Research Brief

A Comparison of the French and English Versions of the Computerized Lifestyle Assessment Instrument

No. B-09

John R. Weekes, Susan A. Vanderburg, and William A. Millson

April, 1995

Executive Summary

The Computerized Lifestyle Assessment Instrument or CLAI was originally developed in English (see Robinson, Fabiano, Porporino, Millson, & Graves, 1992; Skinner, 1994). When the Correctional Service of Canada adapted the CLAI for offenders, the instrument was translated into French for use with francophone offenders. An important research and operational question is the reliability, validity, and the general appropriateness of the translated version of the CLAI for administration to French-speaking offenders. This report examined some of these issues by comparing the psychometric structure of the French- and English-language versions of the CLAI, by assessing francophone (N=1,333) and anglophone offenders' (N=4,453) feedback on their impressions of the instrument, and by examining the nature and characteristics of substance abuse problems for francophone and anglophone offenders. The CLAI was administered to offenders within their first few weeks of incarceration as part of the routine initial assessment process.

The internal consistencies of the two versions were also the same. Our results clearly indicated that the structure of the two versions of the CLAI were virtually identical with one another. Specifically, factor analysis of the alcohol and drug screening measures revealed that the French and English versions of the instrument were very similar. Finally, offenders' reactions to completing the instrument were extremely positive and were very similar for francophone and anglophone offenders.

Our analyses of the severity of alcohol problems, drug problems, and a combined alcohol and drug severity index, revealed that the two groups of offenders did not differ dramatically from one another, although we did identify a number of differences between francophone offenders and anglophone offenders in terms of the characteristics of their alcohol and drug use.

In summary, this study adds to previous research on the CLAI (e.g., Robinson, Porporino, & Millson, 1991; Vanderburg, Weekes, & Millson, 1994; Weekes, et al., 1993) and supports the use of the French and English versions of the CLAI as a means of accurately identifying substance abuse among offenders regardless of their cultural and linguistic backgrounds.

TABLE OF CONTENTS

• EXECUTIVE SUMMARY
• LIST OF FIGURES
• INTRODUCTION
• METHOD
• CLAI Respondents
• CLAI Administration Procedure
• RESULTS
• Comparison of the Psychometric Properties of the French and English Versions of the CLAI
• Internal Reliability of the Alcohol and Drug Screening Measures
• Factor Structure of the Alcohol and Drug Screening Measures
• Offender Feedback
• Nature and Severity of Substance Abuse Among Francophone and Anglophone Offenders
• Alcohol Use
• Drug Use
• Combined Alcohol and Drug Use
• CONCLUSION
• REFERENCES

LIST OF FIGURES

Figure 1 Severity of Alcohol Problems

Figure 2 Severity of Drug Problems

Figure 3 Severity of Combined Alcohol and Drug Problems

 

Introduction

The accurate assessment of the nature and severity of offenders' alcohol and drug problems is of pivotal importance for the development of effective treatment plans (see Fabiano, 1993). In 1990, the Correctional Service of Canada (CSC) began the national implemention of a computerized assessment system, the Computerized Lifestyle Assessment Instrument or CLAI, as an initial substance abuse assessment tool in each of its institutions across Canada. The CLAI was originally developed by the Addictions Research Foundation (we Skinner, 1994). In turn, the CSC modified the original instrument to reflect the offender population by providing a detailed assessment of respondents' criminal behaviour and by examining the relationship between alcohol and drug use and criminality.

The CLAI includes standardized alcohol and drug screening measures, the Alcohol Dependence Scale, or ADS (see Hodgins & Lightfoot, 1989; Skinner & Horn, 1984) and the Drug Abuse Screening Test or DAST (Skinner, 1992). The offender version of the CLAI that was developed by CSC includes a substantial section consisting of questions relating to offenders' present crime, criminal history, and questions which examine potential relationships between substance use and criminal behaviour. In addition, the CLAI is designed to assess previous substance abuse treatment, motivation for treatment, as well as a range of other lifestyle areas including physical and mental health, social functioning, education, work, and finances. Presently, the CLAI is in active use in most CSC facilities. For additional details the interested reader is referred to a recent report by Robinson, Fabiano, Porporino, Millson and Graves (1993).

A concern arising from the use of standard assessment instruments such as the CLAI is the appropriateness of their use with offenders with different cultural backgrounds and linguistic profiles. Strong cultural differences may interfere with the way in which respondents from cultural backgrounds that are different from the group on which the instrument was developed interpret and respond to individual items. Indeed, Van de Vijver and Poortinga (1982) have pointed out that the underlying psychological constructs of tests which have been translated may not be the same as those of the original test due to cultural differences. Accordingly, it is important that basic research be conducted to determine whether or not a particular instrument or procedure can be administered to individuals from different cultural backgrounds. For instance, a recent study conducted by the Research and Statistics Branch examined the use of the CLAI with Native offenders - a group of offenders who differ from Non-native offenders in terms of language and cultural background (Vanderburg, Weekes, & Millson, 1994). This study concluded that the CLAI was appropriate for use with Native offenders.

Another important issue is the accuracy of the French language version of the CLAI with francophone offenders. About 30% of Canada's incarcerated federal offenders are incarcerated in the Province of Quebec (Solicitor General of Canada, 1994) - the majority of whom are francophone. Although the CLAI has been translated into French, it is important to determine the comparability of the French and English versions of the instrument as well as the appropriateness of the French version for use with Francophone Canadians. Indeed, recent research has underscored the necessity for tests and measures that have been developed for a particular group or clientele (e.g., English-speaking offenders), but have been translated into another language (e.g., French), to be validated to ensure that the translation is accurate and that the translated version of the instrument measures the same phenomena and constructs as the original instrument (Bourque, Blanchard, & Vézina, 1990; Pelletier & Vallerand, 1990; Vallieres, & Vallerand, 1990; Vézina, Landreville, Bourque, & Blanchard, 1991). For instance, the instructions used by some instruments may be too technical or difficult to translate accurately. This limitation may result in respondents interpreting the purpose and nature of the measure incorrectly.

In addition, the underlying meaning or concept which a particular item examines may be functionally altered or lost through the process of translation. As a result, researchers must use on a number statistical techniques to ensure that accurate translations have been made, that there are no substantive differences between the two versions of the test, and that the translated instrument remains intact.

In this report, we examined the reliability and appropriateness of the French version of the CLAI with a large sample of francophone offenders. First, we examined the psychometric structure of the primary alcohol and drug measures (i.e., ADS and DAST) that are used to identify the presence of substance abuse problems. Next, we examined offenders' responses to a number of questions regarding their impressions of the CLAI. Finally, we compared data on the nature and severity of substance abuse and criminal behaviour using a group of francophone offenders who completed the French version of the CLAI with a group of anglophone offenders who completed the English version of the instrument.

Method CLAI Respondents

The database for this study consisted of 5,786 federal offenders (5,690 male and 96 female) who completed the CLAI as part of a routine intake assessment protocol administered to all offenders within the first few weeks of arrival at the institution. A total of 1,333 Francophone offenders (23% of the total sample) completed the French version of the CLAI and 4,453 anglophobes offenders (77% of the sample) responded to the English version. At time of testing, the average age of the Francophone offenders was 30 years (SD = 8.40 years) with a range between 18 and 69 years. The average age of the anglophone offenders was also 30 years (SD = 8.98 years) with a range between 18 and 68 years.

CLAI Administration Procedure

Offenders were seated at a micro-computer and completed the CLAI instrument by responding to multiple-choice questions presented to them on the computer screen. Offenders were supervised by an institutional staff member who had been trained to administer and supervise testing sessions using the CLAI system. The CLAI is designed to provide respondents with immediate feedback on their performance using easy-to-read graphs that were presented on the computer screw at key points during the session. At the end of the testing session, offenders were given a hard-copy to keep and review on their own which summarized their test results. A second copy was forwarded to their institutional case management officer for use in treatment planning.

Results Comparison of the Psychometric Structure and Properties of the French and English Versions of the CLAI

As mentioned previously, several standardized alcohol and drug screening instruments have been incorporated into the CLAI. The ADS is a 25-item scale that assesses the severity of alcohol dependence experienced by an individual. The DAST is a 20-item measure that assesses the number and severity of problems associated with drug use.

Due to the fact that the CLAI is used primarily to assess the severity of offenders' substance abuse problems and to match offenders to appropriate substance abuse treatment, the following analyses paid specific attention to the nature of offenders' responses to the alcohol and drug screening measures.

Internal Reliability of the Alcohol and Drug Measures. The reliability of the CLAI was assessed by examining the internal consistency of the French and English versions the ADS and DAST measures. The statistic of reliability (Cronbach's Alpha) was high for both the ADS (.94) and the DAST (.87) for francophone offenders. Comparably high reliability values were also generated by the anglophone offenders who completed the English version of the CLAI (ADS=.96; DAST=.88). These data suggest that the instruments used to screen drug and alcohol problems for the French and English versions are almost identical with one another and are uniformly reliable.

Factor Structure of the Alcohol and Drug Screening Measures. We conducted separate factor analyses of the ADS and DAST administered in the French and English versions of the CLAI. Factor analysis is a sophisticated statistical technique used to
examine the underlying structure of questionnaires and other psychological measures (Tabachnick & Fidell, 1989). Factor analysis groups together questionnaire items that tap the same construct into common factors which may be used as scales.

The results of these analyses indicated that the factor structures of the primary factors were very similar for the French-speaking and English-speaking offender groups (i.e., those factors that accounted for the greatest proportion of variance). Indeed, a subsequent test of the actor congruence (Everett & Entrekin, 1980) confirmed that the factor structures of the main factors were virtually indistinguishable for the two groups of offenders. In other words, the ADS and DAST measures generated similar information regarding alcohol and drug problems for francophone and anglophone offenders, alike. These findings further support the conclusion that, in general, the CLAI provides accurate representations of the alcohol and drug problems of both francophone and anglophone offenders, and, that these two groups of offenders interpreted the items contained in these screening measures in a similar manner.

Offender Feedback

We also examined offenders' responses to a series of questions administered at the end of the CLAI session which asked them about what they thought of doing the survey. There was little, if any, difference in the patterns of responses provided by offenders who completed the French version of the CLAI and those who completed the English version. Moreover, the results were overwhelmingly positive and were consistent with the results obtained from other research on the CLAI (e.g., Robinson, Porporino, & Millson, 1991; Vanderburg, Weekes, & Millson, 1994).

Specifically, we found that about 94% of anglophone offenders and 97% of francophone offenders indicated that understood the instructions and questions. The same percentages of anglophone and francophone offenders reported that they understood the feedback they received on their lifestyle habits. 80% of francophones and 79% of anglophones, felt that they CLAI session was about the right length and 79% of francophones and 80% of anglophones did not experience any difficulty completing the CLAI on the computer. 92% of francophone offenders reported that the CLAI provided an clear picture of their lifestyle habits whereas 76% of anglophone offenders felt that the CLAI was accurate. About 4 out of 5 offenders in each group reported that they learned something about themselves by completing the CLAI. Finally, 94% of francophone offenders and 90% of anglophone offenders indicated that they liked doing the CLAI, 94% of francophones and about 87% of anglophones indicated that they would encourage a friend to do the CLAI.

Nature and Severity of Substance Abuse Among Francophone and Anglophone Offenders

Alcohol Use. Approximately the same proportions of francophone (50.3%) and anglophone offenders (53.5%) reported alcohol problems. The severity of francophone and anglophone offenders' alcohol problems were then categorized into one of the following four levels: "no alcohol problem", "low level problem", "intermediate problem", and "severe problem". The proportions of offenders falling in each of the four categories is displayed in Figure 1. As is evident in this first figure, although there were minor variations in the proportion of francophone and anglophone offenders in each of the severity levels (e.g., 8.9% of anglophone offenders with severe alcohol problems compared with 3.2% of francophones), there were no substantive differences in the severity of alcohol problems between the two groups of sufficient magnitude to question the ADS or to have an impact on treatment resources.

Francophone and anglophone offenders responded similarly on a number of alcohol-related variables. For example, the two offender groups did not differ when we examined the relationship between the severity of offenders' alcohol problem and the age at which they first consumed alcohol, the age at which they began consuming alcohol on a regular basis (i.e., at least once a week), the most recent age at which they voluntarily abstained from consuming alcohol, and the number of times an offender had voluntarily quit using alcohol.

We then correlated individual items relating to alcohol use with various measures of social functioning, education, work, finances, and other aspects of criminal behaviour in order to examine potential differences among the two offender groups. Compared with their francophone counterparts, anglophone offenders with lengthy criminal histories were more likely to squander money on alcohol. Anglophone offenders were also somewhat more likely to drink to unconsciousness than francophones. In addition, visible intoxication or drunkenness (e.g., stumbling, staggering, or weaving) among anglophone offenders was more strongly associated with their ability to recognize problems associated with substance abuse, recognize their need for help, express their desire for treatment, the likelihood that they used alcohol on the day of the offense, and the number of crimes committed under the influence of alcohol than for francophone offenders.

Drug use. Overall, 52.5% of francophone offenders and 46.4% of anglophone offenders reported drug problems. Figure 2 displays the distribution of drug severity scores for the two groups when offenders were classified into one of the four drug severity categories (i.e., "no drug problem", "low level drug problem", "intermediate drug problem", "or severe drug problem"). No meaningful differences were evident between severity of drug problem for francophone and anglophone offenders.

The relationship between drug severity score (i.e., DAST) and a number of other drug-related factors varied according to whether respondents were francophone or anglophone. For example, the frequency with which an offender used heroin during the six month period prior to arrest was more related to the severity of drug problem for anglophone offenders than for francophone offenders. In other words, anglophone offenders with more severe drug problems were more likely to have used heroin during that time period than their francophone counterparts.

Interestingly, the relationships between how old an offender was when he or she first used drugs, how old an offender was the most recent time he or she voluntarily quit drug use, and how many times an offender voluntarily quit using drugs and the severity of drug problem did not differ for the two offender groups.

Combined Alcohol and Drug Use. As shown in Figure 3, when francophone and anglophone offenders were categorized according to their highest score on either the ADS or DAST we found that the two groups did not differ from one another. These results indicate that identical proportions (69%) of francophone offenders and anglophone offenders reported substance abuse problem (i.e., alcohol problems, drug problems, or alcohol and drug problems) of sufficient severity to warrant formal treatment intervention.

Conclusion

In this study, we elected to focus attention specifically on those components of the CLAI that are most important for use by institutional staff (e.g., case managers) in the development of individualized offender treatment plans. Although our study was not designed to directly assess the convergent validity of the CLAI in relation to other measures of substance abuse, the results presented suggest that the French-language version of the CLAI can be used with francophone offenders to produce interpretable information regarding the nature and severity of alcohol and drug problems that is very similar to information obtained for anglophone offenders who completed the English-language version of the CLAI.

We arrived at this conclusion for three main reasons. First, we found that the basic psychometric properties of the French and English versions of the CLAI were virtually indistinguishable. Second, there was a high degree of internal consistency in the alcohol and drug screening instruments. Third, the factor structure of the two instruments used to identify alcohol and drug problems in offenders were very similar to one another. Importantly, the findings regarding the ADS are consistent with recent research which supports the validity and structure of the ADS when it is administered to offenders (Hodgins & Lightfoot, 1989). To our knowledge, the DAST has never been fully evaluated regarding its validity for use with offender clinical populations.

In addition, francophone and anglophone offenders alike offered very positive comments regarding their participation in the CLAI Their responses serve to further underscore the conclusion that the CLAI'S utility lies not only in providing vital information for institutional staff regarding the nature of offenders' substance abuse problems but that the CLAI also serves a useful role as a self-administered learning tool that provides offenders with important feedback regarding their lifestyle habits and behaviour. As well, the offenders appear to enjoy participating in the procedure perhaps relative to more traditional interview and paper-and-pencil measures.

It may be that, in some instances, the CLAI brings additional value to the assessment process by providing offenders with important feedback regarding their alcohol and drug use may alert them to the existence of substance abuse problems that they previously had not recognized.

We found consistent patterns in anglophone and francophone offenders' responses to the screening measures and other CLAI items that related to their present drug and alcohol-taking behaviour, their past substance abuse behaviour, and their criminal behaviour.

Although the distribution of alcohol severity, drug severity, and combined alcohol and drug severity scores were ostensibly the same for the two offender groups, we did unearth some differences between those offenders who completed the CLAI in French and those who completed the CLAI in English. In general our results indicated that we found that alcohol and drug use was more problematic for anglophone offenders than for their francophone counterparts. Although we believe that it is somewhat premature to draw any firm conclusions regarding the implications of these findings, the pattern of results may be reflective of cultural differences in the use of alcohol and drugs.

Taken together, we failed to unearth any patterns in either the psychometric structure of the substance abuse screening measures or anomalies in other key variables relating to substance abuse and criminal behaviour which would cause us to question the use of the French version of the CLAI with francophone offenders. In conclusion, the CLAI continues to offer an efficient, accurate, and inexpensive method of assessing substance abuse problems for either francophone or anglophone offenders. Moreover, the findings presented in this report coupled with recent research supporting the appropriateness of the use of the CLAI with Aboriginal offenders (see Vanderburg, Weekes, & Millson, 1994), suggest that the French and English versions of the CLAI can be administered to offenders from a variety of cultural and linguistic backgrounds.

References

Bourque, P., Blanchard, L., & Vézina, J. (1990). Étude psychométrique de l'Échelle de Dépression Gériatrique. Canadian Journal on Aging, 9, 348-355.

Everett, J. E., & Entrekin L. V. (1980). Factor comparability and the advantages of multiple group factor analysis. Multivariate Behavioral Research, 2, 165-180.

Fabiano, E. (1993). Developing a model for the provision of substance abuse treatment. Ottawa: Correctional Service of Canada.

Hodgins, D. C., & Lightfoot, L. 0. (1989). The use of the Alcohol Dependence Scale with incarcerated male offenders. International Journal of Offender Therapy and Comparative Criminology, 33, 59-67.

Pelletier, L. G., & Vallerand, R. J. (1990). L'Échelle Révisée de Conscience de Soi: Une traduction et une validation canadienne-française du Revised Self-Consciousness Scale. Revue Canadienne des Sciences du Comportement, 22, 191-206.

Robinson, D., Porporino, F. J., & Millson, W. A. (1991). Patterns of alcohol and drug use among federal offenders as assessed by the Computerized Lifestyle Screening Instrument. Ottawa: Correctional Service of Canada.

Skinner, H. A. (1994). The Computerized Lifestyle Assessment Manual. Toronto: Multi-Health Systems.

Skinner, H. A. (1982). The Drug Abuse Screening Test. Toronto: Addiction Research Foundation.

Skinner, H. A., & Horn, J. (1984). Alcohol Dependence Scale User's Guide. Toronto: Addiction Research Foundation.

Solicitor General of Canada. (1994). Basic facts about corrections in Canada 1993. Ottawa: Communications Branch.

Tabachnick, B. G., & Fidell, L. S. (1989). Using Multivariate Statistics. New York: Harper Collins Publishers, Inc.

Vallieres, E. F., & Vallerand, R. J. (1990). Traduction et validation canadienne-français de l'echelle de l'Estime de Soi de Rosenberg. International Journal of Psychology, 25, 305-316.

Vanderburg, S. A., Weekes, J. R, & Millson, W. A. (1994). Assessment of substance abuse problems in Native offenders. The Computerized Lifestyle Assessment Inventory. Ottawa: Correctional Service of Canada.

Van de Vijver, F. J. R, & Poortinga, Y. R (1982). Cross-cultural generalization and universality. Journal of Cross-Cultural Psychology, 13, 387-408.

Vézina, J., Landreville, P., Bourque, P., & Blanchard, L. (1991). Questionnaire de Dépression de Beck: Étude de psychométrique auprès d'une population ƒgée francophone. Canadian Journal on Aging, 10, 29-39.

Weekes, J. R., (1993). Paper presented at the meeting of the Canadian Psychological Association, Montreal, Quebec.

SHEET1.XLS Chart 3

Figure 1

Severity of Alcohol Problem (ADS)

SHEET2.XLS Chart 1

Figure 2

Severity of Drug Problem (DAST)

SHEET4.XLS Chart 1

Figure 3

Severity of Combined Alcohol and Drug Problems