Article

A method for testing group differences of scale validity in multiple population studies.

Department of Psychology, Fordham University, Bronx, NY 10458, USA.
British Journal of Mathematical and Statistical Psychology (Impact Factor: 1.26). 06/2005; 58(Pt 1):173-84. DOI: 10.1348/000711005X38005
Source: PubMed

ABSTRACT A method for testing equality in validity of multi-component measuring instruments across populations is outlined. The approach is developed within the framework of covariance structure modelling and complements earlier research on examining group differences in scale reliability. The procedure is particularly useful for purposes of ascertaining comparability of validity when constructing and developing measuring instruments. The method also provides ranges of plausible values for differences in composite validity across several populations and allows one to evaluate group discrepancies in validity of behavioural scales. The approach is illustrated using data from a cognitive intervention study.

0 Bookmarks
 · 
53 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: A method of composite reliability estimation using covariance structure analysis with nonlinear constraints is outlined. To motivate the developments, initially a short overview of research is presented, demonstrating that in many cases the widely used coefficient alpha is an unsatisfactory index of scale reliability already at the population level. As an alternative, the proposed covariance structure analysis procedure is based on the theoretical formula of the scale reliability coefficient in terms of parameters pertaining to a given set of congeneric components. The described approach is illustrated with several numerical examples and its performance compared with that of coefficient alpha.
    British Journal of Mathematical and Statistical Psychology 12/2001; 54(Pt 2):315-23. · 1.26 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A covariance structure modelling method for examining group differences in scale reliability of multi-component measuring instruments is proposed. The procedure is based on a set of appropriate parameter constraints imposed in a multiple-population structural model. Unlike tests of group differences in coefficient alpha that in general do not evaluate the discrepancies in scale reliability coefficients, the described approach permits one to examine differences in the latter quantities of actual interest when concerned with questions pertaining to reliability of composite scores. The method can be used to test whether a given measuring instrument has identical reliabilities in studied populations, and/or whether distinct instruments, or modes of administration of the same instrument, have equal reliabilities across groups. The proposed procedure is illustrated with a pair of examples.
    British Journal of Mathematical and Statistical Psychology 06/2002; 55(Pt 1):145-58. · 1.26 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Fluid intelligence belongs to that cluster of intellectual abilities evincing aging loss. To examine further the range of intellectual reserve available to aging individuals and the question of replicability in a new cultural and laboratory setting, 204 healthy older adults (mean age = 72 years; range = 60-86) participated in a short-term longitudinal training study. For experimental subjects, 10 sessions consisted of cognitive training involving two subability tests (Figural Relations, Induction) of fluid intelligence. The pattern of outcomes replicates and expands on earlier studies. Older adults have the reserve to evince substantial increases in levels of performance in fluid intelligence tests. Transfer of training, however, is narrow in scope. Training also increases accuracy of performance and the ability to solve more difficult test items. Difficulty level was estimated in a separate study, with a comparable sample of N = 112 elderly adults. Future research is suggested to examine whether intellectual reserve extends to near-maximum levels of performance.
    Psychology and Aging 07/1986; 1(2):172-7. · 2.73 Impact Factor