Cultural redevelopment and validation of the emotion regulation goals scale – culturally redeveloped version (ERGS-CRV): evidence from Iran

Abstract

Introduction

Emotion regulation goals (ERGs) represent an important yet understudied aspect of self-regulation, particularly in non-Western cultural contexts. Existing measures primarily focus on emotion-regulation strategies, and culturally appropriate instruments for assessing ERGs remain limited. The present study aimed to culturally redevelop and validate the Emotion Regulation Goals Scale – Culturally Redeveloped Version (ERGS-CRV) for use in an Iranian population.

Method

A total of 1,475 Iranian adults (67.9% women; M = 29.86, SD = 10.24; age 18–71) completed the ERGS-CRV. The validation process included exploratory and confirmatory factor analyses using split-sample cross-validation, reliability assessment, convergent and discriminant validity analyses, test–retest reliability, and measurement invariance testing across gender groups.

Results

Exploratory and confirmatory factor analyses supported a five-factor structure consisting of Pro-Hedonic, Contra-Hedonic, Performance, Pro-Social, and Impression Management dimensions. Reliability estimates ranged from 0.54 to 0.91 across subscales, with stronger internal consistency observed for the Performance, Pro-Social, and Impression Management dimensions. Evidence of convergent validity was strongest for the Pro-Social and Impression Management dimensions, acceptable for the Performance dimension, and comparatively weaker for the Pro-Hedonic and Contra-Hedonic dimensions. Discriminant validity was confirmed using both the Fornell–Larcker criterion and the heterotrait–monotrait ratio of correlations (HTMT). Test–retest coefficients indicated good to excellent temporal stability. Multi-group confirmatory factor analyses supported configural, metric, scalar, and strict measurement invariance across gender groups.

Conclusions

The ERGS-CRV provides a culturally informed measure of emotion-regulation goals in Iranian adults, with stronger psychometric support for the Pro-Social and Impression Management dimensions, acceptable support for the Performance dimension, and comparatively weaker support for the Pro-Hedonic and Contra-Hedonic dimensions. The findings support its use for research on emotion-regulation motives in Persian-speaking and potentially other non-Western populations.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s40359-026-05480-8.

Introduction

In navigating daily challenges, individuals continually regulate their emotions—often with implicit goals that shape their psychological and social functioning. This process involves modifying one’s emotional state to achieve a more desirable condition [1]. Like all forms of self-regulation, emotion regulation occurs for a specific reason. To fully understand self-regulation, it is essential to explore the motivations behind this process [2].

Initially, research on emotion regulation primarily focused on its processes—how individuals regulate their emotions. However, recent studies have shifted towards examining the content of regulation, specifically the goals that drive these processes [1]. Emotion regulation goals (ERGs) refer to the reasons individuals manage their emotions, highlighting the underlying motivations rather than the strategies employed [3, 4]. A key aspect of emotion regulation is activating a goal that influences the emotional regulation process [5]. Emotion regulation is fundamentally goal-driven: goals determine whether people attempt to regulate, guide the choice and implementation of strategies [6, 7], and shape the consequences of regulation [8].

To assess ERGs, researchers have used psychometric tools, such as the Hedonic and Eudaimonic Motives for Activities Scale HEMA [9], which measure individuals’ typical pursuit of hedonic (seeking pleasure) and eudaimonic (pursuing meaning) goals, rather than experimentally manipulating ERGs. Huta and Waterman [10] categorized these motives into four conceptual classes: directional (broad orientations toward hedonic pleasure or eudaimonic meaning), behavioral (the specific activities people choose to pursue these orientations), experiential (the emotional or cognitive states they aim to achieve, such as relaxation vs. personal growth), and functional (the expected outcomes of these pursuits, such as immediate enjoyment vs. long-term fulfillment). These categories align with the broader goal framework and clarify how hedonic and eudaimonic motives manifest. Huta [11] emphasized that hedonic and eudaimonic motives are best distinguished when assessed as directional. In a later revision, Huta [11] added an item, producing the HEMA-R. Cross-cultural studies on HEMA have yielded mixed results; some found no significant gender differences in the US and Japan [12, 13], others reported that women exhibit a stronger hedonic orientation than men, as seen in an Italian sample [14]. Other research indicates that women tend to experience a wider range of emotions and sustain them longer [15, 16], and that boys score higher on Contra-Hedonic motives while girls exhibit stronger Performance motives [17].

Eldesouky and English [4] classified emotion regulation goals (ERGs) into five categories: Pro-Hedonic, Contra-Hedonic, Performance, Impression Management, and Pro-Social, and developed the ERGS, validated in American undergraduates and community adults. Despite growing interest, culturally adapted and psychometrically validated measures for non-Western populations, including Persian speakers, remain scarce.

The psychometric properties of the ERGS have been examined in several cultural contexts. In Portugal [18], tested two independent samples—the general population and university students—and found that a refined 16-item version demonstrated strong reliability and validity. In Germany, Wilms et al. [19] modified the pro-hedonic and contra-hedonic subscales to reflect emotion expression rather than experience and adjusted the response format to six points; confirmatory factor analysis supported the scale’s five-factor structure. In China, Chen et al. [20] conducted multiple studies, including exploratory and confirmatory factor analyses and test-retest reliability, initially finding a four-factor structure but ultimately reaffirming the five-factor model after minor adjustments.

Although prior studies have adapted the ERGS in Western Europe and parts of Asia, its applicability in Middle Eastern populations remains largely unexplored. Iranian participants present a unique linguistic and cultural context, in which several original ERGS items were conceptually challenging, reflecting limited familiarity with nuanced emotional vocabulary, culture-specific interpretations, and aspects of social performance and impression management [21, 22]. Cultural norms in non-Western settings further shape emotional expression, affecting the relevance of Western-developed tools [23]. To address these challenges, the ERGS was culturally redeveloped to capture a broad spectrum of emotional goals—including hedonic and instrumental aims—while ensuring linguistic and cultural appropriateness. Although focused on Iran, the adapted scale may inform research in other Middle Eastern populations with similar characteristics. Accordingly, this study aimed to culturally redevelop and psychometrically validate the Emotion Regulation Goals Scale – Culturally Redeveloped Version (ERGS-CRV) in Iran, examining reliability, factor structure, and measurement invariance across genders, thus providing a culturally sensitive tool for future research on emotion regulation goals in non-Western contexts.

Method

Participants

Participants were 1,475 Iranian adults (67.9% women; Mage = 29.86 years, SD = 10.24; range = 18–71). This sample size exceeds commonly recommended thresholds for robust psychometric and factor-analytic research [24, 25] and allowed for split-sample cross-validation procedures.

Data were collected between March and September 2024 using a mixed recruitment strategy designed to increase demographic diversity. Participants were recruited through both in-person administration at Kharazmi University and several universities in Tehran, as well as online distribution via commonly used Persian-language social media and messaging platforms (e.g., Telegram, WhatsApp, and Instagram). Recruitment materials invited Persian-speaking adults to participate in a study examining everyday emotion-regulation goals.

Eligibility criteria included being at least 18 years old, fluent in Persian (Farsi), and willing to provide informed consent. Cases showing excessive missing data or patterned responding were excluded prior to analyses.

The study was approved by the Ethics Committee of Kharazmi University (IR.KHU.REC.1402.104), and all participants provided informed consent before participation.

Additional demographic characteristics of the sample are presented in the Supplementary Materials (Appendix A).

Measures

Emotion regulation goals scale – culturally redeveloped version (ERGS-CRV)

The Emotion Regulation Goals Scale – Culturally Redeveloped Version (ERGS-CRV) was developed to assess emotion-regulation goals within the Iranian cultural context. Although it was based on the original Emotion Regulation Goals Scale (ERGS) [4], the present version incorporated theoretically informed cultural and conceptual modifications. Although the original five-factor framework—Pro-Hedonic, Contra-Hedonic, Performance, Pro-Social, and Impression Management—was retained, several dimensions were refined to improve cultural relevance, contextual sensitivity, and conceptual precision for Persian-speaking respondents.

Importantly, the redevelopment process was not limited to linguistic translation. Rather, modifications were theoretically guided by cultural perspectives suggesting that emotion-regulation goals may vary depending on emotional valence, social context, and culturally shaped interpersonal norms (e.g., [26, 27]). Accordingly, the Performance and Impression Management dimensions were expanded to differentiate regulation of positive versus negative emotions, reflecting the possibility that individuals in collectivistic cultural contexts may regulate emotions differently depending on whether emotions facilitate task engagement, social harmony, or interpersonal evaluation.

Similarly, the Pro-Hedonic and Contra-Hedonic dimensions were refined by incorporating concrete and culturally resonant emotional examples to improve emotional salience and contextual comprehensibility. These modifications aimed to preserve the theoretical integrity of the original framework while allowing culturally meaningful expression of emotion-regulation goals.

The final ERGS-CRV includes 25 items across five subscales: Pro-Hedonic (3 items), Contra-Hedonic (3 items), Performance (6 items), Pro-Social (5 items), and Impression Management (8 items), rated on a 5-point Likert scale (1 = never to 5 = always). Subscale scores are calculated as the mean of items. Psychometric properties were evaluated in this study.

Hedonic and eudaimonic motives for activities-revised (HEMA-R)

To assess the convergent validity of the ERGS-CRV, the Hedonic and Eudaimonic Motives for Activities-Revised (HEMA-R; [9]) was administered. This measure was selected as it has previously been validated in Iran, demonstrating good reliability with Cronbach’s alpha coefficients of 0.88 for Eudaimonic and 0.78 for Hedonic motivation [28]. The HEMA-R assesses Hedonic (5 items) and Eudaimonic (4 items) motivations for activities, and the revised version [11] includes an additional item explicitly representing meaning. Participants respond on a 7-point Likert scale (1 = very low to 7 = very high).

Procedure

The ERGS-CRV was developed through a multi-phase process, including theoretical adaptation, translation, cognitive interviews, expert review, and psychometric evaluation. The original ERGS [4] five-factor structure—Pro-Hedonic, Contra-Hedonic, Performance, Pro-Social, and Impression Management—was retained, with targeted modifications to enhance cultural relevance and clarity for Persian-speaking respondents.

The scale was translated into Persian using forward-backward procedures [29] and piloted with 30 participants. Cognitive interviews identified ambiguities and potential response biases, which were addressed by refining item wording (e.g., replacing “emotion” with “feeling”), splitting Performance and Impression Management items into positive and negative emotion contexts. Revised items were carefully reviewed by psychologists specializing in emotion regulation and psychometrics to ensure conceptual accuracy and cultural appropriateness.

The finalized Persian version was back-translated into English, and any discrepancies were resolved to maintain alignment with the original theoretical framework. Large-scale administration involved 1,475 participants via online and in-person methods, ensuring demographic diversity.

As part of the cultural redevelopment process, the original seven-point response format was reduced to five response categories (Never, Infrequently, Sometimes, Often, Always). This modification was made to improve comprehensibility and response consistency among Iranian participants, as distinctions between several adjacent categories in the original format were considered difficult to differentiate reliably in Persian. Furthermore, previous research has shown that five-point response formats often provide psychometric properties comparable to longer scales while reducing respondent burden and improving category discrimination [30, 31]. Although this modification may somewhat limit direct score comparability with studies using the original seven-point ERGS, it was considered necessary to maximize cultural appropriateness, response quality, and interpretability within the target population.

Detailed item-level revisions and complete English and Persian versions of the ERGS-CRV are provided in the Supplementary Materials (Appendices B–E).

Revision of ERGS items

Item revisions were conducted to improve cultural relevance, emotional clarity, and conceptual sensitivity within the Iranian sociocultural context. Modifications were guided not only by linguistic considerations, but also by theoretical perspectives emphasizing cultural variation in emotion-regulation goals, emotional expression, and interpersonal functioning.

The Pro-Hedonic and Contra-Hedonic subscales retained their original structural organization; however, item wording was refined through the inclusion of concrete and culturally resonant emotional examples (e.g., joy, sadness, anger, and hope). These additions were intended to enhance emotional salience, reduce interpretive ambiguity, and facilitate more contextually grounded responses among Persian-speaking participants.

More substantial modifications were applied to the Performance and Impression Management dimensions. Specifically, these subscales were expanded to differentiate the regulation of positive versus negative emotions. This distinction was theoretically informed by cross-cultural research suggesting that emotional valence may differentially shape goal-directed behavior, self-regulation, and interpersonal functioning in collectivistic cultural contexts [26, 27]. For example, individuals may attempt to maintain positive emotions to sustain motivation and task engagement, while simultaneously downregulating emotions perceived as disruptive to concentration, social harmony, or self-presentation.

The Pro-Social subscale underwent comparatively minor revisions focused primarily on linguistic clarity and cultural relatability without altering its underlying conceptual framework.

Overall, the redevelopment process sought to preserve the theoretical architecture of the original ERGS while refining the operationalization of emotion-regulation goals in ways that were culturally meaningful for Persian-speaking respondents. These modifications extend the application of the ERGS framework beyond direct translation by incorporating culturally informed distinctions in emotional regulation and social functioning. Detailed item-level modifications and complete English and Persian versions of the ERGS-CRV are provided in the Supplementary Materials (Appendices B–E).

Data analysis

All statistical analyses were performed using SPSS version 26, R (versions 4.0.3 and 4.6.0). Initially, descriptive statistics (means, standard deviations, skewness, and kurtosis) were calculated for all items and subscales to examine the distributional properties of the data. Internal consistency was assessed with Cronbach’s alpha coefficients, ordinal alpha, and McDonald’s omega and omega3 coefficients, and test–retest reliability was evaluated over a one-month interval in a subsample. A one-month interval was selected because it is commonly recommended in psychometric research as a balance between minimizing recall effects and limiting the likelihood of substantial change in the underlying construct.

To evaluate the factorial structure of the ERGS-CRV, the total sample was randomly split into two independent subsamples of 738 and 737 participants, respectively, following recommended cross-validation procedures in psychometric research. An exploratory factor analysis (EFA) was conducted on the first subsample using principal axis factoring with oblimin rotation to identify the underlying dimensions. Sampling adequacy and factorability were evaluated using the Kaiser–Meyer–Olkin (KMO) index and Bartlett’s test of sphericity. Confirmatory factor analysis (CFA) was subsequently conducted on the second independent subsample to test and cross-validate the hypothesized five-factor structure identified in the EFA. Given the ordinal nature of the ERGS-CRV items and the use of five-point Likert-type response categories, the CFA models were estimated in R (lavaan package) using the robust weighted least squares estimator with mean and variance adjustment (WLSMV), which is widely recommended for ordered categorical indicators and non-normally distributed Likert-type data [32, 33]. Model fit was evaluated using multiple complementary fit indices, including the chi-square statistic divided by degrees of freedom (χ²/df), Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Incremental Fit Index (IFI), Normed Fit Index (NFI), Goodness-of-Fit Index (GFI), Standardized Root Mean Square Residual (SRMR), and Root Mean Square Error of Approximation (RMSEA). Conventional cutoff criteria were adopted to evaluate model adequacy (e.g., CFI/TLI ≥ 0.90 and RMSEA/SRMR ≤ 0.08) [34].

Measurement invariance (MI) across gender groups was examined using a multi-group confirmatory factor analysis (MGCFA) with the robust weighted least squares estimator (WLSMV) and theta parameterization, appropriate for ordinal indicators [35]. Following the sequential strategy recommended by Cheung and Rensvold (2002) and Chen (2007), four increasingly restrictive models—configural, metric, scalar, and residual variance invariance (strict invariance)—were tested hierarchically. Invariance at each step was evaluated based on changes in multiple fit indices rather than solely on the chi-square difference test, with thresholds of ΔCFI ≤ 0.01 and ΔRMSEA ≤ 0.015 considered evidence of invariance [36, 37].

Finally, convergent and discriminant validity were evaluated using multiple complementary approaches. Convergent validity was assessed by examining Pearson correlation coefficients between the ERGS-CRV subscales and corresponding dimensions of the Hedonic and Eudaimonic Motives for Activities-Revised (HEMA-R). In addition, Composite Reliability (CR) and Average Variance Extracted (AVE) were calculated based on standardized CFA loadings. Discriminant validity was examined using both the Fornell–Larcker criterion, comparing the square root of AVE values with inter-factor correlations, and the Heterotrait–Monotrait ratio of correlations (HTMT) [38], with values below 0.85 generally considered indicative of adequate discriminant validity. These analyses provided a comprehensive evaluation of the psychometric properties of the ERGS-CRV among Iranian adults.

Results

Descriptive statistics

Table 1 presents descriptive statistics for all 25 items (N = 1,475), including means, standard errors, standard deviations, skewness, and kurtosis. Item means ranged from 1.99 (Item 5) to 3.71 (Item 2), indicating moderate endorsement across most items. Skewness and kurtosis values were generally within ± 1, suggesting no extreme deviations from normality [39, 40].

Table 1.

Descriptive statistics for the ERGS-CRV items (N = 1,475)

	Mean	SE	SD	skewness	kurtosis	Min	max
Item 1	3.22	0.026	1.002	-0.287	-0.337	1	5
Item 2	3.71	0.024	0.936	-0.533	0.012	1	5
Item 3	3.38	0.027	1.036	-0.313	-0.515	1	5
Item 4	2.97	0.031	1.199	0.067	-0.916	1	5
Item 5	1.99	0.028	1.067	0.938	0.116	1	5
Item 6	2.86	0.028	1.086	0.176	-0.705	1	5
Item 7	3.09	0.028	1.066	-0.169	-0.549	1	5
Item 8	3.31	0.026	0.990	-0.372	-0.214	1	5
Item 9	3.25	0.028	1.093	-0.283	-0.612	1	5
Item 10	3.33	0.027	1.020	-0.280	-0.445	1	5
Item 11	3.14	0.028	1.068	-0.157	-0.613	1	5
Item 12	3.27	0.026	1.002	-0.272	-0.351	1	5
Item 13	3.39	0.027	1.048	-0.324	-0.480	1	5
Item 14	3.13	0.028	1.063	-0.143	-0.641	1	5
Item 15	3.40	0.026	1.004	-0.407	-0.295	1	5
Item 16	3.44	0.025	0.966	-0.449	-0.120	1	5
Item 17	3.23	0.026	0.990	-0.168	-0.413	1	5
Item 18	2.81	0.029	1.127	0.105	-0.750	1	5
Item 19	2.80	0.029	1.113	0.084	-0.759	1	5
Item 20	3.14	0.027	1.021	-0.115	-0.540	1	5
Item 21	3.06	0.027	1.035	-0.149	-0.464	1	5
Item 22	2.81	0.029	1.097	0.129	-0.648	1	5
Item 23	2.84	0.029	1.095	0.094	-0.670	1	5
Item 24	2.97	0.028	1.073	-0.101	-0.673	1	5
Item 25	2.94	0.028	1.068	0.000	-0.597	1	5

Note. SE = Standard Error; SD = Standard Deviation. Min = minimum possible score; Max = maximum possible score. Skewness and kurtosis values are based on standardized scores

Although Kolmogorov–Smirnov tests were statistically significant for all items, significance tests of normality are highly sensitive to sample size and may detect trivial deviations in large samples. Therefore, distributional assumptions were evaluated primarily using skewness and kurtosis values, which were generally within acceptable ranges [40]. Given the large sample size and the robustness of the statistical procedures employed, these minor deviations were not considered sufficient to preclude the use of parametric analyses [41]. These descriptive statistics provide an overview of response tendencies and distributional properties, thereby forming a solid basis for the reliability and validity analyses that follow.

Internal consistency and reliability estimates

Internal consistency was evaluated using Cronbach’s alpha, ordinal alpha, McDonald’s omega (ω), and omega3 coefficients for each ERGS-CRV subscale (Table 2). Overall, the Pro-Hedonic and Contra-Hedonic subscales demonstrated comparatively lower reliability estimates across indices, whereas the Performance, Pro-Social, and Impression Management subscales showed acceptable to excellent internal consistency.

Table 2.

Reliability estimates for the ERGS-CRV subscales

Construct	Cronbach’s α	Ordinal α	McDonald’s ω	McDonald’s ω3
ProH	0.561	0.610	0.486	0.493
ContraH	0.544	0.590	0.458	0.458
Performance	0.792	0.822	0.806	0.820
ProSocial	0.835	0.863	0.841	0.850
Impression	0.891	0.908	0.907	0.964

Note. Cronbach’s α values ≥ 0.70 are generally considered acceptable indicators of internal consistency. Ordinal alpha coefficients were computed based on polychoric correlations to account for the ordinal nature of the response categories. Given the small number of items in the ProH and ContraH subscales, relatively lower reliability coefficients were expected. McDonald’s ω3 represents a model-based reliability estimate derived from observed-score variance as implemented in the semTools package and should not be interpreted as a strict hierarchical omega coefficient from a bifactor model

Specifically, Cronbach’s alpha coefficients for the Pro-Hedonic and Contra-Hedonic subscales were 0.561 and 0.544, respectively. Ordinal alpha values were slightly higher (0.610 and 0.590), indicating improved reliability estimates when the ordinal nature of the Likert-type response format was taken into account. In contrast, the Performance (α = 0.792), Pro-Social (α = 0.835), and Impression Management (α = 0.891) subscales demonstrated satisfactory to excellent internal consistency across all reliability indices.

A similar pattern was observed for McDonald’s omega coefficients, which ranged from 0.458 to 0.907 across subscales. The comparatively lower reliability estimates for the Pro-Hedonic and Contra-Hedonic dimensions may partly reflect the limited number of items within these subscales, as reliability coefficients are sensitive to scale length. In addition, these dimensions assess relatively heterogeneous motivational tendencies involving both the amplification and maintenance of emotional states, which may naturally reduce item homogeneity. Such reductions in internal consistency for broad motivational constructs have also been reported in prior cross-cultural research (e.g., 42).

Item–total correlations further supported these findings, ranging from 0.309 to 0.440 for Pro-Hedonic, 0.346 to 0.477 for Contra-Hedonic, 0.435 to 0.598 for Performance, 0.619 to 0.651 for Pro-Social, and 0.617 to 0.718 for Impression Management. Overall, these results indicate satisfactory internal consistency for most ERGS-CRV subscales, while suggesting that the Pro-Hedonic and Contra-Hedonic dimensions may benefit from further psychometric refinement and potential expansion of item content in future research.

Test-retest reliability

Table 3 presents the test–retest reliability coefficients, assessed in an independent sample of 86 participants who completed the ERGS-CRV twice, with a one-month interval between administrations. The Pro-Hedonic, Contra-Hedonic, and Performance subscales demonstrated good stability (r = .809, 0.806, and 0.888, respectively), whereas the Pro-Social (r = .902) and Impression Management (r = .903) subscales showed excellent stability. These results indicate high temporal stability across all subscales, comparable to the original ERGS validation [4], and support the scale’s suitability for longitudinal research in Iranian contexts.

Table 3.

Test–retest reliability of the ERGS-CRV subscales (N = 86, one-month interval)

Subscale	r	95% CI	Interpretation
Pro-Hedonic	0.809	[0.720, 0.872]	Good
Contra-Hedonic	0.806	[0.716, 0.870]	Good
Performance	0.888	[0.837, 0.923]	Good
Pro-Social	0.902	[0.859, 0.933]	Excellent
Impression Management	0.903	[0.861, 0.934]	Excellent

Note. r = Pearson’s correlation coefficient. Interpretation based on Cohen’s (1988) guidelines: 0.50–0.69 = moderate, 0.70–0.89 = good, ≥ 0.90 = excellent

Factor structure

Exploratory factor analysis

Exploratory factor analysis (EFA) was conducted on the first half of the sample (n = 738) as part of the cross-validation procedure. Bartlett’s test of sphericity was significant (p < .001), and the Kaiser–Meyer–Olkin (KMO) measure indicated sampling adequacy (0.882), supporting factorability of the correlation matrix. Scree plot and eigenvalue analysis supported a five-factor structure, explaining 54.69% of the total variance. Principal axis factoring with direct oblimin rotation showed that all items loaded on their expected factors (factor loadings > 0.40; Table 4), supporting the hypothesized five-factor model.

Table 4.

Factor loadings from the exploratory factor analysis for the total sample (N = 738)

Item		Component 1	Component 2	Component 3	Component 4	Component 5
Item 1		0.208	0.058	0.055	0.095	0.418
Item 2		-0.096	0.057	-0.094	0.175	0.769
Item 3		0.045	0.023	0.045	0.038	0.728
Item 4		-0.042	0.065	-0.083	0.724	0.064
Item 5		0.224	0.095	0.153	0.451	-0.321
Item 6		0.001	-0.059	-0.035	0.754	0.109
Item 7		0.048	0.617	0.043	0.111	-0.110
Item 8		-0.024	0.637	-0.021	-0.008	0.107
Item 9		-0.021	0.767	-0.007	-0.033	-0.080
Item 10		-0.047	0.711	-0.049	-0.031	0.138
Item 11		0.032	0.741	-0.011	0.007	-0.064
Item 12		0.018	0.667	-0.021	-0.037	0.148
Item 13		-0.006	0.038	-0.761	0.066	-0.037
Item 14		0.205	-0.073	-0.655	0.101	-0.002
Item 15		-0.007	0.001	-0.825	-0.064	-0.047
Item 16		-0.043	0.065	-0.796	-0.011	0.056
Item 17		0.259	0.002	-0.616	0.010	-0.003
Item 18		0.693	-0.042	-0.127	0.062	-0.070
Item 19		0.720	-0.029	-0.075	0.071	-0.043
Item 20		0.721	0.115	-0.080	0.097	-0.025
Item 21		0.673	0.079	-0.084	-0.078	0.085
Item 22		0.821	-0.049	0.051	0.117	0.045
Item 23		0.788	-0.054	0.030	0.086	0.066
Item 24		0.696	0.030	-0.033	-0.085	0.015
Item 25		0.698	0.038	-0.034	-0.030	0.015

Note. Factor loadings greater than 0.40 are bolded. Extraction method: principal axis factoring; rotation method: direct oblimin

Detailed EFA results for female and male subsamples, including factor loadings and cross-loadings, are provided in Appendix F-1.

Confirmatory factor analysis

CFA was conducted on the second half of the sample (n = 737), independent from the EFA dataset, to evaluate the hypothesized five-factor structure of the ERGS-CRV. Given the ordinal nature of the ERGS-CRV items and the use of five-point Likert-type response categories, the model was estimated in R (lavaan package) using the robust weighted least squares estimator with mean and variance adjustment (WLSMV), which is recommended for ordered categorical indicators [32, 33].

Three competing models were evaluated. Model 1 specified the original five-factor structure without correlated residuals. Model 2 retained the five-factor structure while allowing residual correlations between four pairs of items within the Impression Management factor (Items 18–19, 20–21, 22–23, and 24–25). These item pairs share nearly identical content and differ only with respect to the emotional valence referenced (negative emotions vs. positive emotions), making correlated residuals theoretically plausible. Specifically, each pair assesses the same interpersonal goal (e.g., avoiding rejection, avoiding unfavorable impressions, gaining approval, or creating a positive impression) across different emotional contexts. Model 3 tested an alternative method-factor explanation in which shared wording effects were modeled using an additional latent method factor.

As shown in Table 5, Model 2 demonstrated the best overall fit to the data (CFI = 0.987, TLI = 0.986, RMSEA = 0.053, SRMR = 0.045), outperforming both the original five-factor model (Model 1) and the alternative method-factor model (Model 3). These findings suggest that the residual associations are more parsimoniously explained by the paired wording structure of the Impression Management items than by a broader method effect. These findings also supported model adequacy according to conventional recommendations [34]. Although the chi-square test was statistically significant, this result is common in large samples and does not necessarily indicate misfit [40].

Table 5.

Comparison of Competing CFA Models for the ERGS-CRV (Total Sample, n = 737)

Model	χ²	df	CFI	TLI	RMSEA	SRMR
Model 1: Original five-factor model	2602.61	265	0.973	0.970	0.077	0.053
Model 2: Five-factor model with correlated residuals	1354.55	261	0.987	0.986	0.053	0.045
Model 3: Five-factor model with method factor	2332.79	257	0.976	0.972	0.074	0.052

Note. Model 2 allowed correlated residuals between Items 18–19, 20–21, 22–23, and 24–25 within the Impression Management factor. Model 3 included an additional latent method factor to account for potential wording-related variance. CFI = Comparative Fit Index; TLI = Tucker–Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Square Residual

Standardized factor loadings ranged from 0.47 to 0.82 (Fig. 1) and were statistically significant. Overall, the findings provide strong support for the structural validity of the five-factor ERGS-CRV model.

Fig. 1.

This figure shows the standardized path coefficients for the confirmatory factor analysis model based on the total sample

Gender-specific CFA results, including fit indices and correlated errors, are presented in Appendix F-1, with Figures S1–S2 depicting the five-factor model for female and male subsamples. Overall, these findings support the five-factor structure and provide evidence for the structural validity of the ERGS-CRV in the Iranian context.

Construct validity

Convergent validity

To comprehensively evaluate the convergent validity of the ERGS-CRV subscales, both internal construct reliability and criterion-related validity were examined.

First, Composite Reliability (CR) and Average Variance Extracted (AVE) were computed for each subscale based on standardized CFA loadings to assess the consistency and amount of variance explained by the latent constructs [42]. As shown in Table 6, CR values ranged from 0.501 to 0.891, indicating moderate to excellent construct reliability across the five dimensions.

Table 6.

Composite Reliability and Average Variance Extracted (AVE) for ERGS-CRV Subscales (N = 1,475)

Factor	No. of Items	Composite Reliability (CR)	Average Variance Extracted (AVE)	Interpretation
Pro-Hedonic	3	0.549	0.271	CR and AVE below conventional thresholds; may reflect conceptual complexity and culturally nuanced emotional goals
Contra-Hedonic	3	0.501	0.246	CR and AVE below conventional thresholds; may reflect heterogeneous motivations toward negative emotional experience
Performance	6	0.791	0.453	CR acceptable (> 0.70); AVE slightly below 0.50, consistent with broad performance-oriented emotion regulation goals
Pro-Social	5	0.833	0.568	Strong CR and AVE; evidence of good convergent validity
Impression Management	8	0.891	0.597	Excellent CR and AVE; strong convergent validity

Note. CR = Composite Reliability; AVE = Average Variance Extracted. Composite reliability and AVE values were calculated using standardized factor loadings obtained from the confirmatory factor analysis model. According to Fornell and Larcker (1981), AVE values above 0.50 indicate adequate convergent validity; however, convergent validity may still be considered acceptable when CR exceeds 0.60 despite AVE values below 0.50. Lower CR and AVE estimates for the Pro-Hedonic and Contra-Hedonic subscales may partly reflect the limited number of items and the conceptual heterogeneity of emotion-regulation goals across cultural contexts

Convergent validity was clearly supported for the Pro-Social (AVE = 0.568) and Impression Management (AVE = 0.597) subscales, both of which exceeded the conventional 0.50 threshold and demonstrated strong composite reliability. These findings suggest that the items within these dimensions capture relatively cohesive and socially consistent emotion-regulation goals.

The Performance subscale demonstrated acceptable composite reliability (CR = 0.791), although its AVE (0.453) remained slightly below the recommended cutoff. This pattern may reflect the conceptual breadth of the construct rather than inadequate measurement quality. Specifically, the Performance items assess emotion regulation across both positive and negative emotional states in the service of maintaining concentration, task persistence, and resistance to distraction. Because the subscale encompasses regulation of multiple emotional experiences within achievement-oriented contexts, moderate heterogeneity among items is theoretically expected. According to Fornell and Larcker (1981), convergent validity may still be considered acceptable when AVE is below 0.50 but composite reliability exceeds 0.60.

In contrast, the Pro-Hedonic (CR = 0.549, AVE = 0.271) and Contra-Hedonic (CR = 0.501, AVE = 0.246) subscales demonstrated comparatively weaker convergent validity. Importantly, these dimensions assess highly nuanced motivational orientations toward emotional experience, including attempts to increase, decrease, or maintain both pleasant and unpleasant affective states. Such goals may carry different interpersonal, cultural, and contextual meanings across individuals, particularly in collectivistic cultural settings where emotional moderation and social appropriateness often shape emotional preferences and regulation strategies.

Moreover, the relatively small number of items within these subscales (three items each) likely contributed to lower AVE and reliability estimates. The heterogeneity observed in these dimensions may therefore reflect the conceptual complexity and contextual variability of hedonic and contra-hedonic emotion-regulation goals rather than solely poor psychometric functioning. Similar patterns have been reported in cross-cultural studies of affective and motivational constructs, where culturally shaped emotional meanings can reduce internal coherence despite retaining theoretical relevance (e.g [43]).

Second, criterion-related convergent validity was assessed by examining correlations between ERGS-CRV subscales and corresponding dimensions of the Hedonic and Eudaimonic Motives for Activities-Revised (HEMA-R) scale [9] in a subsample of 67 participants. Although this subsample was smaller than the main validation sample, previous methodological work suggests that bivariate correlation estimates in the effect-size range commonly observed in psychological research tend to stabilize within approximately this range of sample sizes [44]. Accordingly, the results should be interpreted as preliminary and warrant replication in larger independent samples.

To provide more informative effect-size estimates, 95% bootstrap confidence intervals (1,000 resamples) were calculated for all correlation coefficients.

Pearson correlation analyses revealed a significant positive association between the Pro-Hedonic subscale of ERGS-CRV and the Hedonic subscale of HEMA-R (r = .477, p < .001, 95% CI [0.301, 0.629]), supporting the convergence hypothesis. In contrast, the correlation between the Performance subscale of the ERGS-CRV and the Eudaimonic subscale of the HEMA-R was small and non-significant (r = .078, p = .265, 95% CI [− 0.221, 0.379]). This finding may indicate that the two constructs are related only indirectly or represent partially distinct motivational processes. Whereas the HEMA-R Eudaimonic dimension primarily assesses motives related to personal growth, authenticity, self-development, and meaning in life, the Performance dimension of the ERGS-CRV focuses on regulating emotional states to facilitate concentration, task persistence, and task completion. Thus, although both constructs may be broadly goal-oriented, they operate at different levels of motivational functioning and should not necessarily be expected to demonstrate strong convergence.

Overall, the findings provide partial to strong evidence for the convergent validity of the ERGS-CRV subscales, with particularly robust support observed for the Pro-Social and Impression Management dimensions.

Discriminant validity

Inter-factor correlations between ERGS-CRV Subscales

Table 7 displays the intercorrelations among the ERGS-CRV subscales. Pro-Social and Impression Management showed a strong association (r = .570, p < .001), supporting the hypothesis that socially oriented regulation goals are closely linked [4]. Pro-Hedonic and Contra-Hedonic were unrelated (r = –.038, p > .05), corroborating their theoretical distinction. The remaining correlations were weak to moderate (r = .107 to 0.309, p < .001), providing evidence of discriminant validity and supporting the construct validity of the ERGS-CRV in line with previous cross-cultural findings [1].

Table 7.

Correlation matrix and square root of average variance extracted (√AVE) for ERGS-CRV subscales (N = 1,475)

Pro-Hedonic	Pro-Hedonic	Contra-Hedonic	Performance	Pro-Social	Impression Management
	0.493	-0.038	0.309**	0.207**	0.171**
Contra-Hedonic	-0.038	0.458	0.107**	0.146**	0.261**
Performance	0.309**	0.107**	0.628	0.217**	0.261**
Pro-Social	0.207**	0.146**	0.217**	0.709	0.570**
Impression Management	0.171**	0.261**	0.261**	0.570**	0.712

Note. **p** < 0.001. Diagonal values (in bold) represent the square root of the average variance extracted (√AVE) for each subscale. Off-diagonal values are Pearson correlations between subscales. According to Fornell and Larcker (1981), discriminant validity is supported when the √AVE for each subscale is greater than its correlations with other subscales

To provide a more rigorous test of discriminant validity, the square root of the average variance extracted (√AVE) for each latent construct was compared against the inter-construct correlations, following the Fornell and Larcker (1981) criterion. Specifically, discriminant validity is established when the √AVE for each construct exceeds its highest correlation with any other construct in the model [42]. As detailed in Table 7, the diagonal elements display the √AVE values, while the off-diagonal elements show the Pearson correlations among the ERGS-CRV subscales. This criterion was met across all subscales, thus confirming that each latent factor is empirically distinct and measures a unique construct.

HTMT criterion

To evaluate the discriminant validity of the ERGS-CRV subscales, the Heterotrait–Monotrait ratio of correlations (HTMT) was computed as a more stringent criterion for assessing construct distinctiveness (Henseler et al., 2015). Unlike traditional approaches such as the Fornell–Larcker criterion, HTMT provides a more sensitive assessment of the degree to which constructs are empirically distinct.

As shown in Table 8, all HTMT values were well below the conservative threshold of 0.85, indicating adequate discriminant validity across all latent constructs. The highest observed value was between the Pro-Social and Impression Management subscales (HTMT = 0.660), suggesting a moderate level of conceptual relatedness but remaining clearly below the critical cutoff for discriminant validity concerns. All other inter-construct correlations ranged from 0.182 to 0.514, further supporting the empirical distinctiveness of the ERGS-CRV dimensions.

Table 8.

Heterotrait–Monotrait (HTMT) Ratio of Correlations for ERGS-CRV Subscales

Pro-Hedonic	Pro-Hedonic	Contra-Hedonic	Performance	Pro-Social	Impression Management
	—	0.261	0.514	0.336	0.268
Contra-Hedonic	0.261	—	0.182	0.259	0.414
Performance	0.514	0.182	—	0.269	0.312
Pro-Social	0.336	0.259	0.269	—	0.660
Impression Management	0.268	0.414	0.312	0.660	—

Note. HTMT = Heterotrait–Monotrait ratio of correlations used to assess discriminant validity among latent constructs. Values below 0.85 indicate adequate discriminant validity. All HTMT coefficients in the present study were below the recommended threshold, supporting the distinctiveness of the ERGS-CRV subscales

Overall, these findings provide strong evidence that the five subscales of the ERGS-CRV measure related but empirically distinguishable constructs, supporting the factorial structure of the scale and confirming that each dimension captures a unique aspect of regulatory goal orientation.

Measurement invariance across gender

To examine whether the ERGS-CRV operated equivalently across gender groups, measurement invariance was evaluated using multi-group confirmatory factor analysis (MGCFA). Given the ordinal nature of the items, analyses were conducted using the WLSMV estimator with theta parameterization. Following recommended procedures for ordinal indicators, configural, metric, scalar, and strict invariance models were tested sequentially (Table 9).

Table 9.

Model Fit Indices for the Measurement Invariance Testing Across Gender (N = 1475)

Model	χ²	df	CFI	TLI	RMSEA	SRMR	ΔCFI	ΔRMSEA
Configural	2889.11	530	0.974	0.971	0.078	0.057	—	—
Metric	3016.18	550	0.973	0.971	0.078	0.058	−0.001	0.000
Scalar	2997.43	600	0.974	0.974	0.074	0.057	+ 0.001	−0.004
Strict	3181.91	625	0.972	0.973	0.075	0.058	−0.002	+ 0.001

Note. χ² = chi-square statistic; df = degrees of freedom; CFI = Comparative Fit Index; TLI = Tucker–Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Square Residual; ΔCFI = change in CFI relative to the previous model; ΔRMSEA = change in RMSEA relative to the previous model. Measurement invariance was evaluated primarily based on changes in approximate fit indices (ΔCFI ≤ 0.010 and ΔRMSEA ≤ 0.015)

As shown in Table 9, the configural invariance model demonstrated an acceptable fit to the data (χ² = 2889.11, df = 530, CFI = 0.974, TLI = 0.971, RMSEA = 0.078, SRMR = 0.057), indicating that the five-factor structure of the ERGS-CRV was replicated across male and female groups. Factor loading patterns were highly similar across groups, supporting the conceptual equivalence of the latent structure.

Metric invariance was then evaluated by constraining factor loadings to equality across gender groups. Although the chi-square difference test was statistically significant (Δχ² = 56.17, Δdf = 20, p < .001), changes in approximate fit indices were negligible (ΔCFI = − 0.001, ΔRMSEA < 0.001, ΔSRMR = 0.001), remaining well below recommended cutoff criteria for noninvariance. These findings support metric invariance and indicate that item-factor relationships were comparable across men and women.

Next, scalar invariance was tested by additionally constraining item thresholds to equality across groups. The scalar model demonstrated comparable and slightly improved approximate fit indices relative to the metric model (χ² = 2997.43, df = 600, CFI = 0.974, TLI = 0.974, RMSEA = 0.074, SRMR = 0.057). Changes in fit indices were minimal (ΔCFI = + 0.001, ΔRMSEA = − 0.004, ΔSRMR = − 0.001), providing strong evidence for scalar invariance. This result suggests that latent mean comparisons across gender groups can be interpreted meaningfully.

Finally, strict invariance was evaluated by constraining residual variances across groups. Although the chi-square difference test remained statistically significant (Δχ² = 152.49, Δdf = 25, p < .001), changes in practical fit indices were again trivial (ΔCFI = − 0.002, ΔRMSEA = 0.001, ΔSRMR = 0.001). Therefore, strict invariance was also supported.

Overall, the findings indicate that the ERGS-CRV demonstrates configural, metric, scalar, and strict invariance across gender groups [36]. The stability of fit indices across increasingly constrained models suggests that the questionnaire functions equivalently for men and women and that observed differences can be interpreted as reflecting substantive differences in latent constructs rather than measurement bias [45].

Discussion

Emotion regulation goals are an important yet relatively understudied aspect of self-regulation, compared with extensively researched strategies [4, 8, 46]. This study aimed to culturally redevelop and psychometrically validate the Emotion Regulation Goals Scale – Culturally Redeveloped Version (ERGS-CRV) in Iran. Rather than a direct translation, items were carefully adapted to address cultural and linguistic challenges identified through cognitive interviews. Many original items were conceptually ambiguous due to limited familiarity with nuanced emotional vocabulary and culture-specific interpretations [21, 22]. To enhance clarity and cultural relevance, abstract terms such as “positive” and “negative” emotions were replaced with concrete examples reflecting Iranian emotional experiences. The final 25-item scale comprises five theoretically grounded subscales—Pro-Hedonic, Contra-Hedonic, Performance, Pro-Social, and Impression Management—preserving the original conceptual framework [3] while ensuring clarity, cultural relevance, and psychometric adequacy across diverse Iranian populations.

ERGS-CRV subscales demonstrated acceptable to excellent internal consistency, except for the Pro-Hedonic and Contra-Hedonic subscales, consistent with previous research [17, 20]. Lower reliability may partly reflect the small number of items within these subscales, as shorter scales often yield lower inter-item correlations [47]. In addition, these dimensions assess relatively complex and heterogeneous motivational orientations toward emotional experience, including preferences for increasing, maintaining, or decreasing affective states across different contexts, which may naturally reduce internal consistency. Cultural factors may also contribute to this pattern. In collectivistic contexts such as Iran, emotional goals are often shaped by interpersonal expectations and social considerations, potentially increasing variability in how individuals interpret and endorse hedonic and contra-hedonic goals. Together, these findings suggest that further refinement and expansion of item content may strengthen the reliability of these dimensions in future research.

Test-retest reliability analyses indicated good to excellent temporal stability across all ERGS-CRV subscales, demonstrating that the scale consistently measures the same constructs over time. This finding contrasts with Chen et al. [20], who reported adequate stability only for the Pro-Hedonic subscale. Overall, these results support the ERGS-CRV as a reliable instrument for assessing emotion regulation goals in the Iranian context.

Exploratory factor analysis provided strong support for the structural validity of the ERGS-CRV. The expected five-factor solution emerged clearly in the exploratory sample, with all items loading on their theoretically intended dimensions. Importantly, this structure was identified in one randomly selected subsample and subsequently replicated in an independent confirmatory sample, providing particularly strong evidence for the stability and generalizability of the factor structure. These findings suggest that the original conceptual framework proposed by Eldesouky and English [3] can be meaningfully extended to Persian-speaking populations and are consistent with previous cross-cultural validation studies [18, 19].

Confirmatory factor analyses provided strong support for the five-factor structure of the ERGS-CRV. Importantly, the model allowing theoretically justified correlated residuals among paired Impression Management items demonstrated superior fit relative to both the original five-factor model and an alternative method-factor model. These findings suggest that the shared variance among these item pairs is more parsimoniously explained by their common interpersonal content across positive and negative emotional contexts than by a broader wording or method effect. These results are consistent with previous research [4, 19, 20] and demonstrate the scale’s theoretical robustness and cultural adaptability for assessing emotion regulation goals across genders and diverse populations.

Contrary to Chen et al. [20], who reported a single factor combining Pro-Social and Impression Management goals, our findings supported their conceptual distinction. Although Iran exhibits collectivist tendencies [48], its intermediate level of collectivism relative to China may allow clearer separation between goals aimed at social harmony (Pro-Social) and self-presentation (Impression Management). Moreover, cultural emphasis on honor, dignity, and social status (aberoo [49]) likely increases the salience of Impression Management as an independent motivational construct. These cultural dynamics may explain why our results align more closely with Western research patterns, where differentiation between social and self-presentational goals is more pronounced, than with Chinese studies, where stronger collectivism may obscure this distinction [7].

Intercorrelations among ERGS-CRV subscales showed positive associations for all pairs except Pro-Hedonic and Contra-Hedonic. Contra-Hedonic exhibited the weakest correlations, supporting its independent functioning [20]. The strongest correlation was between Pro-Social and Impression Management, consistent with prior research [4, 20]. Although the association between these dimensions was relatively strong, several findings supported their empirical distinctiveness. The correlation accounted for approximately 32% shared variance (r² ≈ 0.32), HTMT values remained well below recommended thresholds, the Fornell–Larcker criterion was satisfied, and both EFA and CFA consistently supported separate latent factors. Together, these findings suggest that socially oriented emotion-regulation goals and self-presentational goals are closely related but distinguishable constructs within the Iranian context. This pattern indicates that individuals regulating emotions for social purposes may also be motivated by self-presentation concerns, reflecting the interplay of social harmony and personal dignity in Iranian culture.

Beyond the observed intercorrelations, more stringent tests of discriminant validity further supported the distinctiveness of the ERGS-CRV dimensions. Both the Fornell–Larcker criterion and the HTMT approach indicated that all five subscales represent empirically separable constructs. Notably, although Pro-Social and Impression Management demonstrated the strongest association, their HTMT value remained well below recommended thresholds and the Fornell–Larcker criterion was fully satisfied. These findings suggest that emotion-regulation goals aimed at social harmony, performance, hedonic experience, and self-presentation are related yet conceptually distinguishable motivational orientations. The convergence of evidence across correlational, factor-analytic, and discriminant-validity analyses strengthens confidence in the multidimensional structure proposed by the original ERGS framework [3, 4].

Convergent validity of the ERGS-CRV received mixed but generally supportive evidence. At the construct level, Composite Reliability (CR) and Average Variance Extracted (AVE) values provided particularly strong support for the Pro-Social and Impression Management dimensions, both of which exceeded conventional thresholds and demonstrated substantial shared variance among their indicators. The Performance dimension also showed acceptable construct reliability, although its AVE remained slightly below the recommended criterion. According to Fornell and Larcker [42], convergent validity may still be considered adequate when composite reliability remains satisfactory despite AVE values slightly below 0.50. The present pattern may therefore reflect the conceptual breadth of the Performance construct rather than inadequate measurement quality, as performance-related emotion-regulation goals encompass the regulation of both positive and negative emotional states in the service of concentration, persistence, and goal attainment [1].

In contrast, convergent validity evidence was comparatively weaker for the Pro-Hedonic and Contra-Hedonic dimensions. Although their CR and AVE values fell below conventional standards, these findings should be interpreted in light of the conceptual complexity of hedonic emotion-regulation goals. These dimensions assess diverse motivational tendencies related to increasing, maintaining, or reducing emotional experiences across different contexts, which may naturally reduce internal homogeneity. Furthermore, the limited number of items within each subscale likely contributed to lower reliability and AVE estimates, as reliability coefficients are partly dependent on scale length [47]. Similar patterns have been reported in research on broad affective and motivational constructs, where theoretically meaningful dimensions may demonstrate modest internal consistency because they encompass heterogeneous emotional preferences and context-dependent regulatory motives [1, 43]. Consequently, the relatively lower convergence observed for these dimensions may reflect conceptual heterogeneity rather than solely psychometric weakness. Accordingly, evidence for the Pro-Hedonic and Contra-Hedonic dimensions should be considered preliminary, and further item refinement and validation in independent samples are warranted.

Additional support for convergent validity was obtained through associations with the Hedonic and Eudaimonic Motives for Activities-Revised (HEMA-R). As expected, the Pro-Hedonic dimension demonstrated a moderate positive association with the HEMA Hedonic subscale, supporting the theoretical convergence between emotion-regulation goals aimed at promoting desirable emotional states and broader hedonic motivational orientations. In contrast, the Performance dimension was not significantly associated with the HEMA Eudaimonic subscale. This finding suggests that the two constructs may represent related but distinct motivational domains. Whereas the Eudaimonic dimension of the HEMA-R focuses on personal growth, authenticity, excellence, and meaning in life [9, 11], the Performance dimension of the ERGS-CRV reflects instrumental emotion-regulation goals aimed at facilitating concentration, persistence, and task accomplishment. Consistent with theoretical models of instrumental emotion regulation [1], regulating emotions for performance-related purposes may not necessarily coincide with broader eudaimonic motivations.

Nevertheless, because the criterion-related validity analyses were conducted in a relatively small subsample and one theoretically expected association was not statistically significant, these findings should be interpreted cautiously and replicated in larger independent samples. Overall, the pattern of results provides partial to strong support for the convergent validity of the ERGS-CRV, with the most robust evidence observed for the Pro-Social and Impression Management dimensions.

Beyond demonstrating factorial validity, the ERGS-CRV achieved configural, metric, scalar, and strict invariance across gender groups. These findings indicate that the factor structure, item loadings, thresholds, and residual variances operate similarly for men and women. Consequently, observed gender differences are unlikely to reflect measurement bias and may be interpreted as meaningful differences in underlying emotion-regulation goals. This finding is particularly important because it supports the meaningful comparison of latent ERGS-CRV constructs across male and female respondents [50].

Taken together, findings from reliability analyses, factor analytic procedures, discriminant and convergent validity assessments, and measurement invariance testing support the psychometric adequacy of the ERGS-CRV, with stronger evidence for the Performance, Pro-Social, and Impression Management dimensions and more preliminary support for the Pro-Hedonic and Contra-Hedonic dimensions. A notable contribution of the present study is that the ERGS-CRV represents a culturally redeveloped instrument rather than a direct linguistic translation of the original ERGS. The redevelopment process involved conceptual refinements, contextual elaborations, differentiation of emotional valence within selected dimensions, and adaptation of the response format. These modifications were designed to preserve the theoretical foundations of the original model while improving cultural relevance and interpretability for Persian-speaking populations. The findings suggest that culturally informed redevelopment may provide a promising approach for extending emotion-regulation theories across diverse cultural contexts, although replication in more representative and demographically diverse samples is needed.

Limitations and future directions

Several limitations should be noted. First, the sample was exclusively Iranian, which may limit generalizability to other cultural contexts. Future research should examine the ERGS-CRV in diverse populations, including non-Western samples, to enable cross-cultural comparisons. Second, additional qualitative studies could provide deeper insights into how individuals interpret and prioritize emotion regulation goals. Third, evidence for the Pro-Hedonic and Contra-Hedonic subscales remains preliminary. Their relatively low internal consistency and convergent-validity indices suggest that further item refinement, expansion, and validation in independent samples are warranted. Fourth, Convergent validity was examined in a relatively small subsample (n = 67), and the theoretically expected association between the Performance and Eudaimonic dimensions was not statistically significant. Therefore, these findings should be interpreted cautiously until replicated in larger independent samples. Finally, participants were recruited using a non-probability sampling strategy, the sample was predominantly female, and information regarding socioeconomic status and urban versus rural residence was not available. Therefore, the generalizability of the findings to the broader Iranian population should be interpreted cautiously. Future studies would benefit from more representative sampling strategies and the inclusion of sociodemographic indicators.

Despite these limitations, the findings have important implications. Validating the ERGS-CRV in Iran extends its cross-cultural applicability and highlights both universal and culture-specific aspects of emotion regulation goals. The scale provides a reliable tool for researchers and clinicians to assess regulatory goals across populations, supporting culturally informed interventions and prevention programs in educational and clinical settings, which may enhance psychological well-being and adaptive functioning. Although the present findings support the validity of the ERGS-CRV within the Iranian context, future cross-cultural research should examine measurement invariance across language versions and cultural groups to determine the extent to which the ERGS-CRV and the original ERGS can be directly compared.

Conclusion

This study provides evidence for the cultural redevelopment and psychometric validation of the Emotion Regulation Goals Scale – Culturally Redeveloped Version (ERGS-CRV) in Iran. Through rigorous adaptation and testing, the scale demonstrated satisfactory reliability, temporal stability, and factorial validity across gender groups, although evidence for the Pro-Hedonic and Contra-Hedonic dimensions remains preliminary and requires further validation. These findings support the theoretical framework of Eldesouky and English (2019a). Findings highlight both universal and culture-specific features of emotion regulation goals: the five-factor structure was replicated, yet cultural dynamics influenced the distinctiveness of subscales, particularly Pro-Social and Impression Management. The ERGS-CRV offers a reliable instrument for assessing motivational goals in Persian-speaking and broader non-Western populations — although additional refinement and independent validation of the Pro-Hedonic and Contra-Hedonic dimensions are needed before broader generalization, and further validation in more representative samples is warranted given the non-probability recruitment strategy and demographic limitations of the present study—, supporting cross-cultural research and culturally informed interventions.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

M.M. conceptualized and supervised the study, led the cultural adaptation and redevelopment of the ERGS-CRV, designed the psychometric analysis strategy, conducted the final statistical analyses, interpreted the results, and prepared the final manuscript. K.S. contributed to data collection, data organization, and performed initial data screening. L.E. provided feedback on the manuscript and offered general evaluation and advice regarding the study design. All authors reviewed and approved the final version of the manuscript.

Funding

No financial support was received for this study.

Data availability

The present study was not preregistered. To enhance transparency and reproducibility, all study materials and anonymized data are available via a private, view-only link on the Open Science Framework (OSF): https://osf.io/d7eqf/?view_only=cc4ae8eb9e954adfa6982324a50936f5. This link allows access for viewing and downloading without making the project publicly editable.

Declarations

Ethics approval and consent to participate

This study received ethical approval from the Ethics Committee of Kharazmi University (Approval ID: IR.KHU.REC.1402.104, Approval Date: 2024-01-16). All procedures involving human participants were performed in accordance with the ethical standards of the institutional research committee and with the Declaration of Helsinki and relevant guidelines and regulations. All participants provided informed consent before taking part in the study. Data were collected and stored anonymously to ensure participant confidentiality and privacy.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.