December 16, 2015, 10:00 h
Mannheim, Conference room B2,8
A latent variable modeling method for examining the difference between maximal reliability and composite reliability for homogenous multi-component measuring instruments with uncorrelated errors is outlined. The procedure allows point and interval estimation of the discrepancy between the reliability coefficients associated with the optimal linear combination and with the popular unit-weighted (unweighted, overall, simple) sum of the scale components. The approach permits a researcher to make an informed choice if needed between the maximal reliability and composite reliability coefficients and concepts in an empirical setting as indexes of quality of measurement with an instrument under consideration. This choice also informs and allows optimal selection of a scoring scheme for a given measuring instrument. Parallels for the comparison between maximal validity and scale (overall sum score) validity are also drawn. The discussed method is illustrated using numerical data.
About the speaker
Tenko Raykov is Professor of Measurement and Quantitative Methods at Michigan State University, East Lansing, USA, and specializes in a several applied statistical and measurement related areas. These include latent variable and structural equation modeling, reliability and validity evaluation, missing data, longitudinal modeling, scale construction and development, multilevel modeling, survival analysis, and multivariate statistics. He has authored or co-authored more than 100 papers in leading quantitative behavioral science journals, including “Structural Equation Modeling”, “British Journal of Mathematical and Statistical Psychology”, “Journal of Educational and Behavioral Statistics”, “Multivariate Behavioral Research”, “Applied Psychological Measurement”, “Educational and Psychological Measurement”. He is on the editorial boards of most of these journals”, and has co-authored several textbooks in applied statistics and measurement in the behavioral and social sciences.