Regression discontinuity design: a guide for strengthening causal inference in HRD

Abstract

Purpose Regression discontinuity (RD) design is a sophisticated quasi-experimental approach used for inferring causal relationships and estimating treatment effects. This paper aims to educate human resource development (HRD) researchers and practitioners on the implementation of RD design as an ethical alternative for making causal claims about training interventions. Design/methodology/approach To demonstrate the key features of RD designs, a simulated data set was generated from actual pre-test and post-test diversity training scores of 276 participants from three organizations in the USA. Parametric and non-parametric analyses were conducted, and graphical presentations were produced. Findings This study found that RD design can be used for evaluating training interventions. The results of the simulated data set yielded statistically significant results for the treatment effects, showing a positive causal effect of the training intervention. The analyses found support for the use of RD models with retrospective training intervention data, eliminating ethical concerns from random group assignment. The results of the non-parametric model provided evidence of the plausibility of finding the right balance between precision of estimates and generalizable results, making it an alternative to experimental designs. Practical implications This study contributes to the HRD field by explicating the implementation of a sophisticated, statistical tool to strengthen causal claims, contributing to an evidence-based HRD approach to practice and providing the R syntax for replicating the analyses contained herein. Originality/value Despite the growing number of scholarly articles being published in HRD journals, very few have used experimental or quasi-experimental design approaches. Therefore, a very limited amount of research has been devoted to uncovering causal relationships.

Creating Societal Impact Through Learner Empowerment

“…This outcome aligns with typical findings in HRD literature, which questions the efficacy of employee training. In addition, an array of literature questions whether simply making training available truly supports organizational learning (Chambers, 2016; Kreismann & Talaulicar, 2021; Lancaster & Milia, 2014). However, despite an initial failed attempt to affect the perception through intensive exploration of social issues, we reimagined our course, which ultimately showed an effect on 58% (14 of 24) of the learners’ perceptions of the importance of social issues.…”

The regression discontinuity design: Methods and implementation with a worked example in health services research

“…If the slopes of the regression lines differ (interaction effect) the researcher can also simply calculate the differences of intercepts but needs to shift the assignment variable by the cut-off value to zero (x-c). Another possibility would be to conduct a simple t-test on the groups within the bandwidth b around the cut-off value [c-b; c+b], but the treatment effect then might be underestimated [3,7,16].…”

“…Since the assignment variable can be any continuous measure, many possible uses of the RDD in medicine or health service research are imaginable. For example, Maciejewski & Basu [16] mentioned the use of the RDD in conjunction with the neonatal intensive care unit (NICU). New-borns underweight (<1500g) more often need intensive care then heavier infants.…”

“…The following assumptions are made for the example: HADS T1 will be assignment variable and HADS T2 the outcome variable Patients in the intensive care group improve their HADS (T2) (decreases from at least 2 up to 7 points) Patients in the standard care group improve their HADS (T2) (decreases 0 up to 1 points) For simplification, the HADS can also assume non-integer values between 0 and 42 and underlies a uniform distribution. The pre-defined cut-off value of c = 14.5 and a bandwidth b = [13][14][15][16] is assumed.…”

Guest editorial

“…In the third paper, "Regression Discontinuity Design: A Guide for Strengthening Causal Inference in HRD", Chambers (2016) reviewed studies in a broad set of HRD-related journals and found that regression discontinuity (RD) has been used very infrequently. Chambers demonstrated the features of RD using a simulated dataset generated from published descriptive statistics.…”