Distress tolerance (DT) has been defined as an individual’s ability to withstand negative emotional, cognitive, and/or physical states. DT has been identified as a contributing factor to the development and maintenance of many different forms of psychopathology. However, DT has been both conceptualized and operationalized in a variety of different ways. For instance, despite limited empirical data to support this notion, DT is often presumed to be a stable, trait-like characteristic (McHugh et al., 2011), and many DT measures reflect this presumed stable, trait like characteristic by asking the respondent to rate their general tendency to tolerate distress. Thus, some self-report measures of DT may need to be adjusted to assess DT over a shorter period of time in order to accurately measure the construct for potential changes.
The purpose of the current study is to evaluate the factor structure of a brief, 16-item version of the Frustration Discomfort Scale (FDS; Harrington, 2005) when the rating instructions are reframed so the participant is asked to focus the ratings on the previous two weeks. The original FDS is a 28 item self-report measure of frustration intolerance and self-worth attitudes and beliefs. Items (attitudes and beliefs) are rated on a five-point Likert scale ranging from 1 (“absent”) to 5 (“very strong”). The original instructions, which gave no time frame for rating the statements, were modified to indicate the degree each statement applied to the participant over the past two weeks. Additionally, the top four items with the highest factor loadings on each of the four factors were retained to form a brief, 16-item version in order to reduce participant burden.
Confirmatory factor analysis (CFA) was conducted on the correlation matrix of the 16 FDS items using MPlus v7.11 (Muthen & Muthen, 2012). Based on previous CFA of the original FDS (Harrington, 2005), a four factor model was tested in which items loaded on one of the following factors: Emotional Intolerance, Entitlement, Discomfort Intolerance, or Achievement. The four factor solution was fit to the data yielding χ2 = 213.72, p < 0.001, df = 98 with RMSEA = .07 (90% CI = .06, .09); CFI = .93; SRMR = .05, indicating good model fit.
These results provide factor analytic support for the four factor solution previously suggested by Harrington (2005) and provides initial support for this factor solution in a two-week version of the DIS. Furthermore, the results provide support for a brief 16-item version of the FDS that adequately assesses all four factors while reducing participant burden and item redundancy.
Clinical Psychology Doctoral Student
Texas Tech University