<Order DETECT or Books> <Home Page> <Info. Page> <Contact us: Info@levelsofanalysis.com> <Return to list of comparisons>

Comparison of WABA and DETECT to Structural Equation Modeling (SEM)

Description of Structural Equation Modeling

Heck and Thomas (2000) provide an overview of SEM as it applies to multilevel Modeling. They describe the two key features of SEM: Confirmatory factor analysis and Structural equation models. They do so for both the traditional and multilevel approaches to SEM.

Compatibility of Within and Between Analysis (WABA/DETECT) with Structural Equation Modeling

The traditional uses of SEM tie in easily with WABA and DETECT. Once the levels of analysis are identified by WABA as appropriate, all of SEM can be used. For example, as Mumford (1998, p 426) points out "Essentially, Dansereau et al. (1998) propose a two stage approach, WABA is used to identify appropriate levels of analysis. In the second stage, formal causal modeling analyses are conducted at the appropriate levels of analysis." In the case described, dyads were examined. More recent approaches to multi-level models include the use of within-group and between-group matrices. It remains possible to use WABA to test the plausibility of the levels input to SEM.

Responding to Misconceptions and misunderstandings about WABA and DETECT

Statement #1. WABA does not allow causal or longitudinal modeling

Response: From the points made above WABA involves a two step process. It does not do causal modeling in the sense that SEM does. SEM computer packages would be required. WABA does allow longitudinal analyses that allow entities to change over time. Most longitudinal modeling assumes that the entities are stable but that variables change. (See the tutorial about longitudinal analysis and a paper by Dansereau, Yammarino, & Kohles 1999))

References

Dansereau, F., Yammarino, F., and Kohles, J. (1999). Multiple-levels of analysis from a longitudinal perspective. Academy of Management Review, 24, 346-357.

Dansereau, F., Yammarino, F., Markham S., Alutto J., Newman J., Dumas M., Nachman S., Naughton, T, Kim, K, Al-Kelabi, A., Lee S., and Keller, T.(1995). Individualized leadership: A new multiple level approach. Leadership Quarterly, 6, 413-450..

Heck, R., & Thomas, S. (2000) An introduction to Multilevel modeling techniques. Mahwah, NJ: Lawrence Erlbaum.

Mumford, M. (1998) Situations, interactions, and followers: The case of individualized leadership. In F. Dansereau and F. Yammarino (Eds.) Leadership: The multiple level approaches (Part B). Stamford CT: JAI Press (pp.421-427).

<Order DETECT or Books> <Home Page> <Info. Page> <Contact us: Info@levelsofanalysis.com> <Return to list of comparisons>