What is it about?
A simulation technique is a useful method in detecting a selection bias problem, and yet the method has not been widely applied to public management research. This study showed how selection bias can increase around a cut point and how a modeling mechanism can improve the estimation of each service effect by utilizing a Monte Carlo simulation.
Photo by Priscilla Du Preez on Unsplash
Why is it important?
Given the wide usability of survey studies in public management research, the misestimation of what the studies are supposed to measure has disastrous potential. Although researchers can set models that provide the causal relations between substantive variables of interest, these variables are frequently unapproachable to direct measures (Bollen and Pearl, 2013). Our vulnerability lies in the development of measurement tools for data types that do not accurately capture the object of interest to be measured and the continuation of its practice. Measurement modeling research used to capture the nuance of the misestimated care effect is largely an untapped area for public management research. One objective of this study is to contribute to these efforts to improve causal inference by identifying potential sources of selection bias due to the types of questions asked and the placement changes of service users. In this context, the purpose of this study is to develop an alternative modeling technique to capture the nuance of the self-selection problem at the cut point.
Read the Original
This page is a summary of: Simulating self-selection in public management research: implications from caseworker discretion in the child welfare system, International Journal of Public Sector Management, August 2021, Emerald, DOI: 10.1108/ijpsm-10-2020-0292.
You can read the full text:
The following have contributed to this page