The WRMR and RMSEA indicators and chi-square statistics in the evaluation
of goodness-of-fit in SEM models based on ordinal data Cover Image

Wskaźniki WRMR i RMSEA oraz statystyka chi-kwadrat w ocenie dobroci dopasowania modeli SEM dla danych porządkowych
The WRMR and RMSEA indicators and chi-square statistics in the evaluation of goodness-of-fit in SEM models based on ordinal data

Author(s): Piotr Tarka
Subject(s): Economy
Published by: Wydawnictwo Uniwersytetu Ekonomicznego we Wrocławiu
Keywords: structural equation models (SEM); ordinal variabels; estimation; goodness-of-fit in SEM; WLMR and RMSEA indices

Summary/Abstract: In the article, the author discusses the problems of evaluation of goodness-of-fit in structural equation models (SEM) from the perspective of ordinal categorical variables. On their background a SEM model with limited information is considered and the estimation method DWLS(M-V) is diagnosed. Next two indices WRMR and RMSEA are set in comparison with statistic chi-square to observe the efficiency of the model fit. As an example illustrating the strategy for the appropriate dealing with data measured on ordinal scale, we refer to empirical data on researchers and analysts’ perception (in companies) of marketing research effectiveness under the influence of specific organizational conditions.

  • Issue Year: 2017
  • Issue No: 469
  • Page Range: 197-207
  • Page Count: 11
  • Language: Polish
Toggle Accessibility Mode