PRACTICAL PROBLEMS OF THE SEM APPLICATION IN SOCIAL SCIENCES – REPORTING RESEARCH RESULTS
PRACTICAL PROBLEMS OF THE SEM APPLICATION IN SOCIAL SCIENCES – REPORTING RESEARCH RESULTS

Summary/Abstract: In this paper we present problems in reporting the results of structural equation modeling (SEM). We present the basic characteristics of structural modeling, from which arises the need for proper presentation of the results. SEM is an umbrella term that covers a broad family of procedures with great potential for modeling complex causal relationships. An overview of studies about problems in the reporting SEM results is given, as well as an overview of recommendations on the essential elements of research reports. Given the increasing use of SEM in the social sciences, it is necessary to establish standards for the application and reporting of results in order to ensure the quality and comparability of results. We highlight the issue of choosing model fit indices and point to influential studies addressing it. The aim of the paper is to point out the most common mistakes in presenting the results and to provide basic guidance on how to avoid them, with reference to studies that can shed some light on certain aspects of the problem. Deficiencies in the reports may indicate deficiencies in the research itself, so it is important to avoid them, in order to allow for a valid interpretation, evaluation and replication of the achieved results by the wider scientific community. In order to avoid common application and reporting errors, an attempt was made to form a checklist of basic criteria for reports for studies utilizing SEM, based on the relevant literature. However, it should be noted that SEM procedures are in the process of further development, and further monitoring of their development and refinement of these criteria is necessary.

• Details
• Contents