Either linear or quadratic rules may be used to derive classification equations in discriminant analysis for the purpose of predicting group membership. Generally, the decision about which rule to use is governed by the degree to which the separate group covariance matrices are unequal. An example is presented that supports the superior internal…

The use of multivariate statistics in the social sciences is essential given that the studied effects are seldom influenced by only a single variable, and instead are usually influenced by multiple variables. One multivariate technique, discriminant function analysis, has potential for many applications in the social sciences. Discriminant…

This article develops a model for potential working-class applicants to university, using multinomial logit regression to analyse a national UK sample, drawn from lower middle and working-class adults aged 16-30. Influential factors are belief in ability, preference for university rather than earning money, qualifications and family encouragement;…

The purpose of this study was to investigate the practicality of multiple discriminant function analysis for deriving item response weights. Item responses on a job satisfaction questionnaire administered to 219 professional workers and 242 semi-skilled customer workers were analyzed. Discriminant functional analysis was conducted on the total…