GlaxoSmithKline.

reliability bounds, coefficient alpha, non-homogeneous composites,

reliability bounds, non-homogeneous composites, search procedure, trace minimization with constraints,

One hundred and sixteen female adolescent smokers were asked about withdrawal symptoms experienced during past attempts to give up smoking for good. Sixty-three percent reported difficulties during abstinence of the kind experienced by adult smokers. Daily smokers were more likely to report withdrawal effects than non-daily smokers (74% versus 47%, P less than 0.005). Reported experience of withdrawal symptoms was positively related to self-reports of cigarette consumption and depth of inhalation and nicotine intake as indexed by salivary cotinine concentrations. Reported occurrence of at least one withdrawal effect correlated positively with nicotine intake after controlling for behavioural variables. Our results indicate that teenage smokers are likely to suffer withdrawal symptoms when they try to give up. Behavioural factors and expectations based on observations of adults may have played a part in their experience of withdrawal, but it is also likely that pharmacological factors are implicated even at this early stage.

An improved Bayesian method, due to Lindley, for the simultaneous estimation of multiple regressions in m groups is studied by applying the method to test data. Evidence is found to support the belief that in many testing applications the collateral information obtained from each subset of m − 1 colleges will be useful for estimating the regression in the mth college, especially when the sample sizes are small. Using a 25 per cent sample, the Bayesian prediction equations achieved an average 9.7 per cent reduction in mean square error, as compared with the within‐group least squares equations, when cross‐validated with a later sample. More importantly, the mean square error for the Bayesian equations based on the 25 per cent sample was only barely greater than that for the least squares equations based on the full sample data. Thus the main virtue of the method is that it permits predictions to be made separately for relevant subpopulations (e.g. male‐female) where sample sizes would otherwise be too small to achieve an acceptable degree of accuracy.