Formulae are required for estimating the reliability and validity of scales constructed to measure a single factor, when irrelevant common and specific factors also exist among the data. Equations are here developed for this purpose, to cover scales with single members or many members, and with both weighted and unweighted components: the use of suppressor patterns is also taken into account. Relations to existing formulae, e.g., those of Spearman and Brown, are worked out, and an improved nomenclature is proposed.
No takes yet. Share an insight, caveat, or question.
Cattell et al. (1962) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: