Elsevier

Drug and Alcohol Dependence

Volume 153, 1 August 2015, Pages 78-85
Drug and Alcohol Dependence

Positive and negative metacognitions about alcohol use among university students: Psychometric properties of the PAMS and NAMS French versions

https://doi.org/10.1016/j.drugalcdep.2015.06.003Get rights and content

Highlights

  • French versions of the Positive and Negative Alcohol Metacognitions Scales were validated using EFA and CFA.

  • Findings showed support for a two-factor structure of the PAMS and NAMS.

  • Patterns of alcohol use were differentially associated with metacognitions.

  • Uncontrollability was consistently associated with self-report measures of alcohol use.

Abstract

Background

Metacognitions about the positive and negative effects of alcohol use have been associated with various patterns of drinking. The aim of the present study was to validate French versions of the Positive Alcohol Metacognitions Scale (PAMS) and the Negative Alcohol Metacognitions Scale (NAMS) developed by Spada and Wells (2008, Addict. Behav. 33, 515) and to investigate the relationship between metacognitions and patterns of alcohol use among university students.

Methods

Responses of 1600 university students who participated in an internet survey-based study on alcohol use were submitted to confirmatory (N = 800, mean age 20.40 years, 45.50% male) and exploratory (N = 800, mean age 20.34 years, 45.38% male) factor analyses in two separate samples. Alcohol use, binge drinking and mood were also assessed.

Results

In line with the original versions of the scales, results provided support for a two-factor structure of the French PAMS and NAMS. Both scales revealed adequate internal reliability. Good temporal stability was found for the two factors of the NAMS, whereas one factor of the PAMS showed weakness across time. Predictive validity revealed that negative alcohol metacognitions about the uncontrollability of alcohol use were found to be consistently associated with alcohol use and binge drinking, whereas positive metacognitions about alcohol use were found to be differentially associated with alcohol use and binge drinking.

Conclusions

The French versions of the PAMS and NAMS exhibited suitable psychometric properties. This study also emphasized the role of metacognitions about alcohol use in drinking behaviour among university students.

Introduction

The Self-regulatory Executive Function (S-REF) model of psychopathology (Wells and Matthews, 1994, Wells and Matthews, 1996) implicates maladaptive metacognitive processes in the predisposition towards, and maintenance of, psychological dysfunction. In this model, Wells and Matthews propose that psychological dysfunction is driven and maintained by maladaptive coping strategies (e.g., perseverative thinking, threat monitoring, thought suppression), collectively termed the Cognitive Attentional Syndrome (CAS), which cause negative thoughts and emotions to persist. The activation and persistence of the CAS is dependent on maladaptive metacognitions (or metacognitive beliefs). Metacognitions are defined as “stable knowledge or beliefs about one's own cognitive system, and knowledge about factors that affect the functioning of the system; the regulation and awareness of the current state of cognition, and appraisal of the significance of thought and memories” (Wells, 1995, p. 302). In support of this view, metacognitions have been found to be associated with a wide array of psychological and behavioural problems (for a review, see Wells, 2013).

In the domain of alcohol use, metacognitions can be divided into two broad sets: (1) positive metacognitions about the impact of alcohol use on cognitive (e.g., “Drinking helps me to control thoughts”) and emotional (e.g., “Drinking helps me to improve my mood”) self-regulation; and (2) negative metacognitions concerning the perception of lack of executive control over alcohol use (e.g., “My drinking persists no matter how I try to control it”) and the negative impact of alcohol use on cognitive functioning (“Drinking will damage my mind”; Spada and Wells, 2006).

Positive metacognitions about alcohol use can be conceptualised as a specific form of expectancy relating to the use of alcohol as a means of controlling and regulating cognition and emotion (Spada and Wells, 2006). They differ from positive alcohol expectancies in their explicit focus on capturing how alcohol use can help achieve mental control by enhancing problem-solving, acting as a form of thought control, helping to regulate attention, and managing self-image (Spada and Wells, 2008, Spada et al., 2015). Negative metacognitions about alcohol use capture judgements concerning the perception of lack of executive control over alcohol use and the negative impact of alcohol use on cognitive functioning. From a metacognitive standpoint such beliefs are thought to play a crucial role in the perpetuation of alcohol use (Spada and Wells, 2006, Spada et al., 2013, Spada et al., 2015) by becoming activated during and following a drinking episode, and triggering negative emotional states that compel a person to use more. Negative alcohol expectancies differ from such beliefs in as much as they mainly measure general negative outcomes arising from alcohol use (e.g., “I get a hangover”; “I feel guilty”). Research has indicated that metacognitions about alcohol use are an independent predictor of drinking behaviour when controlling for alcohol expectancies (Spada and Wells, 2008).

Spada and Wells (2008) have developed instruments to specifically measure positive and negative metacognitions about alcohol use: the Positive Alcohol Metacognitions Scale (PAMS) and the Negative Alcohol Metacognitions Scale (NAMS). The PAMS consists of 12 items and an exploratory factor analysis conducted among non-clinical and clinical samples yielded a two-factor solution (Spada and Wells, 2008) representing positive metacognitions about using alcohol in emotional and cognitive self-regulation. The authors also reported adequate internal reliability and test–retest reliability over an 8-week interval. The NAMS consists of 6 items and an exploratory factor analysis conducted among the same samples as the PAMS (Spada and Wells, 2008) yielded a two-factor solution representing negative metacognitions about the uncontrollability of alcohol use and about cognitive harm related to alcohol use. The authors reported an adequate internal reliability for both factors, but a poor test–retest reliability for the uncontrollability factor (Spearman's rho = .42).

Exploration of the predictive validity of the two scales revealed significant correlations between factor scores and measures of alcohol use among healthy adults and problem drinkers (Spada et al., 2007a, Spada et al., 2007b; Spada and Wells, 2008). Comparisons of PAMS and NAMS scores among groups of social drinkers, problem drinkers and alcohol-dependent patients (Spada and Wells, 2010) revealed significantly increased scores for each factor of the two scales among the three groups, with alcohol-dependent patients having higher scores and problem drinkers having intermediate scores. More recently, Clark et al. (2012) investigated the relative contribution of personality factors and metacognitions about alcohol use in predicting weekly levels of alcohol use in a sample of 138 binge drinking university students. They showed that positive alcohol metacognitions about cognitive self-regulation, negative alcohol metacognitions about uncontrollability and negative alcohol metacognitions about cognitive harm were significantly correlated with weekly levels of alcohol use. A hierarchical regression analysis revealed that metacognitions about alcohol use accounted for 6% of variance in weekly levels of alcohol use over and above the variance accounted for by gender, and personality factors (agreeableness and conscientiousness). More specifically, findings from this study showed that males with low conscientiousness and high positive alcohol metacognitions about cognitive self-regulation had increased weekly levels of alcohol use.

The aim of the current study was to translate into French the PAMS and NAMS and to examine their psychometric properties in a large non-clinical sample of university students. We first evaluated the goodness of fit of the two-factor model of each scale using confirmatory factor analysis (CFA). Then, we further conducted an exploratory examination of the two scales’ factor structure (EFA). We also investigated reliability (i.e., both internal consistency and test–retest stability) and predictive validity. In line with Clark and colleagues’ (2012) study, we investigated whether metacognitions about alcohol use were related with weekly levels of alcohol use and a more refined measure of binge drinking. Binge drinking involves several aspects of drinking such as drinking quickly, and heavily, to get drunk, and is characterised by repeated bouts of drinking leading to high levels of blood alcohol concentration (BAC) followed by periods in which BAC level return to zero (Scaife and Duka, 2009). For instance, consuming 21 units by week can be the resultant of different patterns of alcohol use such as consuming three drinks by day or consuming them on two occasions. We therefore included a measure of binge drinking developed by Townshend and Duka (2002) which encompasses the speed of drinking, the frequency of being drunk, and the percentage of times getting drunk when drinking.

Section snippets

Translation of the scales

The process of translation of the two scales was carried out as follows. We first forward translated the items of the PAMS and NAMS from English to French. After this stage an independent professional translator (native English speaker) back translated the items into English. Discrepancies emerging between the back translated and the original English versions were debated and translation adjustments were consensually made in agreement with one the originators of the scales (MMS). Final versions

Confirmatory factor analysis

To investigate whether items of the PAMS were related to a single underlying factor we first tested a single factor model. Results suggested that the model had substandard fit: χ2(54) = 276.959, p < .001, χ2/df = 5.13; SRMR = .089; RMSEA = .072 (90% IC: .064–.080); CFI = .913. We then tested the original two-factor model of the PAMS. Results revealed an improved fit of the model to the data with acceptable goodness of fit values: χ2(53) = 268.417, p < .001, χ2/df = 5.06; SRMR = .082; RMSEA = .071 (90% IC:

Discussion

The current study investigated the psychometric properties of the French version of the PAMS and NAMS. Our aim was to validate a version of the scales and to determine whether its factors were related to drinking behaviour. To do this, we used CFA and EFA with robust nonparametric estimators in large sample of data obtained from an internet survey among university students.

Confirmatory factor analyses carried out on the first sample confirmed the two-factor solution of the French PAMS, though a

Role of funding source

Funding for this study was provided by the Mission Interministérielle de Lutte Contre les Drogues et les Conduites Addictives (MILDECA) grant (PREVDROG-11). MILDECA had no further role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

Conflict of interest statement

The authors report no potential conflicts of interest.

Contributors

Gierski and Naassila designed the study. Gierski and Picard conducted the survey process. Gierski undertook the statistical analysis. Spada, Fois, and Van der Linden contributed to data interpretation and manuscript preparation. Gierski wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.

Acknowledgements

The authors would like to thank the anonymous reviewers for their valuable comments and suggestions to improve the manuscript, Elizabeth Wiles-Portier for the backtranslation of the scales and Dr Ana Nikcevic for her careful proofreading.

References (35)

  • M. Bouvard et al.

    Inventaire de dépression (Beck Depression Inventory): Protocoles et échelles d’évaluation en psychiatrie et en psychologie

    (1996)
  • A.L. Comrey et al.

    A First Course in Factor Analysis

    (1992)
  • K.A. Hallgren

    Computing inter-rater reliability for observational data: an overview and tutorial

    Tutor. Quant. Methods Psychol.

    (2012)
  • F.P. Holgado-Tello et al.

    Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables

    Qual. Quant.

    (2008)
  • J.L. Horn

    A rationale and test for the number of factors in factor analysis

    Psychometrika

    (1965)
  • D. Hooper et al.

    Structural equation modelling: guidelines for determining model fit

    Electron. J. Bus. Res. Methods

    (2008)
  • L.-T. Hu et al.

    Cut off criteria for fit indices in covariance structure analysis: conventional criteria versus new alternatives

    Struct. Equ. Model.

    (1999)
  • Cited by (16)

    • The effects of emotion dysregulation and negative affect on urge to smoke and nicotine dependence: The different roles of metacognitions about smoking

      2022, Addictive Behaviors
      Citation Excerpt :

      Research evidence has reported the mediating role of generic metacognitions in the associations of negative affect and emotion dysregulation with addictive behaviors including problematic internet use (Akbari, 2017), nicotine dependence (Spada, Nikčević, Moneta, & Wells, 2007), problem drinking (Moneta, 2011), and the problematic Facebook use (Marino et al., 2016). In addition to generic metacognitions, evidence has shown the role of metacognitions specific to various addictive behaviors including alcohol use (Clark et al., 2012; Gierski et al., 2015; Spada & Wells, 2008), gaming (Caselli, Marino, & Spada, 2020; Spada & Caselli, 2017), social media use (Casale, Rugai, & Fioravanti, 2018), Internet use (Spada, Langston, Nikčević, & Moneta, 2008), and smartphone use (Akbari, Zamani, Fioravanti, & Casale, 2020). Therefore, recent studies have highlighted the importance of both generic (Nikcevic & Spada, 2008; Spada et al., 2007) as well as specific metacognitions (e.g., Spada & Wells, 2008; Spada et al., 2008) in addictive behaviors.

    • Toward a better understanding of the influence of repetitive negative thinking in alcohol use disorder: An examination of moderation effect of metacognitive beliefs and gender

      2020, Addictive Behaviors
      Citation Excerpt :

      It is possible that this heterogeneity might have contributed to the absence of the moderation effect. Furthermore, the PAMS french validation study reports low factor loading (0.37) for item 4 within the cognitive self-regulation factor (Gierski et al., 2015), which calls into question its relevance within the factor. Future research will therefore need to test the specific moderating effect of beliefs about attention regulation and thought regulation.

    • Positive and negative metacognitions about alcohol: Validity of the Norwegian PAMS and NAMS

      2020, Addictive Behaviors
      Citation Excerpt :

      Differences may have occurred because of the nature of the samples studied. The French non-clinical sample also found shared variance between two items of the PAMS (Gierski et al., 2015), while the Persian study used a clinical sample and found support for the original factor structure (Khosravani et al., 2020). The study supported the validity of the PAMS-r and NAMS.

    • Psychometric properties of the Persian versions of the Positive Alcohol Metacognitions Scale (Persian-PAMS) and the Negative Alcohol Metacognitions Scale (Persian-NAMS) in alcohol-dependent individuals

      2020, Addictive Behaviors
      Citation Excerpt :

      These results are in line with previous studies (Gierski et al., 2015; Spada & Wells, 2008), which showed good reliability for the scale. Moreover, similarly to English and French versions of these scales (Gierski et al., 2015; Spada & Wells, 2008), test-retest reliability for the Persian-PAMS and Persian-NAMS and their subscales were found to be good, so that ICC coefficients were between 0.77 and 0.90. Evidence has shown that ICC coefficients between 0.75 and 0.90 refer to good reliability (Koo & Li, 2016).

    View all citing articles on Scopus
    View full text