AN EMPIRICAL ANALYSIS OF ALCOHOL ADDICTION: RESULTS FROM THE MONITORING THE FUTURE PANELS

In a panel of young adults, we find that alcohol consumption is addictive in the sense that increases in past or future consumption cause current consumption to rise. The positive and significant future consumption effect is consistent with the hypothesis of rational addiction. The long-run price elasticity is approximately 60% larger than the short-run price elasticity and twice as large as the elasticity that ignores addiction. Thus, a tax hike policy to curtail consumption or abuse may not have a favorable cost-benefit ratio unless it is based on the long-run price elasticity. (JEL 110)     

THE IMPACT OF CLASSROOM EXPERIMENTS ON THE LEARNING OF ECONOMICS: AN EMPIRICAL INVESTIGATION

This paper provides limited evidence on the effectiveness of a simple classroom experiment. In different courses in environmental economics or public finance, a brief take-some game was performed. Students who took part in this classroom experiment, as well as those who just watched it, were more successful in answering a multiple-choice test on the "tragedy of the commons" than control groups from the same courses.     

Construct Validity and Error Components of Survey Measures: A Structural Modeling Approach

Measurement errors can have profound effects on statisticalrelationships, and better information on the quality of measuresseems needed. This study uses a new technology—structuralmodeling of data from special supplements to regular surveys—togenerate estimates of construct validity, method effects (amajor source of correlated error), and residual error (mainlyrandom error) for a broad set of measures obtained from fivenational surveys and an organizational survey (total respondents= 7,706). Analysis of these estimates suggested that a typicalsurvey item, when administered by a respected survey organizationto a general population sample, can be expected to yield 50–83percent valid variance, 0–7 percent method effects variance,and 14–48 percent residual variance. Multivariate analysisshowed that over two-thirds of the variation in measurementquality could be explained by 13 survey design characteristics;characteristics of respondents explained a small additionalportion. Results provide: (a) information on design conditionsassociated with better (or worse) measurement quality, (b) empiricallybased suggestions for improving measurement quality in futuresurveys, and (c) a set of coefficients for predicting the qualityof measures not studied here.     

SKILL AS A DIMENSION OF OCCUPATIONAL CLASSIFICATION

L'habilité est présentée comme une dimension qui règle les catégories sommaires d'une classification socioéconomique des professions, qui a été développée pour l'Etude sur la Mobilité des Canadiens. Cette analyse repose sur les valeurs ged et svp qui reflètent, respectivement, les différents niveaux d'éducation générale et d'apprentissage professionnel spécial requis pour les professions qui constituent chacune des catégories de l'emc. La classification des catégories de l'emc est reliée largement aux classifications basées sur les valeurs moyennes ged et svp. L'analyse des variations dans les valeurs ged et svp, entre et parmi les catkgories de I'emc, rCvkle que la plupart des catCgories sont homogenes, en ce qui conceme I'habileté. L'homogénéite, cependant, varie positivement suivant le niveau d'habileté. Le chercheur qui se concentre sur l'habileté comme dimension saillante de la variabilité entre les professions trouvent utile la classification de l'emc.
Skill is shown to be a dimension which orders the broad categories of a socioeconomic classification of occupations developed for the Canadian Mobility Study. The analysis is based on ged and svp values which, respectively, reflect different levels of general education and special vocational training required of the component occupations of each cms category. The cms ordering of the categories is highly correlated with orders based on average ged and svp values. Analysis of variation in ged and svp values between and within cms categories reveals that most categories are homogeneous with respect to skill. Homogeneity, however, varies positively with skill level. Researchers who focus on skill as a salient dimension of variability among occupations will find the cms classification useful.