https://dx.doi.org/10.24016/2026.v12.512

ORIGINAL ARTICLE

 

 

Can the Difficulties in Emotion Regulation Scale be even shorter? Development of the DERS-7

 

Pablo D. Valencia1*, Anabel De la Rosa-Gómez2, Mariana Cabral-Familiar3, Alejandrina Hernández-Posadas2, Lorena A. Flores-Plata2

1 Coordinación de Universidad Abierta y Educación Digital, Universidad Nacional Autónoma de México, Mexico City, Mexico.

2 Facultad de Estudios Superiores Iztacala, Universidad Nacional Autónoma de México, Tlalnepantla, Mexico.

3 Portobello Institute, Dublin, Ireland.

 

* Correspondence: pablo.valencia@ired.unam.mx

 

Received: February 01, 2026 | Revised: March 03, 2026 | Accepted: April 23, 2026 | Published Online: June 14, 2026.

 

CITE IT AS:

Valencia, P. D., De la Rosa-Gómez, A., Cabral-Familiar, M., Hernández-Posadas, A., & Flores-Plata, L. A. (2026). Can the Difficulties in Emotion Regulation Scale be even shorter? Development of the DERS-7. Interacciones, 12, e512. https://doi.org/10.24016/2026.v12.512

 

 

ABSTRACT

Background: A 15-item version of the Difficulties in Emotion Regulation Scale (DERS-15) has previously been tested in Latin America. However, its factor structure remains unclear. Previous evidence suggests that, after removing the Awareness item, the scale is essentially unidimensional.

Objective: This study aimed to shorten the DERS-15 and develop a clearly unidimensional version that includes items from all the original dimensions of the scale.

Methods: Two Mexican samples seeking psychological care were analyzed (n₁ = 1,383; n₂ = 2,464). A series of exploratory factor analyses was conducted to develop a shorter version of the scale. The resulting 7-item version, referred to as the DERS-7, was then evaluated through confirmatory factor analysis, reliability estimation, measurement invariance testing by sex, and associations with related variables.

Results: The DERS-7 showed excellent model fit in the confirmatory phase (CFI = .99, RMSEA = .06) and adequate internal consistency reliability (ω = .88). The scale was invariant between sexes, and its associations with other variables were very similar to those observed for the longer version.

Conclusions: The DERS-7 appears to be a promising alternative for measuring emotional dysregulation when a global score is desired and administering longer versions of the scale is not feasible.

Keywords: Emotion regulation, emotional dysregulation, validation study, factor analysis, Mexico.

 

 

INTRODUCTION

Emotions comprise three interrelated elements: cognitive, physiological, and behavioral. A change in one can potentially lead to changes in the others (Reyes & Tena, 2016). Gross and Thompson (2007) define emotional regulation as the process by which individuals influence the emotions (positive and negative) experienced by themselves and others.

The contemporary functionalist perspective stresses the important role of emotions in priming necessary behavioral responses, fine-tuning decision-making, enhancing the memory of important events, and facilitating interpersonal interactions (Beauchaine & Crowell, 2020). However, they could also be detrimental when emotional responses occur at the wrong intensity and time. This phenomenon is known as emotional dysregulation, which is defined as the lack of ability to regulate emotional experiences, behavioral responses, and expressions (verbal and nonverbal) in the presence of an emotional stimulus. It is often the result of emotional vulnerability and the inadequate selection of modulation strategies, leading to maladaptive responses (Linehan et al., 2007). These maladaptive emotional responses are present in various forms of psychopathologies, social difficulties, and mental disorders (Gross & Thompson, 2007).

The Difficulties in Emotion Regulation Scale (DERS) (Gratz & Roemer, 2004), has strong empirical support (Hallion et al., 2018; Kämpf et al., 2023). The objective of this self-report scale is to evaluate emotional difficulties through six factors: Non-acceptance, Goals, Impulse, Awareness, Strategies, and Clarity. One of the main characteristics of the DERS is its adaptability to different contexts and languages. In Colombia, Muñoz-Martínez et al. (2016), based on an exploratory factor analysis, identified two dimensions instead of six, resulting in a reduction of items from 36 to 15. The factors were organized in the following manner: Factor 1 (composed of the subdimensions Non-acceptance, Goals, Impulsivity, Strategies, and Clarity) and Factor 2 (composed of the dimension Awareness).

Other brief versions, such as the DERS-16 (Bjureberg et al., 2016), the DERS-SF (Kaufman et al., 2016), and the DERS-18 (Victor & Klonsky, 2016), were developed primarily in English-speaking clinical and community samples, typically retaining multi-factor structures by selecting top-performing items from each dimension. More recently, an analysis of the DERS-18 in Peruvian university students suggested that the scale is essentially unidimensional when Awareness items—which consistently show poor factor loadings—are removed (Blancas-Guillen et al., 2024). Despite the availability of these alternatives, we chose to focus on refining the DERS-15 due to its linguistic and cultural proximity to the Mexican context, and because it was already established as a baseline assessment in ongoing national telepsychological intervention trials. One challenge with this version (DERS-15) is that, despite a previous study in the Mexican adult population, the results on its factorial structure were inconclusive (De la Rosa-Gómez et al., 2021). Regarding the Awareness dimension specifically, previous studies have consistently found that its items tend to load poorly on the general dysregulation factor and often behave as separate, weakly related component. This pattern has been attributed to the distinct conceptual nature of Awareness—which refers to the ability to attend to and acknowledge one’s emotions—as opposed to the other dimensions that capture maladaptive responses to emotional distress (Bardeen et al., 2012; Hallion et al., 2018; Muñoz-Martínez et al., 2016). As a result, the Awareness subscale has been treated as psychometrically problematic and excluded in several validated short versions of the scale. However, previous research (Bardeen et al., 2012; Hallion et al., 2018) suggests that a DERS-14 (i.e., the DERS-15 without the Awareness item) demonstrates essential unidimensionality, implying a general factor and possible residual factors. Thus, it would be feasible to construct a reduced version that adequately measures the general dysregulation factor and can be explained by a single dimension, without residual items (i.e., a strictly unidimensional scale) (Reise et al., 2013; Slocum-Gori et al., 2009).

Furthermore, it is often assumed that psychometric instruments assess equivalent constructs across women and men. As a result, invariance testing is infrequently conducted, allowing potential biases in research outcomes to remain undetected (Steyn & de Bruin, 2020). Therefore, evaluating measurement invariance is essential to ensure that observed differences reflect true variation in the underlying construct rather than measurement bias. In the case of the DERS, evidence suggests the presence of strict measurement invariance (Gómez-Simón et al., 2014; Gouveia et al., 2022; Muñoz-Troncoso et al., 2024).

In view of the above, we examined DERS-15 and, based on it, developed a briefer version that measured a single dimension of emotional dysregulation. Specifically, (a) the dimensionality of the instrument was examined; (b) a final brief version was evaluated in terms of its factor structure and psychometric properties; (c) measurement invariance of this brief version was analyzed in relation to sex; and (d) validity evidence based on relations to anxiety and depression was obtained.

 

METHODS

Design

The present study was instrumental, as its purpose was to examine the psychometric properties of a measure and to generate a brief version of it (Ato et al., 2013).

 

Participants

Sample 1 (n = 1,383) had a mean age of 31.6 years (SD = 9.94), ranging from 18 to 76 years, and Sample 2 (n = 2,464) had a mean age of 32.4 years (SD = 9.98), ranging from 18 to 83 years (for complete sociodemographic characteristics, see Table 1).

 

 

Table 1. Sociodemographic characteristics of the samples.

Characteristic

Sample 1

Sample 2

 

(n = 1,383)

(n = 2,464)

Sex

 

 

  Women

82.4%

79.1%

  Men

17.6%

20.9%

Marital Status

 

 

  Single

53.3%

57.8%

  Cohabiting

14.6%

14.2%

  Married

19.2%

15.6%

  Divorced/Separated

9.2%

10.9%

  Other

3.7%

1.5%

Educational level

 

 

  High school

21.4%

32.0%

  Undergraduate

58.7%

50.8%

  Postgraduate

10.1%

13.2%

  Other

9.8%

4.0%

Occupation

 

 

  Students

27.3%

26.6%

  Employed

48.1%

48.7%

  Unemployed

9.9%

14.0%

  Homemakers

9.4%

8.9%

  Other

5.4%

1.8%

Place of residence

 

 

  Mexico City

30.6%

27.7%

  State of Mexico

33.6%

22.8%

  Other regions

35.8%

49.5%

 

 

Instruments

Difficulties in Emotion Regulation Scale-15 (DERS-15; Muñoz-Martínez et al., 2016). The DERS-15 is an instrument validated in Colombia, adapted from the original DERS-36 (Gratz & Roemer, 2004). It comprises 15 items grouped into two factors, with response options ranging from 1 (almost never) to 5 (almost always). The first factor encompasses items that in DERS-36 corresponded to the subscales of Non-acceptance, Goals, Impulsivity, Strategies, and Clarity. Specifically, items 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, and 14 belong to this factor. Meanwhile, the second factor corresponds to the Awareness dimension and is represented by a single item (item 15; i.e., "I am attentive to my feelings"), which is reverse coded prior to scoring. Scores for each factor are computed as the sum of the corresponding items, with higher total scores indicating greater difficulties in emotion regulation. According to the study by Muñoz-Martínez et al. (2016), it demonstrated good reliability (α = .90), and evidence of its factorial validity in Mexican adults has also been reported (De la Rosa-Gómez et al., 2021). In the Results section, we will provide a detailed analysis of the psychometric properties of this measure in the present sample.

Beck Depression Inventory II (BDI-II; Beck et al., 1996). It measures depression through 21 items. Individuals can rate their responses on a scale of 0 to 3, resulting in total scores ranging from 0 to 63. The BDI-II has shown strong psychometric properties in Mexican populations. Previous studies reported adequate reliability and validity across clinical and non-clinical samples (González et al., 2015). More recent research has provided further support, demonstrating high internal consistency (alpha greater than .90) as well as adequate sensitivity and specificity in clinical samples (Rodríguez-Pérez et al., 2021). In the current study, reliability was excellent (α = .92).

Beck Anxiety Inventory (BAI; Beck et al., 1988). This inventory was developed to evaluate the severity of anxious symptoms. It is a self-administered scale with 21 items. Each item is rated on a 4-point scale (0 = not at all, 3 = severely). The BAI has demonstrated adequate psychometric properties in Mexican populations. Early validation studies reported good internal consistency and convergent validity across student, clinical, and community samples (Robles et al., 2001). More recent evidence has confirmed its robustness, showing high internal consistency (alpha greater than .90) and adequate construct validity, although some variability in its factorial structure has been reported (Padrós Blázquez et al., 2020). In the current study, its reliability was excellent (α = .93).

 

Procedure

Data collection for Sample 1 was conducted within the framework of a telepsychological intervention study (De la Rosa-Gómez et al., 2023). The DERS-15 was administered through a SurveyMonkey form as a baseline measure prior to the assignment of users to the experimental groups. Participants were recruited through social networks and institutional portals. Data for Sample 2 were collected in a similar manner from another study that sought to test two online psychological intervention modalities (De la Rosa-Gómez et al., 2022). The dissemination process was conducted similarly to Sample 1, and data were obtained through an online platform designed for the project, which required user registration.

 

Analysis plan

First, an exploratory factor analysis (EFA) based on polychoric correlations was conducted in Sample 1. The number of dimensions was determined by an optimized parallel analysis (Timmerman & Lorenzo-Seva, 2011). We used the MORGANA method, an adaptation of the unweighted least squares estimator that accounts for correlated residuals in an exploratory framework (Ferrando et al., 2022). Subsequently, we identified the two items with the highest residual correlation and selected only one item (the one with the highest factor loading) for the next round of analyses. This procedure was repeated until (a) the residual correlation identified could not be interpreted in a straightforward manner, or (b) it fell below an absolute value of .30. The .30 threshold was selected based on the criteria suggested by Ferrando et al. (2022), who classify residual correlations between .20 and .30 as low doublets, and values above .30 as medium or high doublets. Consequently, this cutoff was used to systematically identify and remove items with significant redundant variance not captured by the general factor. Special care was taken to include items from all the original DERS dimensions.

Once a clear factor structure was established, we tested in a new dataset (Sample 2). In this phase, we conducted a confirmatory factor analysis with the WLSMV estimator. Model fit was examined with approximate indices: comparative fit index (CFI), Tucker-Lewis index (TLI), root-mean-square error of approximation (RMSEA), and standardized root-mean-square residual (SRMR). Following Hu and Bentler's (1999) guidelines, the following values were considered as evidence of good fit: CFI > .95, TLI > .95, RMSEA < .06, SRMR < .08. If the model had an acceptable fit, we proceeded to estimate internal consistency reliability using Green and Yang's (2009) categorical omega. Next, measurement invariance was also examined. As recommended for non-linear models, threshold invariance was tested, followed by threshold and loadings invariance (Temme, 2006), using Wu and Estabrook's (2016) method. Finally, associative evidence of validity was obtained from Sample 2. Pearson correlations were calculated between the DERS scores and measures of depression (BDI-II) and anxiety (BAI). Additionally, we estimated the correlation (r) between the longer and shorter forms, and subsequently, this value was later corrected (r′) for spurious correlation (Levy, 1967).

The EFA was conducted using FACTOR (version 12.04.01). For CFA and measurement invariance, we used the packages lavaan (version 0.6-16) and semTools (version 0.5-6) implemented in R (version 4.3.0).

 

Ethical aspects

The study was approved by the Research Ethics Committee of the Facultad de Estudios Superiores Iztacala of the Universidad Nacional Autónoma de México (CE/FESI/082020/1363). Due to the nature of the primary intervention studies, the data was not anonymized; however, access to the databases was restricted to the principal investigator and two assistants. Anonymized versions of the data were created for research purposes. In both samples, informed consent was requested prior to data collection.

 

RESULTS

Item-Level Descriptive Statistics

Table 2 presents descriptive statistics of the DERS-15 items in Sample 1. Notably, the majority of items exhibit means around 3 (i.e. the middle option). The lowest value was observed in item 13 (When I’m upset, I lose control over my behaviors), while the largest one was that of item 7 (When I’m upset, I have difficulty focusing on other things). Standard deviations were similar across all items. Additionally, skewness values were all reasonably close to zero. On the other hand, most kurtosis values were negative and fell below -1, suggesting that the items had a platykurtic distribution. Table 2 provides the percentages of response options.

 

 

Table 2. Item-Level Descriptive Statistics of the DERS-15 (n = 1383).

Item

M

SD

g₁

g₂

Responses to Each Option (%)

 

 

 

 

 

1

2

3

4

5

1. Tengo dificultad para encontrar el significado a mis sentimientos [I have difficulty making sense out of my feelings].

3.06

1.22

0.05

-1.11

9

31

19

27

14

2. Estoy confundido(a) acerca de cómo me siento [I am confused about how I feel].

3.15

1.23

-0.03

-1.17

8

30

17

29

16

3. Cuando estoy alterado(a), tengo dificultad para realizar el trabajo [When I’m upset, I have difficulty getting work done].

3.41

1.24

-0.26

-1.13

6

24

16

30

23

4. Cuando estoy molesto(a), quedo fuera de control [When I’m upset, I become out of control].

2.57

1.38

0.49

-1.06

27

31

13

16

13

5. Cuando estoy alterado(a), creo que seguirá siendo así durante mucho tiempo [When I’m upset, I believe that I will remain that way for a long time].

2.73

1.34

0.29

-1.17

21

31

14

21

13

6. Cuando estoy alterado(a), creo que voy a terminar sintiéndome muy deprimido [When I’m upset, I believe that I’ll end up feeling very depressed].

3.04

1.42

-0.01

-1.39

17

26

12

23

21

7. Cuando estoy alterado(a), tengo dificultad para concentrarme en otras cosas [When I’m upset, I have difficulty focusing on other things].

3.61

1.25

-0.45

-1.08

4

22

13

30

31

8. Cuando estoy alterado(a), me siento fuera de control [When I’m upset, I feel out of control].

2.74

1.40

0.32

-1.21

23

30

14

17

16

9. Cuando estoy alterado(a), me siento avergonzado de mí mismo por sentir de esa manera [When I’m upset, I feel ashamed with myself for feeling that way].

3.00

1.44

0.06

-1.42

18

27

11

21

22

10. Cuando estoy alterado(a), yo tengo dificultades concentrándome [When I’m upset, I have difficulty concentrating].

3.47

1.30

-0.34

-1.18

7

23

14

28

28

11. Cuando estoy alterado(a), tengo dificultades controlando mis comportamientos [When I’m upset, I have difficulty controlling my behaviors].

2.75

1.36

0.34

-1.16

20

32

15

18

15

12. Cuando estoy alterado(a), creo que no hay nada que pueda hacer para sentirme mejor [When I’m upset, I believe that there is nothing I can do to make myself feel better].

2.88

1.37

0.21

-1.25

17

31

15

20

17

13. Cuando estoy alterado(a), pierdo el control sobre mis conductas [When I’m upset, I lose control over my behaviours].

2.38

1.36

0.68

-0.83

34

30

11

14

11

14. Cuando estoy alterado(a), encuentro difícil pensar en algo más [When I’m upset, I have difficulty thinking about anything else].

3.33

1.31

-0.19

-1.25

8

26

14

27

24

15. Yo estoy atento a mis sentimientos [I am attentive to my feelings].

2.97

1.24

0.10

-1.08

12

30

20

25

13

Note. g₁ = skewness. g₂ = kurtosis (zero centered). Response options are as follows: 1 = almost never, 2 = sometimes, 3 = about half of the time, 4 = most of the time, and 5 = almost always.

 

 

Exploratory Factor Analysis

Before conducting the EFA, the adequacy of the correlation matrix was evaluated. The Kaiser-Meyer-Olkin (KMO) measure was .92, and Bartlett’s test of sphericity was significant, χ²(105) = 15843.7, p < .001, indicating that the data were suitable for factor analysis. The parallel analysis suggested the extraction of a single factor. When examining the factor solution, we observed that item 15 (corresponding to the original Awareness dimension) had a factor loading close to zero (λ = .01); therefore, we decided to discard it. In the next round of analysis, a high residual correlation was found between items 1 and 2 (φ = .70, 95% CI: [.66, .75]). Subsequent item eliminations were performed to address redundancy within the original DERS dimensions. Based on the factor loadings, we decided to retain item 1 and discard item 2. In the next round of analysis, with 13 items, a high correlation was found between the residuals of items 7 and 10 (φ = .70, 95% CI: [.64, .76]), both representing the Goals dimension. Consequently, item 10 was eliminated, and a new analysis was performed. This pattern of redundancy within dimensions continued: a high residual correlation was found for items 11 and 13 (φ = .63, 95% CI: [.60, .71]). In this case, we decided to drop item 13 and redo the analysis. Once more, a high residual correlation was observed, this time between items 4 and 8 (φ = .58, 95% CI: [.53, .63]); item 8 was retained for the next analysis. The reanalysis, which included 8 items, showed a moderate correlation between the residuals of items 7 and 14 (φ = .34, 95% CI: [.30, .45]). Following the removal of item 14, a final analysis was performed with the remaining 7 items, however, the residual correlations identified did not meet our interpretability criteria, so this was considered the final model. Table 3 provides the polychoric correlations on which the described factor analysis was based.

 

 

Table 3. Inter-item polychoric correlations of the DERS-15 (n = 1383).

Item

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

1

—

2

0.79

—

3

0.45

0.48

—

4

0.38

0.37

0.50

—

5

0.44

0.44

0.55

0.65

—

6

0.48

0.49

0.55

0.52

0.66

—

7

0.45

0.43

0.75

0.54

0.59

0.65

—

8

0.41

0.38

0.54

0.81

0.68

0.58

0.64

—

9

0.43

0.42

0.50

0.48

0.56

0.59

0.56

0.60

—

10

0.42

0.41

0.74

0.50

0.57

0.56

0.87

0.61

0.61

—

11

0.42

0.39

0.50

0.75

0.63

0.52

0.54

0.78

0.57

0.56

—

12

0.52

0.48

0.58

0.56

0.69

0.67

0.60

0.65

0.60

0.60

0.66

—

13

0.41

0.37

0.46

0.79

0.62

0.50

0.50

0.78

0.54

0.50

0.86

0.65

—

14

0.38

0.37

0.61

0.51

0.61

0.58

0.73

0.59

0.56

0.69

0.55

0.69

0.58

—

15

-0.21

-0.15

0.04

-0.01

0.03

0.10

0.09

0.00

0.05

0.06

-0.03

0.01

-0.06

0.09

—

 

 

Confirmatory Factor Analysis and Internal Consistency Reliability

The confirmatory factor analysis performed on Sample 2 (n = 2464) showed an adequate fit of the previously obtained 7-item model (henceforth called DERS-7): χ²(14) = 140.67, p < .001, CFI = .99, TLI = .99, RMSEA = .06, SRMR = .02. As shown in Figure 1, the factor loadings were between .56 and .83. Likewise, the internal consistency reliability was adequate (α = .87, ωcategorical = .88).

 

 

Figure 1. Factor Structure of the DERS-7

Note. Residuals, thresholds, and intercepts are not presented to ease interpretation.

 

 

Measurement Invariance

As shown in Table 4, both thresholds and factor loadings were equivalent between men and women. Therefore, it is possible to make valid comparisons between sexes using the DERS-7.

 

 

Table 4. Measurement invariance of the DERS-7 by sex (n = 2464).

Model

χ²

df

p

CFI

TLI

RMSEA

Δχ²

Δdf

p

1. Configural (baseline)

162.92

28

<.001

0.99

0.99

0.06

2. Equal thresholds

178.17

42

<.001

0.99

0.99

0.05

11.92

14

0.612

3. Equal thresholds and loadings

155.77

48

<.001

0.99

1.00

0.04

6.30

6

0.390

Note. Wu and Estabrook's (2016) approach was followed.

 

 

Associative Validity Evidence

Both the DERS-7 and the DERS-14 (i.e., DERS-15 after removing the Awareness item) were examined in relation to anxious and depressive symptomatology. For anxiety, the correlations of the DERS-7 (r = .54, 95% CI: [.51, .56]) were virtually identical to those of the DERS-14 (r = .54, 95% CI: [.51, .57]). Similarly, correlations with depression were very similar to the DERS-7 (r = .63, 95% CI: [.61, .66]) and the DERS-14 (r = .63, 95% CI: [.60, .65]). Finally, the association between DERS-7 and DERS-14 was also calculated by applying the correction for spurious correlation. The corrected value indicated a strong association between both versions (r′ = .90, 95% CI: [.90, .91]).

 

 

DISCUSSION

The present study examined the structure of the factor of a short version of the DERS: the DERS-15. Results indicated that, apart from the Awareness item, it was essentially unidimensional, however, it exhibited some residual correlations. Based on this result, we were able to construct a shorter 7-item version that retained the original scale’s psychometric properties while delivering a clearer factor solution with no residual correlations. Importantly, the final DERS-7 comprises items representing the original domains of Non-acceptance, Goals, Impulse, Strategies, and Clarity, thus preserving the conceptual scope of emotional dysregulation while achieving a strictly unidimensional structure.

In the original study in which DERS-15 was proposed, researchers identified two components that best explained the variance of the data. One of these components assessed Dysregulation, while the other (composed of only one item) measured Awareness (Muñoz-Martínez et al., 2016). However, this structure did not obtain a good fit when it was subsequently tested in the Mexican population (De la Rosa-Gómez et al., 2021). The data from the present study allows us to understand the reason for this discrepancy. The Dysregulation subscale does, in fact, measure a global dimension but also exhibits a set of residual correlations. In other words, this subscale does not present strict, but essential unidimensionality (Reise et al., 2013; Slocum-Gori et al., 2009). To address this issue, we propose a reduced version (DERS-7), which not only shortens the scale but also maintains items representing all the original dimensions of the DERS (except for Awareness). The Awareness dimension was excluded given its low factor loadings and weak association with the general dysregulation factor, likely due to its distinct conceptual emphasis on emotional attention rather than dysregulatory responses, thereby improving factorial coherence.

The finding that the DERS is an essentially unidimensional measure is consistent with those reported in other studies. For example, Hallion et al. (2018) identified a bifactor structure in the DERS and various short versions, which included a strong general factor and a set of residual factors (excluding Awareness items). Similarly, the study by Mekawi et al. (2021) also found support for a bifactor model, in which the general dysregulation factor was stronger than the specific ones, after excluding Awareness. Finally, Blancas-Guillen et al. (2024) reached a similar conclusion, and recommended researchers calculate only an overall DERS score (excluding Awareness items), emphasizing the limited utility of the specific dimensions. Notably, all these studies underscore the necessity of excluding the Awareness items when calculating an overall score, thus justifying their elimination in our proposal. In this context, although bifactor models provide a useful representation of the hierarchical structure of the DERS, the consistent dominance of the general factor supports the use of more parsimonious unidimensional approaches when the goal is to obtain a global score.  Accordingly, DERS-7 can be treated as a unidimensional measure of emotional dysregulation for scoring and interpretation.

The item reduction process was guided by both statistical and theoretical criteria. Statistically, items were removed when they exhibited high residual correlations, indicating redundancy beyond the general factor. In each case, the item with the highest factor loading was retained. Theoretically, special care was taken to preserve items that captured the core features of emotional dysregulation, ensuring that the resulting scale-maintained content validity despite being brief. In this context, although bifactor models provide a useful representation of the hierarchical structure of the DERS, the consistent dominance of the general factor supports the use of more parsimonious unidimensional approaches when the goal is to obtain a global score. While this approach offers a more efficient and psychometrically clean global estimate for screening, its primary disadvantage is the loss of face-to-level granularity. Consequently, DERS-7 prioritizes the assessment of general maladaptive reactivity over the ability to monitor emotional states or identify specific regulatory profiles.

 

Limitations

The present study has several limitations that should be acknowledged. First, the predominance of women, constituting approximately 80% of both samples. This imbalance may limit the generalizability of the findings to men, particularly given evidence that emotional regulation processes and their expression can vary by gender. Second, all the data collected relied on self-report measures, we were unable to examine whether the information extrapolated to the level of directly observable behaviors. Additionally, the level of education in both samples, where most participants had university or graduate degrees, may not accurately represent the Mexican educational reality. Finally, the absence of a specific emotion regulation measure prevented a direct assessment of convergent validity. Despite these limitations, the present study does offer several strengths, such as the assessment of individuals seeking psychological help, the use of two study samples (one for construction and the other for replication), as well as substantial sample sizes.

 

Conclusion

In the present study, a 15-item version of the DERS, previously developed in the Colombian population, was used to generate an even shorter version of the instrument. After excluding the Awareness dimension item, the DERS-15 was reduced to a briefer 7-item version that maintained items from the original five dimensions and presented a unidimensional structure. This brief version can prove to be valuable in contexts where time constraints or respondent fatigue, and a global estimate of emotional dysregulation is needed using only a few items. Future studies should test the DERS-7 in other populations and examine its performance.

 

ORCID

Pablo D. Valencia: https://orcid.org/0000-0002-6809-1805

Anabel De la Rosa-Gómez: https://orcid.org/0000-0002-3527-1500

Mariana Cabral-Familiar: https://orcid.org/0009-0007-1143-7303

Alejandrina Hernández-Posadas: https://orcid.org/0000-0001-5753-9785

Lorena A. Flores-Plata: https://orcid.org/0000-0003-1306-0718

 

AUTHORS’ CONTRIBUTION

Pablo D. Valencia: Conceptualization, Data Curation, Formal Analysis, Writing - Review & Editing.

Anabel De la Rosa-Gómez: Conceptualization, Writing - Review & Editing, Supervision, Project administration, Funding acquisition.

Mariana Cabral-Familiar: Writing - Original Draft.

Alejandrina Hernández-Posadas: Writing - Original Draft.

Lorena A. Flores-Plata: Writing - Original Draft.

 

FUNDING SOURCE

This work was supported by both the UNAM-PAPIIT Project (IT300721) and the former Mexican Consejo Nacional de Humanidades, Ciencia y Tecnología (CONAHCyT; Conv. 2020-04: Proyectos de Investigación e Incidencia Social en Salud Mental y Adicciones. #1401). The funding institutions had no role in the design of the study or in the collection, analysis, and interpretation of the data, and had no role in the writing of the manuscript.

 

CONFLICT OF INTEREST

The authors declare that they have no conflicts of interest.

 

ACKNOWLEDGMENTS

Not applicable.

 

REVIEW PROCESS

This study has been reviewed by Andrei Franco-Jimenez, and other reviewers in double-blind mode. The editor in charge was Renzo Rivera. The review process is included as supplementary material 1.

 

DATA AVAILABILITY STATEMENT

Available in: https://osf.io/83jds/

 

DECLARATION OF THE USE OF GENERATIVE ARTIFICIAL INTELLIGENCE

The authors used the AI tool Gemini exclusively for proofreading assistance. Final responsibility for the conclusions and text of this manuscript rests entirely with the authors.

 

DISCLAIMER

The authors are responsible for all statements made in this article.

 

REFERENCES

Ato, M., López-García, J. J., & Benavente, A. (2013). A classification system for research designs in psychology. Anales de Psicología / Annals of Psychology, 29(3), 1038–1059. https://doi.org/10.6018/analesps.29.3.178511

Bardeen, J. R., Fergus, T. A., & Orcutt, H. K. (2012). An examination of the latent structure of the Difficulties in Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 34(3), 382–392. https://doi.org/10.1007/s10862-012-9280-y

Beauchaine, T. P., & Crowell, S. E. (Eds.). (2020). The Oxford Handbook of Emotion Dysregulation (1st ed.). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780190689285.001.0001

Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56(6), 893–897. https://doi.org/10.1037/0022-006X.56.6.893

Beck, A. T., Steer, R. A., Ball, R., & Ranieri, W. F. (1996). Comparison of Beck Depression Inventories-IA and-II in psychiatric outpatients. Journal of Personality Assessment, 67(3), 588–597. https://doi.org/10.1207/s15327752jpa6703_13

Blancas-Guillen, J., Arroyo-Pizarro, P., Quintana, E., Tupa-Belizario, R. S., & Valencia, P. D. (2024). Escala de Dificultades en la Regulación Emocional: Análisis psicométrico de sus versiones breves en universitarios peruanos [Difficulties in Emotion Regulation Scale: A psychometric analysis of its brief versions in Peruvian college students]. Revista Argentina de Ciencias Del Comportamiento, 16(2), 60–73. https://doi.org/10.32348/1852.4206.v16.n2.36158

Bjureberg, J., Ljótsson, B., Tull, M. T., Hedman, E., Sahlin, H., Lundh, L. G., ... & Gratz, K. L. (2016). Development and validation of a brief version of the Difficulties in Emotion Regulation Scale: The DERS-16. Journal of Psychopathology and Behavioral Assessment, 38, 284-296. https://doi.org/10.1007/s10862-015-9514-x

De la Rosa-Gómez, A., Flores-Plata, L. A., Esquivel-Santoveña, E. E., Santillán Torres Torija, C., García-Flores, R., Dominguez-Rodriguez, A., Arenas-Landgrave, P., Castellanos-Vargas, R. O., Berra-Ruiz, E., Silvestre-Ramírez, R., Miranda-Díaz, G. A., Díaz-Sosa, D. M., Hernández-Posadas, A., Flores-Elvira, A. I., Valencia, P. D., & Vázquez-Sánchez, M. F. (2022). Efficacy of a transdiagnostic guided internet-delivered intervention for emotional, trauma and stress-related disorders in Mexican population: Study protocol for a randomized controlled trial. BMC Psychiatry, 22(1), 537. https://doi.org/10.1186/s12888-022-04132-6

De la Rosa-Gómez, A., Hernández Posadas, A., Valencia, P. D., Flores-Plata, L. A., Serrano Zárate, B., Flores Elvira, A. I., Dominguez-Rodriguez, A., Vázquez Sánchez, M. F., & González Santiago, E. (2023). Online transdiagnostic intervention for emotional, trauma- and stressor-related disorders in the Mexican population: A randomized pilot and feasibility trial. Contemporary Clinical Trials Communications, 35, 101204. https://doi.org/10.1016/j.conctc.2023.101204

De la Rosa-Gómez, A., Hernández-Posadas, A., Valencia, P. D., & Guajardo-Garcini, D. A. (2021). Análisis dimensional de la Escala de Dificultades en la Regulación Emocional (DERS-15) en universitarios mexicanos [Dimensional analysis of the Difficulties in Emotion Regulation Scale (DERS-15) in Mexican university students]. Revista Evaluar, 21(2), 80–97. https://doi.org/10.35670/1667-4545.v21.n2.34401

Ferrando, P. J., Hernandez-Dorado, A., & Lorenzo-Seva, U. (2022). Detecting correlated residuals in exploratory factor analysis: New proposals and a comparison of procedures. Structural Equation Modeling: A Multidisciplinary Journal, 29(4), 630–638. https://doi.org/10.1080/10705511.2021.2004543

González, D. A., Reséndiz Rodríguez, A., & Reyes-Lagunes, I. (2015). Adaptation of the BDI–II in Mexico. Salud mental, 38(4), 237–244. https://doi.org/10.17711/SM.0185-3325.2015.033

Gómez-Simón, I., Penelo, E., & De La Osa, N. (2014). Factor structure and measurement invariance of the Difficulties Emotion Regulation Scale (DERS) in Spanish adolescents. Psicothema, 26(3), 401-408. https://doi.org/10.7334/psicothema2013.324

Gouveia, P., Ramos, C., Brito, J., Almeida, T. C., & Cardoso, J. (2022). The Difficulties in Emotion Regulation Scale–Short Form (DERS-SF): psychometric properties and invariance between genders. Psicologia: Reflexão e Crítica, 35(1), 11. https://doi.org/10.1186/s41155-022-00225-z

Gratz, K. L., & Roemer, L. (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(1), 41–54. https://doi.org/10.1023/B:JOBA.0000007455.08539.94

Green, S. B., & Yang, Y. (2009). Reliability of summed item scores using structural equation modeling: An alternative to coefficient alpha. Psychometrika, 74(1), 155–167. https://doi.org/10.1007/s11336-008-9099-3

Gross, J. J., & Thompson, R. A. (2007). Emotion regulation: Conceptual foundations. In J. J. Gross (Ed.), Handbook of emotion regulation. (pp. 3–24). The Guilford Press.

Hallion, L. S., Steinman, S. A., Tolin, D. F., & Diefenbach, G. J. (2018). Psychometric properties of the Difficulties in Emotion Regulation Scale (DERS) and its short forms in adults with emotional disorders. Frontiers in Psychology, 9, 539. https://doi.org/10.3389/fpsyg.2018.00539

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

Kämpf, M. S., Adam, L., Rohr, M. K., Exner, C., & Wieck, C. (2023). A Meta-Analysis of the relationship between emotion regulation and social affect and cognition. Clinical Psychological Science, 11(6), 1159–1189. https://doi.org/10.1177/21677026221149953

Kaufman, E. A., Xia, M., Fosco, G., Yaptangco, M., Waters, T. E., & Crowell, S. E. (2016). The Difficulties in Emotion Regulation Scale Short Form (DERS-SF): Validation and replication in adolescent and adult samples. Journal of Psychopathology and Behavioral Assessment, 38(3), 443-455. https://doi.org/10.1007/s10862-015-9529-3

Levy, P. (1967). The correction for spurious correlation in the evaluation of short-form tests. Journal of Clinical Psychology, 23(1), 84–86. https://doi.org/10.1002/1097-4679(196701)23:1%3C84::AID-JCLP2270230123%3E3.0.CO;2-2

Linehan, M. M., Bohus, M., & Lynch, T. R. (2007). Dialectical behavior therapy for pervasive emotion dysregulation: Theoretical and practical underpinnings. In J. J. Gross (Ed.), Handbook of emotion regulation. (pp. 581–605). The Guilford Press.

Mekawi, Y., Watson‐Singleton, N. N., Dixon, H. D., Fani, N., Michopoulos, V., & Powers, A. (2021). Validation of the Difficulties with Emotion Regulation Scale in a sample of trauma‐exposed Black women. Journal of Clinical Psychology, 77(3), 587–606. https://doi.org/10.1002/jclp.23036

Muñoz-Martínez, A. M., Vargas, R. M., & Hoyos-González, J. S. (2016). Escala de Dificultades en Regulación Emocional (DERS): Análisis factorial en una muestra colombiana [Difficulties in Emotion Regulation Scale (DERS): Factor analysis in a Colombian sample]. Acta Colombiana de Psicología, 19(1), 225–236. https://doi.org/10.14718/ACP.2016.19.1.10

Muñoz-Troncoso, F., Riquelme-Mella, E., Halberstadt, A. G., Montero, I., Sepúlveda-Bernales, V., Fuentes-Vilugrón, G., Miranda-Zapata, E., Legaz-Vladímisrkaya, E., Caamaño-Navarrete, F., & Muñoz-Troncoso, G. (2024). Emotional Regulation Challenges in Chilean Teachers: An analysis of the measurement invariance of the DERS-E and the influence of gender and age. Journal of Intelligence, 12(9), 86. https://doi.org/10.3390/jintelligence12090086

Padrós Blázquez, F., Montoya Pérez, K. S., Bravo Calderón, M. A., & Martínez Medina, M. P. (2020). Propiedades psicométricas del Inventario de Ansiedad de Beck (BAI, Beck Anxiety Inventory) en población general de México [Psychometric properties of the Beck Anxiety Inventory (BAI) in the general population of Mexico]. Ansiedad y Estrés, 26(2–3), 181–187. https://doi.org/10.1016/j.anyes.2020.08.002

Reise, S. P., Bonifay, W. E., & Haviland, M. G. (2013). Scoring and modeling psychological measures in the presence of multidimensionality. Journal of Personality Assessment, 95(2), 129–140. https://doi.org/10.1080/00223891.2012.725437

Reyes, M. A., & Tena, E. A. (2016). Regulación emocional en la práctica clínica [Emotion regulation in clinical practice]. Manual Moderno.

Robles, R., Varela, R., Jurado, S., & Páez, F. (2001). Versión mexicana del Inventario de Ansiedad de Beck: Propiedades psicométricas [The Mexican version of Beck Anxiety Inventory: Psychometric properties]. Revista Mexicana de Psicología, 18(2), 211–218.

Rodríguez-Pérez, V., García-Flores, R., Miranda-Díaz, G. A., & García-Hernández, O. (2021). Beck Depression Inventory (BDI-IA) adapted for HIV: Psychometric properties. Salud Mental, 44(6), 287–294. https://doi.org/10.17711/sm.0185-3325.2021.037

Slocum-Gori, S. L., Zumbo, B. D., Michalos, A. C., & Diener, E. (2009). A note on the dimensionality of quality of life scales: An illustration with the Satisfaction with Life Scale (SWLS). Social Indicators Research, 92(3), 489–496. https://doi.org/10.1007/s11205-008-9303-y

Steyn, R., & De Bruin, G. P. (2020). An investigation of gender-based differences in assessment instruments: A test of measurement invariance. SA Journal of Industrial Psychology, 46(1), 1-12. https://doi.org/10.4102/sajip.v46i0.1699

Temme, D. (2006). Assessing measurement invariance of ordinal indicators in cross-national research. In S. Diehl & R. Terlutter (Eds.), International Advertising and Communication (pp. 455–472). DUV. https://doi.org/10.1007/3-8350-5702-2_24

Timmerman, M. E., & Lorenzo-Seva, U. (2011). Dimensionality assessment of ordered polytomous items with parallel analysis. Psychological Methods, 16(2), 209–220. https://doi.org/10.1037/a0023353

Victor, S. E., & Klonsky, E. D. (2016). Validation of a brief version of the Difficulties in Emotion Regulation Scale (DERS-18) in five samples. Journal of Psychopathology and Behavioral Assessment, 38(4), 582-589. https://doi.org/10.1007/s10862-016-9547-9

Wu, H., & Estabrook, R. (2016). Identification of confirmatory factor analysis models of different levels of invariance for ordered categorical outcomes. Psychometrika, 81(4), 1014–1045. https://doi.org/10.1007/s11336-016-9506-0