Elsevier

Addictive Behaviors

Volume 33, Issue 1, January 2008, Pages 54-68
Addictive Behaviors

The Positive Drinking Consequences Questionnaire (PDCQ): Validation of a new assessment tool

https://doi.org/10.1016/j.addbeh.2007.06.003Get rights and content

Abstract

Expected and experienced negative consequences and expected positive consequences of alcohol use have been widely studied, while little attention has been given to experienced positive drinking consequences. Although existing studies suggest that positive consequences may be important [Park, C.L. (2004). Positive and negative consequences of alcohol consumption in college students. Addictive Behaviors, 29, 311–321.; Park, C.L. & Grant, C. (2005). Determinants of positive and negative consequences of alcohol consumption in college students: Alcohol use, gender, and psychological characteristics. Addictive Behaviors, 30, 755–765.], it is not clear if they are distinct from expected positive outcomes or uniquely associated with drinking behavior. The primary goal of the current study was to develop a measure that directly assessed specific, real life drinking consequences rather than relying on general past tense derivations (“I forgot my worries”) of expectancy items. Such a measure is necessary to determine whether or not positive consequences are distinct from positive expectancies and to assess the unique contribution of positive drinking consequences to drinking behavior. Participants were 423 undergraduate students who completed an online survey; 277 drinkers (56.5% women) completed all data necessary for analyses. Principal components analysis of the Positive Drinking Consequences Questionnaire (PDCQ) identified a single-factor structure with good internal and split-half reliability. The PDCQ also demonstrated discriminant validity relative to a positive expectancy measure and incremental validity in relation to drinking behavior. Although additional studies with heavier drinking populations are needed, the PDCQ may ultimately serve as a valuable research and clinical assessment tool.

Introduction

Alcohol use is prevalent in the U.S., and the majority of young people begin drinking well before they reach the legal drinking age. Alcohol consumption typically peaks during early adulthood when many young people are attending college. The 2005 Monitoring the Future national survey found that 81.2% of college students use alcohol (Johnston, O'Malley, Bachman, & Schulenberg, 2006). Although overall prevalence rates may be a source for concern, those drinking in moderation contribute relatively little to the social and economic burden associated with alcohol use. Thus, it is critical to understand factors that contribute to continued drinking and escalation of drinking behavior over time.

The theory of operant conditioning (Skinner, 1953) suggests that those who experience primarily positive consequences and few negative consequences will persist in their use of alcohol. Presumably, it is the decrease in positive consequences and increase in negative consequences at higher levels of consumption that prevents most individuals from drinking heavily. Although such a simple model is appealing, social learning theory (Bandura, 1986) suggests that the picture is much more complex. Expectations about reward and punishment may develop prior to experience with a behavior via social modeling. Further, these expectations may facilitate the initiation of a behavior and help to sustain it, even when consequences of the behavior are not consistent with positive expectations about outcomes. Thus, both expected positive and negative outcomes and actual consequences experienced are likely to be important drivers of behavior.

In fact, empirical research has consistently suggested that alcohol outcome expectancies (AEs) are associated with the initiation and maintenance of alcohol consumption (Borjesson and Dunn, 2001, Brown, 1993, Goldman, 1994, Brown et al., 1987, Leigh, 1987, Leigh, 1989, Satre and Knight, 2001, Young and Oei, 1993). AEs are beliefs that individuals have about how alcohol affects their mood, emotions, and behavior (Leigh, 1989), and they are typically categorized as either positive (e.g., happy, talkative, romantic, and silly) or negative (e.g., depressed, aggressive, argumentative, and sick). Evidence suggests that AEs can accurately predict drinking patterns and a large number of drinking-related behaviors along the continuum from abstinence to alcohol dependence (Leigh, 1989). In fact, AEs have been found to be at least as important if not more important than the pharmacological effects of alcohol in determining behavioral consequences (Darkes and Goldman, 1993, Darkes and Goldman, 1998, Marlatt and Rohsenow, 1980).

Consistent with social learning theory, correlations between self-reported drinking behavior and positive expectancies are consistently significant and positive (Brown et al., 1987, Brown et al., 1980, Christiansen and Goldman, 1983, Fromme and D'Amico, 2000, Fromme et al., 1993), while individuals tend to report corresponding inverse associations between drinking behavior and negative expectancies (Dunn and Goldman, 1996, McMahon and Jones, 1994). Jones, Corbin, and Fromme (2001) note that, “[Individuals] appear to consume alcohol in a way that delivers the effects they expect.” In addition, increases in the level of alcohol consumed correspond to increases in positive expectancies, such that the heaviest drinkers tend to have the most positive expectancies (Brown et al., 1987). This positive feedback scenario, suggesting a reciprocal relationship between expectancies and drinking, has been demonstrated both cross-sectionally and prospectively (Smith, Goldman, Greenbaum, & Christiansen, 1995).

At the same time, changes in expectancies are not always associated with respective changes in behavior. This lack of correspondence has been demonstrated in both survey and experimental studies with respect to positive expectancies (see Jones et al., 2001). Drinking reductions have been found in the absence of changes in expectancies (Fromme, Kivlahan, & Marlatt, 1986), and interventions targeting expectancies have led to changes in expectancies without consistent reductions in consumption (Corbin, McNair, & Carter, 2001). Although negative expectancies are related to decreased drinking behavior in young, inexperienced drinkers (Dunn & Goldman, 1996) and in problem drinkers (McMahon & Jones, 1996) they are not consistently related to reduced drinking across all developmental time points. For example, negative expectancies have not been consistently associated with drinking reductions in college students (Leigh & Stacy, 1993).

The extent to which expectancies accurately reflect the consequences an individual experiences may help explain inconsistent relations between expectancies and behavior. For example, an increase in negative consequences may lead to reduced drinking via operant conditioning in the absence of changes in expectancies. Similarly, an intervention that targets positive expectancies (based on the assumption that they are inaccurate) may have little impact if the individual's expectancies are based on the actual consequences experienced. Thus, understanding the relations among consequences, expectancies, and their respective influences on drinking behavior may improve our understanding of the factors that maintain or increase alcohol consumption.

At present, there is limited research on positive consequences, and limitations of existing measures make it difficult to determine the nature of the relationship between consequences and expectations (Park, 2004, Park and Grant, 2005). In contrast, there are well-validated measures of negative expectancies (e.g. Comprehensive Effects of Alcohol; CEOA: Fromme et al., 1993; Negative Alcohol Expectancy Questionnaire; NAEQ: McMahon & Jones, 1994) and negative consequences of drinking (e.g. Young Adult Alcohol Problems Screening Test; YAAPST; Hurlbut & Sher, 1992; Rutgers Alcohol Problems Index; RAPI: White & Labouvie, 1989). In addition, a recent study conducted by Noar, LaForge, Maddock, and Wood (2002) found that negative alcohol-related consequences, conceptualized as pros versus cons of alcohol use by the Alcohol Decisional Balance Scale, correlated with negative expectancies (r = .27) as measured by the Comprehensive Effects of Alcohol Scale (CEOA; Fromme et al., 1993). This correlation suggests that while these two constructs do overlap, they are also distinct. Thus, discrepancies between consequences and expected outcomes may be common and may impact the nature of relations between expectancies and behavior.

While still in the early stages, Crystal Park (2004) has begun a program of research examining the positive consequences of alcohol use. Using self-report measures, Park (2004) has examined the frequency and degree of positive and negative alcohol-related consequences, the relationship between consumption level and these consequences, and the impact of these consequences on intentions regarding future drinking behavior. Results show a positive correlation between consumption level and both positive and negative consequences. In addition, participants report more extreme and more frequent encounters with positive consequences than with negative consequences. Collectively, these preliminary results suggest an important role for positive consequences in drinking behavior. However, limitations of the existing measure leave important questions about the distinctiveness of consequences and expectancies and the relative influence of these two constructs unanswered.

The positive consequence measure used by Park (2004) was derived from an expectancy measure developed by Kushner, Sher, Wood, and Wood (1994). Items were adapted to represent past-tense derivations of the original items. However, past tense derivations (e.g., “I forgot my worries” or “I had better ideas”) may not be sufficiently distinct from expectancy items that are in the hypothetical tense (e.g., “Drinking helps me forget my worries” or “Drinking helps me have better ideas”). In a follow-up study, Park and Grant (2005) assessed the relationship between positive consequences and expectancies, and strong positive correlations emerged for both men (r = .72) and women (r = .69). Although these correlations may reflect the true nature of the relationship between expectancies and consequences, they might also be due to the fact that items on the positive consequences measure are not sufficiently distinct from the expectancy items. Thus, while the work of Park and colleagues represents an important first step toward developing an empirically based measure of positive consequences of alcohol use, it is important to continue to work toward the development of a measure that is clearly distinct in content from positive expectancy questionnaires. Clear differentiation from expectancy measures is critical for evaluating the unique contribution of positive drinking consequences to alcohol consumption.

Thus, the goal of the current study was to develop a reliable and valid positive alcohol consequences scale that is sufficiently distinct in content from existing expectancy measures, to allow assessment of the unique contribution of positive drinking consequences to drinking behavior. We derived item content from existing expectancy scales but attempted to make the items more event-specific than the items used by Park and Grant (2005). We believe this type of approach is necessary to identify the true experience of consequences, distinct from general expectations that certain types of outcomes are likely to occur after drinking (expectancies). The factor structure of the new measure was established with principal components analysis (PCA), and analyses of reliability (internal consistency and split-half reliability) and validity (concurrent, discriminant, and incremental) were conducted.

Section snippets

Participants

Participants were 423 students from a private university in the northeastern United States who were recruited during the fall of 2005. A random and representative sample of undergraduates (N = 1440) stratified by gender, year in school, and residential housing units was selected. Within each of twelve residential housing units, a total of 30 students (15 males and 15 females) from each year in school (freshman, sophomore, junior, and senior) was selected. This approach was designed to assure

Data management

A total of 277 participants reported having consumed alcohol and completed all measures leading up to and including the positive consequences measure. Additional participants were excluded from certain analyses because they failed to complete measures included in the analysis. Twenty-one participants were missing negative consequences (RAPI) data, and two participants had missing data on drinking quantity, which reduced the number of cases to 254 for analyses containing both of these variables

Discussion

Consistent with the study goal, a reliable and valid measure of positive drinking consequences was developed with content distinct from existing expectancy measures. Analyses of discriminant and concurrent validity provide evidence of the construct validity of the measure. As expected, positive drinking consequences were more strongly associated with negative consequences (concurrent validity) than they were positive expectancies (discriminant validity). Of critical importance, the PDCQ was

References (45)

  • S.A. Brown

    Drug effect expectancies and addictive behavior change

    Experimental and Clinical Psychopharmacology

    (1993)
  • S.A. Brown et al.

    The Alcohol Expectancy Questionnaire: An instrument for the assessment of adolescent and adult alcohol expectancies

    Journal of Studies on Alcohol

    (1987)
  • S.A. Brown et al.

    Expectations of reinforcement from alcohol: Their domain and relation to drinking patterns

    Journal of Consulting and Clinical Psychology

    (1980)
  • B.A. Christiansen et al.

    Alcohol-related expectancies versus demographic/background variables in the prediction of adolescent drinking

    Journal of Consulting and Clinical Psychology

    (1983)
  • J. Cohen

    Statistical power analysis for the behavioral sciences

    (1988)
  • R.L. Collins et al.

    Social determinants of alcohol consumption: The effects of social interaction and model status on the self-administration of alcohol

    Journal of Consulting and Clinical Psychology

    (1985)
  • J. Darkes et al.

    Expectancy challenge and drinking reduction: Experimental evidence for a mediational process

    Journal of Consulting and Clinical Psychology

    (1993)
  • J. Darkes et al.

    Expectancy challenge and drinking reduction: Process and structure in the alcohol expectancy network

    Experimental and Clinical Psychopharmacology

    (1998)
  • M.E. Dunn et al.

    Empirical modeling of an alcohol expectancy memory network in elementary school children as a function of grade

    Experimental and Clinical Psychopharmacology

    (1996)
  • K. Fromme et al.

    Measuring adolescent alcohol outcome expectancies

    Psychology of Addictive Behaviors

    (2000)
  • K. Fromme et al.

    Comprehensive effects of alcohol: Development and psychometric assessment of a new expectancy questionnaire

    Psychological Assessment

    (1993)
  • L.S. Ham et al.

    Psychometric assessment of the Comprehensive Effects of Alcohol Questionnaire: Comparing a brief version to the original full scale

    Journal of Psychopathology and Behavioral Assessment

    (2005)
  • Cited by (50)

    View all citing articles on Scopus
    View full text