Measuring Beliefs About Automatic Mood Regulation: Development of a Self-Report Scale

In order to further test the unidimensionality of the BAMR, we also ran a confirmatory factor analysis on the sample from Study 1. We excluded the 3 variables that were later eliminated in Study 2, leaving a total of 17 items to be analyzed. One item from the final BAMR was missing from this analysis because this item was added to the measure only after the completion of Study 1. We ran our confirmatory analysis testing a one-factor model using the confa command in Stata (Kolenikov 2009), which uses maximum likelihood estimation. Fit indices suggested that the model did not fit the data well. In particular, the goodness of fit test showed that the model had a likelihood ratio of 268.75, p < .0001. The root mean square error of approximation (RMSEA = 0.11) was above the desired cutoff of 0.05 (Kolenikov 2009). The Tucker–Lewis Index (TLI = 0.43) and the Comparative Fit Index (CFI = .50) were also below the desired cutoffs of 0.95 (Hu & Bentler 1999). Given that the exploratory analysis in Study 2 showed three factors, it is not surprising this one-factor model did not fit the data from Study 1.

In order to better understand the factor structure of the sample from Study 1, we also ran an exploratory factor analysis using the principal axes method of extracting factors and promax oblique rotation. A scree plot suggested the retention of approximately three factors. The first factor accounted for 33.15 % of the variance in BAMR scores, while the second and third factors accounted for only 12.36 and 7.29 % of the variance respectively. Thus, both Studies 1 and 2 show strong first factor variance overall. However, in contrast to Study 2, the rotated factor loadings from Study 1 suggested the possibility of meaningful subscales. Items loading highest on factor two primarily related to feeling better without effort, and items loading highest on factor three were primarily reverse-scored items related to engaging in active coping. However, we are reluctant to make any definitive conclusions based on this analysis for several reasons. A 10:1 ratio of cases to indicators is one of the most common guidelines for sample size in factor analysis, and some experts recommend minimum sample sizes of 200–400 (Kline 2013). In addition, the final version of the BAMR that is factor analyzed in Study 2 has one item not included in Study 1. Given these concerns, we place more emphasis on the exploratory analysis conducted using the final BAMR and larger sample in Study 2. As elaborated in the “Discussion” section, future research should continue to examine the factor structure of the BAMR.

These instructions are a modified version of those from “Measuring Generalized Expectancies for Negative Mood Regulation: Initial Scale Development and Implications,” by S. J. Catanzaro and J. Mearns, 1990, Journal of Personality Assessment, 54, pp. 562–563. © 1990 Lawrence Erlbaum Associates, Inc. (Taylor & Francis Group). Adapted with permission of the authors and the publisher (Taylor & Francis Ltd., http://​www.​tandf.​co.​uk/​journals, http://​www.​informaworld.​com).

This item stem is from “Measuring Generalized Expectancies for Negative Mood Regulation: Initial Scale Development and Implications,” by S. J. Catanzaro and J. Mearns, 1990, Journal of Personality Assessment, 54, p. 552. © 1990 Lawrence Erlbaum Associates, Inc. (Taylor & Francis Group). Reprinted with permission of the authors and the publisher (Taylor & Francis Ltd., http://​www.​tandf.​co.​uk/​journals, http://​www.​informaworld.​com).

(r) = reverse scored item.

Based on item 29 in the Emotional Schema Questionnaire: “Strong feelings only last a short period of time.” (Leahy 2002, p. 181).

Based on item 4 in the Affective Control Scale: “If I get depressed, I am quite sure that I’ll bounce right back.” (Williams et al. 1997, p. 242).

Based on item 24 in the Negative Mood Regulation Scale, “When I’m upset, I believe that I’ll be upset for a long time,” from “Measuring Generalized Expectancies for Negative Mood Regulation: Initial Scale Development and Implications,” by S. J. Catanzaro and J. Mearns, 1990, Journal of Personality Assessment, 54, p. 552. © 1990 Lawrence Erlbaum Associates, Inc. (Taylor & Francis Group). Adapted with permission of the authors and the publisher (Taylor & Francis Ltd., http://​www.​tandf.​co.​uk/​journals, http://​www.​informaworld.​com).

Based on item 38 in the Affective Control Scale: “I don’t really mind feeling nervous; I know it’s just a passing thing” (Williams et al. 1997, p. 242).