|
Article Excerpt
IN ACCORDANCE WITH SOCIAL LEARNING theory of drinking and alcoholism (for a review, see Maisto et al. [1999]), self-efficacy has been shown to predict relapse following treatment for alcohol dependence (e.g., Goldbeck et al., 1997). Moreover, changes in self-efficacy over the course of treatment seem to reflect an individual's future success in achieving long-term abstinence (Burling et al., 1989). Hence, nonpharmacological treatments for alcohol dependence, including cognitive-behavioral therapy (e.g., Marlatt and Gordon, 1985) and client-centered approaches (e.g., Miller and Rollnick, 2002), seek to strengthen clients' confidence in their ability to resist the urge to drink alcohol. However, previous research has revealed that alcoholism treatment clients tend to be overconfident in terms of their ability to resist the urge to drink (e.g., Burling et al., 1989). Therefore, the predictive power of self-efficacy is limited by a ceiling effect (e.g., Goldbeck et al., 1997). Moreover, relapsers continue to make fairly high self-efficacy ratings despite clear evidence of failure (e.g., Maisto et al., 1998). These findings appear to be counterintuitive, because performance experiences are assumed to be the most important sources of self-efficacy (Bandura, 1986).
To date, virtually nothing is known about the sources and correlates of inflated self-efficacy. Most recently, overconfidence following relapse has been shown to be related to defensive coping (Demmel and Rist, 2005). Research on unsuccessful dieting suggests that inflated self-efficacy might be the result of self-serving external attributions for failure (for a review, see Polivy and Herman [2002]). Moreover, because self-efficacy beliefs are positively related to social desirability (e.g., Silverthorn and Gekoski, 1995), self-efficacy ratings of alcoholism treatment clients are likely to be affected by deliberate impression management and self-deception (Paulhus, 1984). Whereas the need to restore self-esteem and support a positive self-concept seems to nurture inflated self-efficacy, an individual's beliefs concerning the causes and outcomes of another person's behavior are likely to be negatively biased because of self-enhancement motives (for a review, see Chambers and Windschitl [2004]). For example, alcohol-dependent inpatients have been shown to be more optimistic in predicting favorable treatment outcomes for themselves when comparing their own ability to refrain from drinking to that of other alcohol-dependent individuals (Demmel and Beck, 2004). However, self-efficacy beliefs are both different from yet related to general beliefs regarding the likelihood of relapse. Findings from a cross-sectional study suggest that respondents use their self-efficacy expectations as a baseline when making predictions about the success of others (Demmel and Beck, 2004). Given that projection is more prevalent in everyday life than introjection (for a review, see Krueger [2002]), pessimistic other-efficacy beliefs may reflect an individual's true expectations more accurately.
The present study seeks to expand previous research on self-efficacy by establishing the relative predictive validity of alcoholism treatment clients' confidence in their ability to resist the urge to drink and general beliefs concerning the likelihood of relapse. Moreover, the concurrent validity of a well-established multi-item self-efficacy scale and two simple single-item measures of clients' confidence is determined. Using multi-item scales such as the Drug Taking Confidence Questionnaire (DTCQ; Sklar et al., 1997) to assess self-efficacy is likely to improve reliability. However, the development and validation of simple measures may encourage both clinicians and researchers to assess self-efficacy more frequently. Previous findings suggest that a one-time comprehensive assessment of self-efficacy might be less useful than a series of simple assessments over the course of treatment (e.g., Shiffman et al., 2000).
Method
Participants
One...
|
