Reporting practices in confirmatory factor analysis: An overview and some recommendations.

Abstract

Reporting practices in 194 confirmatory factor analysis studies (1,409 factor models) published in American Psychological Association journals from 1998 to 2006 were reviewed and compared with established reporting guidelines. Three research questions were addressed: (a) how do actual reporting practices compare with published guidelines? (b) how do researchers report model fit in light of divergent perspectives on the use of ancillary fit indices (e.g., L.-T. Hu & P. M. Bentler, 1999; H. W. Marsh, K.-T., Hau, & Z. Wen, 2004)? and (c) are fit measures that support hypothesized models reported more often than fit measures that are less favorable? Results indicate some positive findings with respect to reporting practices including proposing multiple models a priori and near universal reporting of the chi-square significance test. However, many deficiencies were found such as lack of information regarding missing data and assessment of normality. Additionally, the authors found increases in reported values of some incremental fit statistics and no statistically significant evidence that researchers selectively report measures of fit that support their preferred model. Recommendations for reporting are summarized and a checklist is provided to help editors, reviewers, and authors improve reporting practices.

Factor Validation of the Portuguese Version of the Social Skills Scale of the Preschool and Kindergarten Behavior Scales

“…This study sought to overcome the focus on the child's difficulties (behavior problems), and to investigate their potentials (social skills), using CFA methodologies, considered by Jackson et al (2009) as an indispensible statistical tool for studies associated with construct evidence, namely in the development of rating scales. In this sense, this study sought to validate the factor structure of the Portuguese version of the Social Skills scale of the PKBS-2 (Merrell, 2002a), with the difficulty and complexity of definition and conceptualization of the social skills construct, reported in the literature (Gresham, 1986;Merrell & Gimpel, 1998), eventually becoming apparent in the CFA studies.…”

“…Although considered a controversial practice and a target of debate in the scientific community, based on the under-representation of the model or the appropriateness of the use of normative data based on parcels (Little et al, 2002), the use of parcels in CFA studies has been increasingly implemented in this kind of research (Raimundo et al, 2012). Therefore, the most common method of estimation in the literature was used (Jackson et al, 2009), the maximum likelihood, and the procedures suggested in the PKBS-2 manual were followed, by testing the model fit not only for the entire sample but also for Figure 2. Confirmatory factor analysis of the Social Skills scale of the ECIP-2 (Final parcels model).…”

“…Although there is no consensus in the literature regarding the model fit indices to present in CFA studies (Hu & Bentler, 1999;Jackson et al, 2009;Marôco, 2010), the following goodness-of-fit indices were used: Chi-Square (χ 2 ) and the Chi-Square and degrees of freedom ratio (χ 2 /df).…”

“…The Goodness of Fit Index (GFI), the Comparative Fit Index (CFI) and the Root Mean Square Error of Approximation (RMSEA) were also used, with the respective confidence interval (CI) of 90% (Jackson et al, 2009;Marôco, 2010). The cutoff points are also the subject of some disagreement (Jackson et al, 2009).…”

“…The cutoff points are also the subject of some disagreement (Jackson et al, 2009). Therefore, a good model fit can be assumed when the χ 2 has a small value associated with a level of significance higher than .05 and χ 2 /df less than 2 (Marôco, 2010).…”