Issues in Formative Measurement

Not Merely a Matter of Drawing Arrows: The Empirical Consequences of Measurement Model Specification and Recommendations for Practice

“…For reflective measures, the indicators are viewed as a manifestation of the underlying construct (i.e., the latent construct causes the measure; Schwab, 1980); for formative measures, indicators are seen as a cause of the underlying variable (MacCallum & Browne, 1993). While the nature of this construct-indicators relationship should be determined by theory, there is plentiful evidence across fields where researchers have expressed different opinions as to whether even some well-established measures of theoretical constructs should have formative or reflective indicators (Alvarez & Asugman, 2009;Diamantopoulos & Winklhofer, 2001;Howell et al, 2007;Jarvis et al, 2003;Murphy & Hofacker, 2009). The issue of formative versus reflective measures is a particularly salient issue in hospitality research for two key reasons.…”

Not Merely a Matter of Drawing Arrows: The Empirical Consequences of Measurement Model Specification

“…Methodological papers focused on, for example, management research may not necessarily inform hospitality research on a marketing topic. Second, constructs often used in hospitality research seem to be some of the most controversial constructs in the formative versus reflective measurement debate (Alvarez & Asugman, 2009;Murphy & Hofacker, 2009). Measures like SERVQUAL (Parasuraman, Zeithaml, & Berry, 1988) have been the focus of attention as to whether they are best represented as formative or reflective (cf.…”