Propensity score analysis: an alternative statistical approach for HRD researchers

Abstract

Purpose This paper aims to introduce matching in propensity score analysis (PSA) as an alternative statistical approach for researchers looking to make causal inferences using intact groups. Design/methodology/approach An illustrative example demonstrated the varying results of analysis of variance, analysis of covariance and PSA on a heuristic data set. The three approaches were compared by results and violations of statistical assumptions. Findings Through the illustrative example, it is demonstrated how different statistical approaches can produce varied results. Only PSA mitigated pre-existing group differences without violating the assumption of independence. Originality/value This paper attempts to answer calls in the literature for more robust statistical methodologies to better inform human resource development practice and theory.

Effects of Medicare wellness visits on health promotion outcomes

“…Selection bias was controlled by using propensity score matching (PSM) to apply a mathematical scoring system to each case group member, then mathematically aligning each case group member with a similarly matched control group member based on selected characteristics (Keiffer & Lane, 2016; Rosenbaum, 2004; Rubin & Thomas, 1992). The PSM approach ensured that both groups were balanced, with covariation controlled (Keiffer & Lane, 2016; Rosenbaum, 2004; Rubin & Thomas, 1992). Matching methods increased internal validity by allowing homogeneity between the case and control groups (Austin, 2011).…”

“…Sample exclusion criteria included individuals not enrolled in Medicare, and attributed as nonactive for primary care services, or diagnosed with dementia. Selection bias was controlled by using propensity score matching (PSM) to apply a mathematical scoring system to each case group member, then mathematically aligning each case group member with a similarly matched control group member based on selected characteristics (Keiffer & Lane, 2016; Rosenbaum, 2004; Rubin & Thomas, 1992). The PSM approach ensured that both groups were balanced, with covariation controlled (Keiffer & Lane, 2016; Rosenbaum, 2004; Rubin & Thomas, 1992).…”

Guest editorial

“…In the fifth paper, "Propensity Score Analysis: An Alternative Statistical Approach for HRD Researchers", Keiffer and Lane (2016) presented propensity score analyses and illustrated the technique using an example relevant to HRD. They also contributed R syntax that readers can use to replicate the propensity score analyses.…”