Romanian Version of the Frustration Discomfort Scale (FDS): A Preliminary Validation on a Non-clinical Sample

Low frustration tolerance represents one of the major categories of irrational beliefs within the Rational Emotive Behavior Therapy (REBT) framework (David et al., 2005). It is defined as a demand that the reality should be as we want it to be (Harrington, 2005a, b). People scoring on low frustration tolerance encounter difficulties detecting the threshold between desires and reality (Harrington, 2007). In contrast, people with good frustration tolerance are more willing to accept things as they are rather than demand that things should be different. Trip et al. (2021) have shown that low frustration tolerance uniquely predicts unruly and disruptive behaviors as specific externalizing problems.

However, the REBT literature includes various beliefs in the low frustration tolerance category, such as instant gratification, intolerance of emotions, and problem avoidance (Dryden & Gordon, 1993). In this regard, Harrington (2007) suggested that it would be helpful to approach frustration intolerance as a multidimensional construct because different cognitions related to low frustration tolerance led to different outcomes (e.g., intolerance to discomfort to depression, intolerance of emotional distress to anxiety, and intolerance of unfairness to anger).

In this regard, (Harrington, 2005b) introduced the Frustration Discomfort Scale (FDS) to measure a multidimensional frustration intolerance construct. The scale contains four facets of frustration intolerance: emotional intolerance, discomfort intolerance, achievement frustration, and entitlement. The emotional intolerance facet reflects reduced tolerance to experience emotional distress. The discomfort intolerance dimension deals with beliefs that life should be easy and free from hassle. The achievement frustration intolerance scale is linked to task-related frustration, whereas the entitlement scale refers to the demand for fairness and immediate gratification. Harrington (2005a, 2007) argued that the multidimensional FDS construct has discriminant validity. For instance, discomfort intolerance was a positive predictor of procrastination, whereas emotional intolerance and achievement frustration were negative predictors. Likewise, emotional intolerance was associated with anxiety, discomfort intolerance was associated with depression, and entitlement was associated with anger. Regarding the big-five parallelisms, emotional intolerance is closely related to neuroticism, entitlement with low agreeableness, achievement frustration and discomfort intolerance with high/low conscientiousness.

The factorial structure of the Frustration Discomfort Scale has been investigated in various research that focused on validating the scale in other languages. So far, the instrument has been translated into Spanish on an Argentinian sample (Medrano et al., 2018) and a Chilean sample (Ruiz-Ortega et al., 2021), French (Chamayou et al., 2016), Italian (Filippello et al., 2014; Tripaldi et al., 2018), Portuguese on a Brazilian sample (Silva & Faro, 2021), Serbian (Stanković & Vukosavljević-Gvozden, 2011), Turkish (Ozer et al., 2012), and Urdu (Jibeen, 2013). However, the initial four-factor correlated solution has received mixed empirical support. Most translations reached a four-factor solution, but all of them required modifications (deviations) to obtain adequate or good indices of fit (i.e., CFI > 0.90; RMSEA < 0.06). These modifications varied in form. In some cases, suggestions were made towards dropping some items (i.e., Turkish sample, Argentinian sample). In other cases, different factorial solutions were suggested (e.g., one-factor in the French sample; two-factor in the Brazilian sample). Finally, in some cases, a second-order factor above the four factors or solutions that include correlations among item residuals was suggested to obtain good fit indices. Therefore, the FDS seems to have a less reliable factorial structure across various samples and languages.

The current study aimed to validate the Romanian version of the FDS on a teachers' sample. We selected teachers for convenience, as teachers were candidates to receive a subsequent REBT intervention to decrease teachers' frustration intolerance in school settings. Relying on a sample of teachers as a non-clinical sample has the advantage of departing from the often-used student samples in the FDS validation studies and the disadvantage of not being a representative sample from the general population. We examined the factorial structure of the FDS obtained on the Romanian sample of teachers. We also investigated the relationship between low frustration tolerance and its facets on the one hand and other irrational beliefs on the other hand. First, we looked at the relationship between the FDS score and the central irrational beliefs in the REBT framework: demandingness, self-downing, low frustration tolerance, and awfulizing. This investigation would support the discriminant validity of the FDS, particularly when we compare the association between its overall score with the ABS-2 low frustration tolerance scale and the association between the same FDS overall score with the ABS-2 self-downing scale. Most theoretical accounts highlight the distinction between self-distortions (i.e., self-downing) and reality distortions (i.e., frustration intolerance) (Harrington, 2005a).

Taking advantage of our sample characteristics (i.e., teachers), we also investigated how the FDS is associated with TIBS level of irrational beliefs. In addition, we expected that the FDS score would correlate more intensively with teachers' low frustration tolerance level than with teachers' level of self-downing. We also exploratory analyzed the interplay of the FDS overall score and its associated subscales with TIBS absolutistic demands towards others. The latter concept shares more features with the FDS entitlement scale than other frustration intolerance scales, such as discomfort or emotional intolerance.

Likewise, to the extent to which the FDS successfully distinguishes between entitlement and discomfort intolerance, we expect that the teacher's absolutistic demands on others will correlate more intensively with the FDS entitlement scale than discomfort intolerance. This exploratory endeavor will support the utility of addressing frustration intolerance as a multi-component construct.

Unconditional self-acceptance (Chamberlain & Haaga, 2001) has consistently been found to correlate negatively with irrational beliefs (Davies, 2006, 2008; Samfira & Sava, 2021). Therefore, in our approach to validate the Romanian version of the FDS, we expect a negative association between unconditional self-acceptance and the overall level of frustration intolerance.

Last but not least, we also examined the relationship between the frustration intolerance level and teachers’ style of pupil control. Custodial teachers (Willower et al., 1967) impose a high demand on others and are more likely to become frustrated because students' behavior departs from their idealistic expectations. Teachers who adopt a custodial view of pupil control ideology endorse more dysfunctional beliefs, unrelenting standards, and schemas of entitlement (Samfira & Sava, 2021).

Likewise, to the extent to which the FDS successfully distinguishes between entitlement, on the one hand, and emotional intolerance and discomfort intolerance, on the other hand, we expect that a custodial pupil control ideology will correlate more intensively with the FDS entitlement and achievement frustration scales than with the emotional and the discomfort intolerance scales, to provide further support for the utility of addressing the frustration intolerance as a multi-component construct.

Descriptive values for the instruments included in this study are presented in Table 1, whereas the correlation matrix among our variables of interest is presented in Table 2.

The study aimed to validate a Romanian version of the FDS on a teacher sample. The FDS was developed as a multidimensional measure of frustration intolerance construct, as there were different facets involved, such as the urgency for immediate gratification or the intolerance of emotions. Our main aim was to assess the validity and reliability of the FDS and pay particular attention to the factorial structure of the scale, given that mixed results in this regard were obtained in previous similar studies that were conducted in other countries. Overall, both 1-factor and 4-factor solutions could adequately fit the data. However, the 1-factor solution is more parsimonious since the 4 FDS subscales need to provide additional discriminant validity. Likewise, based on the pattern of correlations with the ABS-2 scale, the most common instrument for measuring irrational beliefs, our results support the specificity perspective of the FDS in focusing on frustration intolerance mainly. The FDS correlates in the expected directions with other relevant measures. It correlates positively with TIBS (the level of irrational beliefs of teachers) and PCI (more likely to become frustrated when pupils misbehave), and it correlates negatively, to a smaller extent, with USAQ (less prone to unconditional self-acceptance). Such findings align with the general perspective of frustration intolerance as being related to a difficulty in distinguishing between reality and a wish.

On the other hand, we needed more support for the relevance of distinguishing different facets of frustration intolerance. Theoretically, it might be relevant in distinguishing, for instance, between immediate gratification needs (a typical problem for externalizing issues) and the higher level of discomfort when experiencing frustration as an emotion. Such differentiation failed to be relevant when we linked them to other constructs (i.e., different types of irrational beliefs).

The factorial structure of the Romanian version of the FDS obtained on a non-clinical sample provided good fit indices when employing an appropriate method of parameter estimation such as DWLS. Notably, all items had high factor loadings in the extracted factor(s), both for the 4-factor and the 1-factor solutions. The factorial structure of the Romanian version translation is in line with some translations of the original FDS, such as the French translation. However, unlike those instances, the factorial structure for the Romanian version based on CFA found similar support for the original FDS 4-factor solution as for the 1-factor solution.

The study presents several limitations. The most important one is that all analyses were conducted on a non-clinical sample. It skewed the entire distribution, as the average score on FDS was 2.28 on a 5-point Likert scale from 1 to 5. We hope that providing the FDS scale for the Romanian-speaking community will facilitate further research on the frustration intolerance construct and its significance in non-clinical and clinical settings. It might be the case that selecting clinical populations (e.g., narcissistic personality disorders vs. obsessive–compulsive personality disorders) would provide a different story on the utility of treating frustration intolerance as a multidimensional construct. Similarly, future studies that would include depression, anger, anxiety, or procrastination measures, as Harrington (2005a, 2007) did. Such constructs might provide more discriminant validity for treating FDS as a multidimensional construct. The type of sample (clinical vs. non-clinical) likely plays a crucial role in determining the factorial structure. In the original study (Harrington, 2005b), clinical participants represented more than 75% of the total sample. Although irrational beliefs are considered to be also present in the general population, they will be more pronounced in a clinical sample. For instance, the average total FDS score in our sample of teachers was 64.01 (SD of 23.59), whereas the average total FDS score in Harrington’s (2005b) study was 90.75 (SD of 20.70). Participants from a non-clinical sample are less likely to endorse FDS items (e.g., I can’t bear to feel I am losing my mind) than participants from a clinical sample. This response style will attenuate the chance of distinguishing among different FDS dimensions, as it will underestimate the true intercorrelations among the items. Unfortunately, almost all previous studies shared this limitation, as they were conducted on student samples such as the Italian (Filippello et al., 2014; Tripaldi et al., 2018), Turkish (Ozer et al., 2012), Serbian (Stanković & Vukosavljević-Gvozden, 2011), Brazilian (Silva & Faro, 2021), French (Chamayou et al., 2016), and Pakistani/Urdu (Jibeen, 2013), or general non-clinical populations in the Italian (Tripaldi et al., 2018), and Chilean studies (Ruiz-Ortega et al., 2021). Therefore, it is highly recommended that future attempts to translate and validate the factorial structure of the FDS should be conducted on a clinical sample or on a mixed sample with at least 50% of participants originating from a clinical sample, a case in which the factorial structure invariance could also be addressed.

In addition to distinguishing between clinical and non-clinical populations, the particularity of our sample (teachers) could have led to additional problems as the content/wording of frustration intolerance beliefs in a stressful classroom setting might be different from the frustration intolerance beliefs in a clinical setting (see DiGiuseppe et al. (2020), for an extended discussion on the content versus process in organizing irrational beliefs). The current study also focuses only on teachers, which was unbalanced regarding gender, as most were women.

Therefore, the instrument's psychometric properties could be tested on other samples. Likewise, future studies should also assess the temporal stability and sensitivity to treatment effects of the FDS scale, as the cross-sectional approach that we have used preclude us from seeing how reliable the FDS results are from a temporal perspective.

Despite these limitations that made us consider our endeavor a preliminary validation study of the Romanian version, the FDS scale has good psychometric properties. It will allow researchers and practitioners to assess and identify individuals scoring high on frustration intolerance, helping mental health specialists provide a tailored intervention by considering the level of frustration intolerance in the target group. For instance, teachers who scored high on FDS overall scores were selected as potential recipients of an REBT group intervention in a subsequent intervention study meant to decrease teachers’ frustration intolerance in school settings.