Robustness of normal theory methods in the analysis of linear latent variate models

The structure of the covariance matrix of sample covariances under the class of linear latent variate models is derived using properties of cumulants. This is employed to provide a general framework for robustness of statistical inference in the analysis of covariance structures arising from linear latent variate models. Conditions for normal theory estimators and test statistics to retain each of their usual asymptotic properties under non-normality of latent variates are given. Factor analysis, LISREL and other models are discussed as examples.     

Family system functioning: behavior in the laboratory and the family treatment setting

The Family System Functioning (FSF) scale is a new instrument for measuring dimensions of the intrafamilial environment thought to be important in recovery from major psychiatric illness. Modest statistical correlations were obtained when FSF ratings of laboratory-based family interactions were compared with researcher-guided therapist ratings of FSF based upon the family's behavior in family therapy sessions during the subsequent month. The data from these two settings provide support for the validity of some of the scales. Because of the modest size of the correlations, however, behavior in the laboratory setting may not always be an accurate indicator of how the family will behave in the early weeks of family therapy.     

Hindsight bias: An interaction of automatic and motivational factors?

If subjects are asked to recollect a former response after having been informed about the correct response, their recollection tends to approach the correct response. This effect has been termedhindsight bias. We studied hindsight bias in an experiment requiring numerical responses to almanac-type questions for physical quantities. We varied (1) the time at which the correct information was provided, (2) the encoding of the original responses by asking/not asking subjects to give a reason for the respective response, and (3) the motivation to recall correctly. We found that hindsight is less biased if reasons are given and if the correct information is provided at an earlier time. Motivation had only interactive effects: (1) With high motivation to recall correctly, the time the correct information was provided had no influence. (2) With reasons given, the variation of motivation showed no effect. These results rule out purely motivational and purely automatic explanations.