external validity, our findings gener- there are currently no clear conventions on how to inter- ally aligned with our expectations. We found that rumi- pret inconsistent results across different fit indices, and nation/negative thinking had positive associations with there is large variability in cut-off levels as well as what anxiety, depression, DE, and NSSI, aligning with data fit indices should be considered most important in evalu- found using the previous version of the AERSQ (Zhou ating measurement invariance (see Putnick & Bornstein, et al., 2020). Rumination, one of two cognitive strategies 2016, for a discussion on conventions of measurement assessed by the AERSQ-E and one of the most studied invariance testing). Nevertheless, scalar invariance is ER strategies, has been extensively linked to several important because it ensures that statistical differences mental health issues, with many studies specifically tar- in group means reflect actual differences in ER strategy geting adolescents (e.g., Calvete et al., 2015; Olatunji preferences and not unintended, biasing properties of et al., 2013; Rood et al., 2009; Royuela-Colomer et al., the scale (S. T. Lee, 2018). We managed to establish par- 2021). Accordingly, we found that rumination/negative tial scalar invariance by releasing some items but have thinking had the strongest associations with both posi- no strong theoretical reasons behind these specific modi- tive and negative aspects of mental health and psycholo- fications. Future work could examine more closely why gical functioning. This aligns with a meta-analytic specifically these items are interpreted differently by girls review by Aldao et al. (2010) who, comparing different and boys and by different levels of high school students. ER strategies, suggested that rumination had the stron- Some potentially important factors are the role played gest effect size in predicting anxiety, depression, eating, by social reinforcement in emotion and ER (e.g., Barrett and substance-related disorders. Moreover, we found & Bliss-Moreau, 2009; Nolen-Hoeksema, 2012) as well that positive reorientation, the second cognitive strategy as levels of cognitive maturation (e.g., Ahmed et al., assessed by the AERSQ-E, was associated with various 2015; Casey et al., 2008). aspects of positive functioning, showing positive asso- Regarding internal consistency, we found that all ciations with life satisfaction and negative associations subscales except distraction had acceptable internal con- with internalizing and externalizing problems. This cor- sistency (i.e., a . .70) and good test–retest reliability roborates the common understanding of strategies (i.e., r . .70). Alpha values for the distraction subscale involving positive reappraisal as adaptive (e.g., Cracco varied between 0.57 and 0.71 across different samples. et al., 2017; Schafer et al., 2016) and having wide- In a recent review on the use of Cronbach’s a in research ranging benefits to mental health (e.g., Aldao et al., on instrument development, Taber (2018) provided illus- 2010; Nowlan et al., 2015). trative examples from the science education literature Social support, the interpersonal dimension assessed showing a wide range of alpha values being treated as with the AERSQ-E, was weakly but negatively associ- acceptable or satisfactory (e.g., as low as a = .45). The ated with internalizing and externalizing problems and article raised concerns with the arbitrary value of .70 as positively associated with life satisfaction. This corrobo- a sufficient measure of acceptable internal consistency, rates findings suggesting that the availability of interper- citing several influential statisticians. For instance, sonal resources can contribute to positive functioning although Cronbach (1951) himself suggested that a high (Dixon-Gordon et al., 2015). However, we urge that value of alpha was ‘‘desirable,’’ he also emphasized the future work interprets this scale within context, as who importance of instrument interpretability which, accord- is providing the support could have important implica- ing to him, was often possible without having high val- tions for the effectiveness and availability of utilizing ues of alpha. Similar conclusions have been made by this strategy to regulate emotion. Contextual interpreta- Schmitt (1996) claiming there is no general level (such as tion is also necessary for the distraction subscale, as it .70) at which a becomes acceptable and that instruments demonstrated only negligible associations (r \ .15) with with quite low alpha values can prove useful. In relation the other variables. Other studies have similarly sug- to coping and ER specifically, early researchers postu- gested that distraction is only weakly associated with lated that a low alpha value is sometimes expected if the concurrent and future levels of depression (Rood et al., use of one type of coping (ER) strategy obviates the use 2009). These findings reinforce the view that distraction of another (Billings & Moos, 1981). Given that the dis- is neither adaptive nor maladaptive outside its context; traction subscale comprised only three items, meaning rather, it can be a predictor of positive functioning when that each item intercorrelation has a considerable effect combined with an attitude of acceptance and a predictor on the a value (the average inter-item correlation for the of negative functioning when combined with an attitude distraction subscale was r = .36), we suggest that a of avoidance (Wolgast & Lundh, 2017). ˚ Radman et al. 15 Finally, the two expressive ER strategies (i.e., creative adolescence. It includes both ER strategies more focused expression and aggressive outlet) were in this study both on changes in cognition (i.e., rumination/negative think- positively related to internalizing and externalizing ing or positive reorientation) and changes in behavior problems and negatively related to life satisfaction. This (i.e., creative expression, aggressive outlet, or distrac- is unsurprising for aggressive outlet, as aggression has tion). The instrument also covers the interpersonal previously been linked to NSSI and deliberate self-harm dimension of ER (i.e., social support) contrasting with (Boxer, 2010; Brunner et al., 2007; Daukantaite _ et al., intrapersonal ER strategies. Moreover, it also captures 2019; Fliege et al., 2009; Sourander et al., 2006; Tang ER strategies commonly viewed as maladaptive (rumi- et al., 2013) and is largely considered a maladaptive ER nation/negative thinking, aggression outlet) and adap- strategy (e.g., Cracco et al., 2017; Grob & Smolenski, tive (e.g., positive reorientation or social support). 2005). Finding consistent, albeit weak, positive links to Finally, one key feature of the AERSQ is the behavioral mental health problems for creative expression was ER strategies focusing on expressions of emotions, more surprising. In the original AERSQ paper (Zhou including the subscales aggressive outlet and creative et al., 2020), the ‘‘cultural activities’’ subscale (compara- expression, which is a less frequently considered dimen- ble to the creative expression subscale of the current ver- sion of ER across measurements, with expressive sup- sion) produced mixed results, showing positive pression, i.e., the inhibition of emotional expression, associations only to NSSI and emotional symptoms, being much more common (e.g., Gross & John, 2003; and only at one out of two measured time-points. Hofmann & Kashdan, 2010; Zeman et al., 2001). Thus, Conversely, research regarding the influence of perform- the AERSQ-E holds promise to be applicable in a wide ing art-based activities tend to highlight the beneficial range of research projects emphasizing different distinc- effect with regard to mental health (e.g., Geipel et al., tions in ER. 2018; Saarikallio, 2010; van Lith et al., 2013; Zhao et al., Beyond the limitations discussed previously, there are 2016), whereas our findings suggest the opposite. It a few others that should be discussed. First, given the should be noted, however, that the current study is fact that this study is cross-sectional and correlational, cross-sectional, meaning the associations found here do we cannot ascertain whether making use of certain ER not necessarily translate to the longitudinal effects of strategies poses a risk of developing mental health issues using art-based activities as ER strategies. It is conceiva- or if it is the other way around, whereby a preexisting ble that there would be prospective beneficial effects of level of distress causes the favoring of certain ER strate- the creative expression subscale when assessed longitud- gies over others. For instance, the cross-sectional nature inally except that art-based creative activities are com- of this study might be the cause of the weak but overall mon among a portion of those with a tendency toward positive association found between creative expression issues of mental health (Cropley, 1990). Future longitu- and mental health issues, contrasting previous findings dinal studies could investigate this further. showing mental health benefits for these types of activi- An important strength of the present study is that the ties (e.g., Geipel et al., 2018; Zhao et al., 2016). sample encompassed 1,104 adolescents in total sourced Longitudinal assessments are needed to further investi- from schools located in Swedish municipalities that were gate the validity of each subscale in this regard. comparable to the Swedish average. In addition, we had Second, we did not manage to collect test–retest data roughly an equal number of both adolescent girls and for the distraction subscale given its late revision in the boys covering a wide range of ages. Together, this lends development process, meaning we could not evidence strong support to the representativity of the sample and the longitudinal reliability of this subscale. Given this generalizability of the findings, at least covering a and its relatively low internal consistency values, future Swedish setting and across countries with comparable studies are warranted to seek improvements to this sub- demographics. Future studies are needed to evaluate the scale in particular. structure and psychometric properties of the AERSQ-E Third, although our goal was to avoid items requiring cross-culturally. Future studies should also examine the complex metacognitive evaluations, the degree to which instrument among those not identifying as either men/ we managed this is uncertain. For instance, we reformu- boys or women/girls. Finally, we did not recruit from lated the distraction subscale to be more general, thereby exclusively clinical populations of adolescents who increasing its level of abstraction. This means the ques- might show unique difficulties regarding ER, implying tions (e.g., ‘‘Try to think about something else’’) require that its applicability in a clinical setting has not been an ability to not only draw connections between felt established, something we suggest for future studies. emotions and behavior, but also to some extent under- Another strength is that the AERSQ-E encompasses stand the cognitive processes that drove that behavior. a broad range of important ER dimensions, making it a Importantly, these items as well as the items included in holistic yet resource-efficient measurement of ER in the other subscale do not require insight about the 16 Assessment 00(0) consequences of these cognitive and behavioral pro- numbers SOL: 2020-05885; VIS: 2021-06695-01; 2022-02093- 02). cesses, which would have added metacognitive self- evaluative requirements we deem particularly proble- matic for assessment in adolescence. ORCID iDs To summarize, the AERSQ-E has several merits as it Gustaf Ra˚ dman https://orcid.org/0000-0002-7042-0196 addresses some limitations posed by the current instru- Benjamin Clare´ us https://orcid.org/0000-0003-0972-6660 ments used to measure adolescent ER, and it showed Daiva Daukantaite https://orcid.org/0000-0002-1994-041X some improvement in psychometric properties com- pared with its predecessor, the AERSQ. However, fur- Data Availability Statement ther improvements could be made, and specifically, The data that support the findings of this study are available regarding the distraction subscale the validity and relia- on request from the corresponding author. The data are not bility could be further investigated. In addition, the publicly available due to privacy or ethical restrictions. instrument should be validated using longitudinal data and for use with clinical samples. Nevertheless, we believe the AERSQ-E has the potential to contribute to Supplemental Material our knowledge of ER across adolescence. Supplemental material for this article is available online. Acknowledgment References The authors wish to extend their sincere gratitude to Lars- Adrian, M., Zeman, J., & Veits, G. (2011). Methodological Gunnar Lundh and Margit Wa˚ ngby Lundh for their invalu- implications of the affect revolution: A 35-year review of able contribution in developing the AERSQ, and for their emotion regulation assessment in children. Journal of assistance in creating the extended version of the AERSQ. Experimental Child Psychology, 110(2), 171–197. https:// doi.org/10.1016/J.JECP.2011.03.009 Afanador, N. L., Tran, T., Blanchet, L., & Baumgartner, R. Author Contributions (2021). Mvdalab: Multivariate data analysis laboratory. DD contributed to the design of the larger project. GR had https://CRAN.R-project.org/package=mvdalab primary responsibility for questionnaire revisions and exten- Ahmed, S. P., Bittencourt-Hewitt, A., & Sebastian, C. L. sions and collected most of the data. GR and BC both con- (2015). Neurocognitive bases of emotion regulation devel- ducted the statistical analyses and wrote the first full draft of opment in adolescence. Developmental Cognitive Neu- the manuscript. All authors worked on several edits of the roscience, 15, 11–25. https://doi.org/10.1016/J.DCN.2015. paper. All authors contributed to and have approved the final 07.006 manuscript. Aldao, A. (2013). The future of emotion regulation research: Capturing context. Perspectives on Psychological Science, 8(2), 155–172. https://doi.org/10.1177/1745691612459518 Declaration of Conflicting Interests Aldao, A., Gee, D. G., de Los Reyes, A., & Seager, I. (2016). The author(s) declared no potential conflicts of interest with Emotion regulation as a transdiagnostic factor in the devel- respect to the research, authorship, and/or publication of this opment of internalizing and externalizing psychopathology: article. Current and future directions. Development and Psycho- pathology, 28(4pt1), 927–946. https://doi.org/10.1017/S095 Funding Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emo- The author(s) disclosed receipt of the following financial sup- tion-regulation strategies across psychopathology: A meta- port for the research, authorship, and/or publication of this analytic review. Clinical Psychology Review, 30(2), 217–237. article: This study was funded by a grant from the Swedish https://doi.org/10.1016/J.CPR.2009.11.004 Research Council for Health, Working Life and Welfare Andover, M. S., & Morris, B. W. (2014). Expanding and clari- (2016-00248); by the Thora Ohlsson Foundation; and by the fying the role of emotion regulation in nonsuicidal self- Lundh Research Foundation. injury. The Canadian Journal of Psychiatry, 59(11), 569–575. https://doi.org/10.1177/070674371405901102 Barrett, L. F., & Bliss-Moreau, E. (2009). She’s emotional. Ethical Approval He’s having a bad day: Attributional explanations for emo- The development of this new version was done as part of two tion stereotypes. Emotion, 9(5), 649–658. https://doi.org/10. project, one on self-harm, emotion regulation and interperso- 1037/A0016821 nal relations in youth (self-harm project; SOL), and one on Beauchaine, T. P. (2015). Future directions in emotion dysre- executive functions and mental health among young adoles- gulation and youth psychopathology. Journal of Clinical cents (self-control project; VIS). Ethical approval was pro- Child & Adolescent Psychology, 44(5), 875–896. https://doi. vided by the Swedish national ethics review board (registration org/10.1080/15374416.2015.1038827 ˚ Radman et al. 17 Bender, P. K., Reinholdt-Dunne, M. L., Esbjørn, B. H., & Cracco, E., van Durme, K., & Braet, C. (2015). Validation of Pons, F. (2012). Emotion dysregulation and anxiety in chil- the FEEL-KJ: An instrument to measure emotion regula- dren and adolescents: Gender differences. Personality and tion strategies in children and adolescents. PLOS ONE, Individual Differences, 53(3), 284–288. https://doi.org/10. 10(9), Article e0137080. https://doi.org/10.1371/journal. 1016/J.PAID.2012.03.027 pone.0137080 Billings, A. G., & Moos, R. H. (1981). The role of coping Cronbach, L. J. (1951). Coefficient alpha and the internal responses and social resources in attenuating the stress of structure of tests. Psychometrika, 16(3), 297–334. https:// life events. Journal of Behavioral Medicine, 4(2), 139–157. doi.org/10.1007/BF02310555 https://doi.org/10.1007/BF00844267 Cropley, A. J. (1990). Creativity and mental health in everyday Bjarehed, J., & Lundh, L. (2008). Deliberate self-harm in 14- life. Creativity Research Journal, 3(3), 167–178. https://doi. year-old adolescents: How frequent is it, and how is it asso- org/10.1080/10400419009534351 Daukantaite, _ D., Lundh, L. G., & Wa˚ ngby-Lundh, M. (2019). ciated with psychopathology, relationship variables, and styles of emotional regulation? Cognitive Behaviour Ther- Association of direct and indirect aggression and victimiza- apy, 37(1), 26–37. https://doi.org/10.1080/1650607070177 tion with self-harm in young adolescents: A person-oriented 8951 approach. Development and Psychopathology, 31(2), Bonanno, G. A., & Burton, C. L. (2013). Regulatory flexibil- 727–739. https://doi.org/10.1017/S0954579418000433 ity: An individual differences perspective on coping and Dingemans, A., Danner, U., & Parks, M. (2017). Emotion reg- emotion regulation. Perspectives on Psychological Science, ulation in Binge eating disorder: A review. Nutrients 2017, 9(11), 1274. https://doi.org/10.3390/NU9111274 8(6), 591–612. https://doi.org/10.1177/1745691613504116 Dixon-Gordon, K. L., Bernecker, S. L., & Christensen, K. Boxer, P. (2010). Covariation of self- and other-directed aggression among inpatient youth: Continuity in the transi- (2015). Recent innovations in the field of interpersonal tion to treatment and shared risk factors. Aggressive Beha- emotion regulation. Current Opinion in Psychology, 3, vior, 36(3), 205–217. https://doi.org/10.1002/AB.20343 36–42. https://doi.org/10.1016/J.COPSYC.2015.02.001 Brockmeyer, T., Skunde, M., Wu, M., Bresslein, E., Rudofsky, Ebesutani, C., Reise, S. P., Chorpita, B. F., Ale, C., Regan, J., G., Herzog, W., & Friederich, H. C. (2014). Difficulties in Young, J., Charmaine, H. M. M., & Weisz, J. R. (2012). emotion regulation across the spectrum of eating disorders. The Revised Child Anxiety and Depression Scale-Short Comprehensive Psychiatry, 55(3), 565–571. https://doi.org/ Version: Scale reduction via exploratory bifactor modeling of the broad anxiety factor. Psychological Assessment, 10.1016/J.COMPPSYCH.2013.12.001 Brunner, R., Parzer, P., Haffner, J., Steen, R., Roos, J., Klett, 24(4), 833–845. https://doi.org/10.1037/A0027283 M., & Resch, F. (2007). Prevalence and psychological corre- Eisenberg, N., Cumberland, A., & Spinrad, T. L. (1998). Par- lates of occasional and repetitive deliberate self-harm in ado- ental socialization of emotion. Psychological Inquiry, 9(4), lescents. Archives of Pediatrics & Adolescent Medicine, 161(7), 241–273. https://doi.org/10.1207/s15327965pli0904_1 641–649. https://doi.org/10.1001/ARCHPEDI.161.7.641 Fliege, H., Lee, J. R., Grimm, A., & Klapp, B. F. (2009). Risk Calvete, E., Orue, I., & Hankin, B. L. (2015). Cross-lagged factors and correlates of deliberate self-harm behavior: A associations among ruminative response style, stressors, systematic review. Journal of Psychosomatic Research, and depressive symptoms in adolescents. Journal of Social 66(6), 477–493. https://doi.org/10.1016/J.JPSYCHORES. 2008.10.013 and Clinical Psychology, 34(3), 203–220. https://doi.org/10. Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative 1521/jscp.2015.34.3.203 Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent life events, cognitive emotion regulation and emotional brain. Developmental Review, 28(1), 62–77. https://doi.org/ problems. Personality and Individual Differences, 30(8), 10.1016/J.DR.2007.08.003 1311–1327. https://doi.org/10.1016/S0191-8869(00)00113-6 Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack Geipel, J., Koenig, J., Hillecke, T. K., Resch, F., & Kaess, M. of measurement invariance. Structural Equation Modeling: (2018). Music-based interventions to reduce internalizing A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/ symptoms in children and adolescents: A meta-analysis. 10.1080/10705510701301834 Journal of Affective Disorders, 225, 647–656. https://doi. Chorpita, B. F., Yim, L., Moffitt, C., Umemoto, L. A., & org/10.1016/J.JAD.2017.08.035 Goodman, R. (1997). The strengths and difficulties question- Francis, S. E. (2000). Assessment of symptoms of DSM-IV anxiety and depression in children: A revised child anxiety naire: A research note. Journal of Child Psychology and Psy- and depression scale. Behaviour Research and Therapy, 38(8), chiatry, 38(5), 581–586. https://doi.org/10.1111/j.1469- 835–855. https://doi.org/10.1016/S0005-7967(99)00130-8 7610.1997.tb01545.x Cole, P. M., Hall, S. E., & Hajal, N. (2017). Emotion dysregu- Gratz, K. L. (2001). Measurement of deliberate self-harm: Pre- lation as a risk factor for psychopathology. Child and Ado- liminary data on the deliberate self-harm inventory. Journal lescent Psychopathology, 2, 341–373. of Psychopathology and Behavioral Assessment, 23(4), Cracco, E., Goossens, L., & Braet, C. (2017). Emotion regula- 253–263. https://doi.org/10.1023/A:1012779403943 Gratz, K. L., & Roemer, L. (2004). Multidimensional assess- tion across childhood and adolescence: Evidence for a maladaptive shift in adolescence. European Child & Adoles- ment of emotion regulation and dysregulation: Develop- cent Psychiatry, 26(8), 909–921. https://doi.org/10.1007/ ment, factor structure, and initial validation of the s00787-017-0952-8 difficulties in emotion regulation scale. Journal of 18 Assessment 00(0) Psychopathology and Behavioral Assessment, 26(1), 41–54. Goldsmith (Eds.), Handbook of affective sciences (pp. 347– https://doi.org/10.1023/B:JOBA.0000007455.08539.94 374). Oxford University Press. Grob, A., & Smolenski, C. (2005). Fragebogen zur Erhebung Larson, R. W., Moneta, G., Richards, M. H., & Wilson, S. der Emotionsregulation bei Kindern und Jugendlichen (2002). Continuity, stability, and change in daily emotional (FEEL-KJ) [Questionnaire for surveying emotion regula- experience across adolescence. Child Development, 73(4), tion in children and adolescents (FEEL-KJ)]. Huber 1151–1165. https://doi.org/10.1111/1467-8624.00464 Verlag. Lee, F. S., Heimer, H., Giedd, J. N., Lein, E. S., Sestan, N., Gross, J. J., & John, O. P. (2003). Individual differences in two Weinberger, D. R., & Casey, B. J. (2014). Adolescent men- emotion regulation processes: Implications for affect, rela- tal health-opportunity and obligation: Emerging neu- tionships, and well-being. Journal of Personality and Social roscience offers hope for treatments. Science, 346(6209), Psychology, 85(2), 348–362. https://doi.org/10.1037/0022- 547–549. https://doi.org/10.1126/science.1260497 3514.85.2.348 Lee, S. T. (2018). Testing for measurement invariance: Does Gullone, E., & Taffe, J. (2012). The emotion regulation ques- your measure mean the same thing for different partici- tionnaire for children and adolescents (ERQ-CA): A psy- pants? APS Observer, 31(8). https://www.psychologi- chometric evaluation. Psychological Assessment, 24(2), calscience.org/observer/testing-for-measurement-invariance Lundh, L., Karim, J., Quilisch, E., & Lundh, L.-G. (2007). 409–417. https://doi.org/10.1037/A0025777 Hofmann, S. G., Carpenter, J. K., & Curtiss, J. (2016). Inter- Deliberate self-harm in 15-year-old adolescents: A pilot personal Emotion Regulation Questionnaire (IERQ): Scale study with a modified version of the Deliberate Self-Harm development and psychometric characteristics. Cognitive Inventory. Scandinavian Journal of Psychology, 48, 33–41. Therapy and Research, 40(3), 341–356. https://doi.org/10. https://doi.org/10.1111/j.1467-9450.2006.00567.x 1007/s10608-016-9756-2 Lundh, L. G., Wa˚ ngby-Lundh, M., & Bja¨ rehed, J. (2008). Self- Hofmann, S. G., & Kashdan, T. B. (2010). The affective style reported emotional and behavioral problems in Swedish 14 questionnaire: Development and psychometric properties. to 15-year-old adolescents: A study with the self-report ver- Journal of Psychopathology and Behavioral Assessment, sion of the Strengths and Difficulties Questionnaire. Scan- 32(2), 255–263. https://doi.org/10.1007/s10862-009-9142-4 dinavian Journal of Psychology, 49(6), 523–532. https://doi. Hofmann, S. G., Sawyer, A. T., Fang, A., & Asnaani, A. org/10.1111/j.1467-9450.2008.00668.x (2012). Emotion dysregulation model of mood and anxiety Lundh, L. G., Wa˚ ngby-Lundh, M., & Bjarehed, J. (2011). disorder. Depression and Anxiety, 29, 409–416. https://doi. Deliberate self-harm and psychological problems in young org/10.1002/da.21888 adolescents: Evidence of a bidirectional relationship in girls. Hooper, D., Coughlan, J., & Mullen, M. (2007). Structural Scandinavian Journal of Psychology, 52(5), 476–483. https:// equation modeling: Guidelines for determining model fit. doi.org/10.1111/J.1467-9450.2011.00894.X The Electronic Journal of Business Research Methods, 6(1), MacCallum, R. C., Roznowski, M., & Necowitz, L. B. (1992). 53–60. Model modifications in covariance structure analysis: The Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes problem of capitalization on chance. Psychological Bulletin, in covariance structure analysis: Conventional criteria ver- 111(3), 490–504. https://doi.org/10.1037/0033-2909.111.3. sus new alternatives. Structural Equation Modeling: A Mul- 490 tidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/ Marroquı´n, B. (2011). Interpersonal emotion regulation as a 10705519909540118 mechanism of social support in depression. Clinical Psy- Huebner, E. S. (1991a). Initial development of the student’s life chology Review, 31(8), 1276–1290. https://doi.org/10.1016/j. satisfaction scale. School Psychology International, 12(3), cpr.2011.09.005 231–240. https://doi.org/10.1177/0143034391123010 McLaughlin, K. A., Hatzenbuehler, M. L., Mennin, D. S., & Huebner, E. S. (1991b). Further validation of the students’ life Nolen-Hoeksema, S. (2011). Emotion dysregulation and satisfaction scale: The independence of satisfaction and adolescent psychopathology: A prospective study. Beha- affect ratings. Journal of Psychoeducational Assessment, viour Research and Therapy, 49(9), 544–554. https://doi.org/ 9(4), 363–368. https://doi.org/10.1177/073428299100900408 10.1016/J.BRAT.2011.06.003 Johnson, D. P., & Whisman, M. A. (2013). Gender differences McRae, K., & Gross, J. J. (2020). Emotion regulation. Emo- in rumination: A meta-analysis. Personality and Individual tion, 20(1), 1–9. https://doi.org/10.1037/EMO0000703 Differences, 55(4), 367–374. https://doi.org/10.1016/J. Milfont, T. L., & Fischer, R. (2010). Testing measurement PAID.2013.03.019 invariance across groups: Applications in cross-cultural Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, research. International Journal of Psychological Research, K. R., & Walters, E. E. (2005). Lifetime prevalence and 3(1), 111–130. https://doi.org/10.21500/20112084.857 age-of-onset distributions of DSM-IV disorders in the Nolen-Hoeksema, S. (2012). Emotion regulation and psycho- national comorbidity survey replication. Archives of Gen- pathology: The role of gender. Annual Review of Clinical eral Psychiatry, 62(6), 593–602. https://doi.org/10.1001/ Psychology, 8(1), 161–187. https://doi.org/10.1146/ ARCHPSYC.62.6.593 annurev-clinpsy-032511-143109 Kopp, C. B., & Neufeld, S. J. (2003). Emotional development Nolen-Hoeksema, S., & Aldao, A. (2011). Gender and age dif- during infancy. In R. J. Davidson, K. R. Scherer & H. H. ferences in emotion regulation strategies and their ˚ Radman et al. 19 relationship to depressive symptoms. Personality and Indi- review. Journal of Youth and Adolescence, 46(2), 61–76. vidual Differences, 51(6), 704–708. https://doi.org/10.1016/ https://doi.org/10.1007/S10964-016-0585-0 Schmitt, N. (1996). Uses and abuses of coefficient alpha. Psy- J.PAID .2011.06.012 Nowlan, J. S., Wuthrich, V. M., & Rapee, R. M. (2015). Posi- chological Assessment, 8(4), 350–353. https://doi.org/10. tive reappraisal in older adults: A systematic literature 1037/1040-3590.8.4.350 review. Aging & Mental Health, 19(6), 475–484. https://doi. Sourander, A., Aromaa, M., Pihlakoski, L., Haavisto, A., org/10.1080/13607863.2014.954528 Rautava, P., Helenius, H., & Sillanpa¨ a¨ , M. (2006). Early Nozaki, Y., & Mikolajczak, M. (2020). Extrinsic emotion regu- predictors of deliberate self-harm among adolescents. A lation. Emotion, 20(1), 10–15. https://doi.org/10.1037/ prospective follow-up study from age 3 to age 15. Journal of emo0000636 Affective Disorders, 93(1–3), 87–96. https://doi.org/10.1016/ Olatunji, B. O., Naragon-Gainey, K., & Wolitzky-Taylor, K. J.JAD.2006.02.015 B. (2013). Specificity of rumination in anxiety and depres- Steinberg, L. (2005). Cognitive and affective development in sion: A multimodal meta-analysis. Clinical Psychology: Sci- adolescence. Trends in Cognitive Sciences, 9(2), 69–74. ence and Practice, 20(3), 225–257. https://doi.org/10.1037/ https://doi.org/10.1016/J.TICS.2004.12.005 H0101719 Sullivan, T. N., Helms, S. W., Kliewer, W., & Goodman, K. L. Paus, T., Keshavan, M., & Giedd, J. N. (2008). Why do many (2010). Associations between sadness and anger regulation psychiatric disorders emerge during adolescence? Nature coping, emotional expression, and physical and relational Reviews Neuroscience, 9(12), 947–957. https://doi.org/10. aggression among urban adolescents. Social Development, 1038/nrn2513 19(1), 30–51. https://doi.org/10.1111/J.1467-9507.2008.00531.X Phillips, K. F. V., & Power, M. J. (2007). A new self-report Swannell, S. v., Martin, G. E., Page, A., Hasking, P.St, & measure of emotion regulation in adolescents: The Regula- John, N. J. (2014). Prevalence of nonsuicidal self-injury in tion of Emotions Questionnaire. Clinical Psychology & Psy- nonclinical samples: Systematic review, meta-analysis and chotherapy, 14(2), 145–156. https://doi.org/10.1002/CPP.523 meta-regression. Suicide and Life-Threatening Behavior, Putnick, D. L., & Bornstein, M. H. (2016). Measurement 44(3), 273–303. https://doi.org/10.1111/SLTB.12070 invariance conventions and reporting: The state of the art Taber, K. S. (2018). The use of Cronbach’s alpha when devel- and future directions for psychological research. Develop- oping and reporting research instruments in science educa- mental Review, 41, 71–90. https://doi.org/10.1016/J.DR. tion. Research in Science Education, 48(6), 1273–1296. 2016.06.004 https://doi.org/10.1007/s11165-016-9602-2 Revelle, W. (2017). Psych: Procedures for personality and psy- Tamres, L. K., Janicki, D., & Helgeson, V. S. (2002). Sex dif- chological research [Statistical software]. ferences in coping behavior: A meta-analytic review and an Roberton, T., Daffern, M., & Bucks, R. S. (2012). Emotion examination of relative coping. Personality and Social Psy- regulation and aggression. Aggression and Violent Behavior, chology Review, 6(1), 2–30. https://doi.org/10.1207/S15327 17(1), 72–82. https://doi.org/10.1016/J.AVB.2011.09.006 957PSPR0601_1 Rood, L., Roelofs, J., Bo¨ gels, S. M., Nolen-Hoeksema, S., & Tang, J., Ma, Y., Guo, Y., Ahmed, N. I., Yu, Y., & Wang, J. Schouten, E. (2009). The influence of emotion-focused (2013). Association of aggression and non-suicidal self rumination and distraction on depressive symptoms in non- injury: A school-based sample of adolescents. PLOS ONE, clinical youth: A meta-analytic review. Clinical Psychology 8(10), Article e78149. https://doi.org/10.1371/JOURNAL. Review, 29(7), 607–616. https://doi.org/10.1016/J.CPR. PONE.0078149 2009.07.001 Thompson, R. A. (1994). Emotion regulation: A theme in Rosseel, Y. (2012). Lavaan: An R package for structural equa- search of definition. Monographs of the Society for Research tion modeling. Journal of Statistical Software, 48(2), 1–36. in Child Development, 59(2/3), 25. https://doi.org/10.2307/ https://doi.org/10.18637/jss.v048.i02 1166137 Royuela-Colomer, E., Ferna´ ndez-Gonza´ lez, L., & Orue, I. Ullman, J. B., & Bentler, P. M. (2013). Structural equation (2021). Longitudinal associations between internalizing modeling. In J. A. Schinka, W. F. Velicer & I. B. Weiner symptoms, dispositional mindfulness, rumination and (Eds.), Handbook of Psychology (2nd ed., pp. 661–690). impulsivity in adolescents. Journal of Youth and Adoles- John Wiley & Sons. https://doi.org/10.1002/9781118133880. cence, 50(10), 2067–2078. https://doi.org/10.1007/s10964- HOP202023 021-01476-2 van Buuren, S., & Groothuis-Oudshoorn, K. (2011). Mice: Saarikallio, S. (2010). Music as emotional self-regulation Multivariate imputation by chained equations in R. Journal throughout adulthood. Psychology of Music, 39(3), of Statistical Software, 45(3), 1–67. https://doi.org/10. 307–327. https://doi.org/10.1177/0305735610374894 18637/jss.v045.i03 Sass, D. A. (2016). Testing measurement invariance and com- van Lith, T., Schofield, M. J., & Fenner, P. (2013). Identifying paring latent factor means within a confirmatory factor anal- the evidence-base for art-based practices and their potential ysis framework. Journal of Psychoeducational Assessment, benefit for mental health recovery: A critical review. Dis- 29(4), 347–363. https://doi.org/10.1177/0734282911406661 ability and Rehabilitation, 35(16), 1309–1323. https://doi. Schafer, J. O., Naumann, E., Holmes, E. A., Tuschen-Caffier, ¨ org/10.3109/09638288.2012.732188 B., & Samson, A. C. (2016). Emotion regulation strategies in Viborg, N., Wa˚ ngby-Lundh, M., & Lundh, L. G. (2014). Reci- depressive and anxiety symptoms in youth: A meta-analytic procal prospective associations between disordered eating 20 Assessment 00(0) and other psychological problems in a community sample 155–168. https://doi.org/10.1097/00004703-200604000- of Swedish adolescent girls. Eating Behaviors, 15(1), 00014 159–163. https://doi.org/10.1016/J.EATBEH.2013.11.002 Zeman, J., Klimes-Dougan, B., Cassano, M., & Adrian, M. Waaddegaard, M., Thoning, H., & Petersson, B. (2003). Vali- (2007). Measurement issues in emotion research with chil- dation of a screening instrument for identifying risk beha- dren and adolescents. Clinical Psychology: Science and viour related to eating disorders. European Eating Disorders Practice, 14(4), 377–401. https://doi.org/10.1111/j.1468- Review, 11(6), 433–455. https://doi.org/https://doi.org/10. 2850.2007.00098.x 1002/erv.537 Zeman, J., Shipman, K., & Penza-Clyve, S. (2001). Develop- Wolff, J. C., Thompson, E., Thomas, S. A., Nesi, J., Bettis, A. ment and initial validation of the children’s sadness man- H., Ransford, B., Scopelliti, K., Frazier, E. A., & Liu, R. T. agement scale. Journal of Nonverbal Behavior, 25(3), (2019). Emotion dysregulation and non-suicidal self-injury: 187–205. https://doi.org/10.1023/A:1010623226626 A systematic review and meta-analysis. European Psychia- Zhao, K., Bai, Z. G., Bo, A., & Chi, I. (2016). A systematic try, 59, 25–36. https://doi.org/10.1016/J.EURPSY.2019.03. review and meta-analysis of music therapy for the older adults with depression. International Journal of Geriatric Psy- Wolgast, M., & Lundh, L. G. (2017). Is distraction an adaptive chiatry, 31(11), 1188–1198. https://doi.org/10.1002/GPS.4494 or maladaptive strategy for emotion regulation? A person- Zhou, Y., Daukantaite, _ D., Lundh, L. G., Wa˚ ngby-Lundh, oriented approach. Journal of Psychopathology and Beha- M., & Ryde, A. (2020). Adolescents’ emotion regulation vioral Assessment, 39(1), 117–127. https://doi.org/10.1007/ strategies questionnaire: Initial validation and prospective s10862-016-9570-x associations with nonsuicidal self-injury and other mental Yih, J., Uusberg, A., Taxer, J. L., & Gross, J. J. (2018). Better health problems in adolescence and young adulthood in a together: A unified perspective on appraisal and emotion Swedish Youth Cohort. Frontiers in Psychiatry, 11, 462. regulation. Cognition and Emotion, 33(1), 41–47. https:// https://doi.org/10.3389/fpsyt.2020.00462 doi.org/10.1080/02699931.2018.1504749 Zimmermann, P., & Iwanski, A. (2014). Emotion regulation Zaki, J., & Williams, W. C. (2013). Interpersonal emotion reg- from early adolescence to emerging adulthood and middle ulation. Emotion, 13(5), 803–810. https://doi.org/10.1037/ adulthood: Age differences, gender differences, and a0033839 emotion-specific developmental variations. International Zeman, J., Cassano, M., Perry-Parrish, C., & Stegall, S. Journal of Behavioral Development, 38(2), 182–194. https:// (2006). Emotion regulation in children and adolescents. doi.org/10.1177/0165025413515405 Journal of Developmental & Behavioral Pediatrics, 27(2),

Journal

Assessment – SAGE

Published: Mar 1, 2024

Keywords: emotion regulation; adolescence; psychopathology; psychometrics; questionnaire

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Adolescents’ Emotion Regulation Strategies Questionnaire–Extended: Further Development and Associations With Mental Health Problems in Adolescence

Adolescents’ Emotion Regulation Strategies Questionnaire–Extended: Further Development and Associations With Mental Health Problems in Adolescence

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Adolescents’ Emotion Regulation Strategies Questionnaire–Extended: Further Development and Associations With Mental Health Problems in Adolescence

Adolescents’ Emotion Regulation Strategies Questionnaire–Extended: Further Development and Associations With Mental Health Problems in Adolescence

References (95)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies