Introduction to SEM

What is Structural Equation Modeling?

A multivariate technique.

Combines factor analysis and multiple regression.

Examines the relationship between observed (manifest) variable and unobserved (latent) variable.

Examines the relationship between latent variables.

In short, SEM seeks to explain the relationship among multiple variable.

Some Examples of Analyses using SEM

Source: Nawanir, G., Lim, K. T., Othman, S. N., & Adeleke, A. Q. (2018). Developing and validating lean manufacturing constructs: an SEM approach. Benchmarking: An International Journal, 25(5), 1382-1405. doi:doi:10.1108/BIJ-02-2017-0029. Click here.

First & Second Generation Multivariate Data Analysis

First Generation Technique

Can only analyze one layer of linkages between independent and dependent variables at once.

Examples: Linear regression, LOGIT, ANOVA, and MANOVA, etc.

Second generation techniques

Can analyze the relationships among multiple variables simultaneously. This refers to Structural Equation Modeling (SEM).

Comparison between 1st and 2nd Generation Techniques

How many regressions that need to be done in SPSS (Direct Relationships)?

Analytical Trend

In line with this extensive use of multi-item measures, most papers (96%) reported at least some degree of measurement analysis, most typically using SEM techniques.

Confirmatory factor analysis (CFA) results were reported for 86.6% of the studies, and exploratory factor analysis (EFA) results for 22.8%.

For papers that do not focused on scale development, the use of both CFA and EFA is likely unnecessary. In general, use of CFA is preferred.

Google Sites

Report abuse