Emotional Antecedents Of Alcohol Use: A Scale Construction Project

The purpose of the present study was to develop a measure of the Emotional Antecedents of Alcohol Use (EAAU) and investigate its psychometric properties. Focus groups and three pilot studies were conducted to develop and define the construct of alcohol consumption as a means of emotion regulation. Items were written based upon focus group results, existing published instruments, and a review of the literature. Independent raters evaluated the initial item pool for face validity and the initial EAAU was given to small pilot samples (N=64, N=39) to investigate initial internal consistency and content validity. Results of pilot studies suggested that the EAAU is a valid measure of alcohol consumption and positively associated with the Alcohol Use Disorders Identification Test (AUDIT; r = 0.445, p \u3c .001; r = 0.607, p \u3c .001). The EAAU was then administered to a larger sample (N=301) to investigate the structure of the instrument. Exploratory factor analysis results implied that a 14-factor model was the best fitting and most parsimonious and sufficient factor solution to represent the structure of the EAAU scores produced by the current sample. These 14 distinct emotional domains were interpreted as representing the following: (1) Relaxation, (2) Fear/Worry, (3) Reservation, (4) Negative High Arousal, (5) Shock, (6) Sedation, (7) Sadness, (8) Fatigue, (9) Anger, (10) Happiness, (11) Alertness, (12) Annoyance, (13) Serenity, and (14) Positive High Arousal. Results suggested the EAAU demonstrates convergent validity as small to moderate positive relationships were noted between the EAAU and measures of alcohol expectancies (r = 0.374, p \u3c .001) and alcohol-related problems (r = 0.292, p \u3c .001). Results supported marginal discriminant validity of the EAAU as negligible to small negative relationships were noted between the EAAU and a measure of self-esteem (r = -0.157, p = .007). Additionally, the EAAU demonstrated initial evidence of predictive validity as EAAU subscale scores were found to be predictive of difficulties in emotion regulation (R2 = .128, F(14, 286) = 2.991, p \u3c .001) and alcohol consumption (R2 = .111, F(14, 286) = 2.539, p = .002). Lastly, gender and drinker status differences in EAAU scores were found. Males (M=267.74, SD=80.07) were found to obtain higher EAAU scores (t(290) = -2.59, p = .010) in comparison to women (M=244.47, SD=73.26), while problem drinkers (M=281.21, SD=69.81) were found to obtain higher EAAU scores (t(299) = 7.459, p \u3c .001) in comparison to social drinkers (M=218.87, SD=74.61). Study limitations, implications, and future research directions are discussed