Exploratory Mediation Analysis via Regularization

Abstract

Exploratory mediation analysis refers to a class of methods used to identify a set of potential mediators of a process of interest. Despite its exploratory nature, conventional approaches are rooted in confirmatory traditions, and as such have limitations in exploratory contexts. We propose a two-stage approach called exploratory mediation analysis via regularization (XMed) to better address these concerns. We demonstrate that this approach is able to correctly identify mediators more often than conventional approaches and that its estimates are unbiased. Finally, this approach is illustrated through an empirical example examining the relationship between college acceptance and enrollment.

Keywords:

• exploratory mediation analysis
• lasso
• mediation
• regularization

FUNDING

Sarfaraz Serang was supported by funding from the National Institute on Aging, Grant Number 3R37AG007137. Ross Jacobucci was supported by funding through the National Institute on Aging, Grant Number T32AG0037. Kim C. Brimhall was funded by U.S. Department of Health and Human Services Agency for Healthcare Research and Quality Grant Number 1R36HS024650-01. Kevin J. Grimm was funded by National Science Foundation Grant REAL-1252463 awarded to the University of Virginia, David Grissmer (Principal Investigator), and Christopher Hulleman (Co-Principal Investigator).

Notes

¹ This two-stage approach has also been advocated as post-selection inference (Chernozhukov, Hansen, & Spindler, 2015; Lee, Sun, Sun, & Taylor, 2016).

² Information criteria are fully available with RegSEM and are calculated as usual.

³ Although percent bias is typically defined as the negative of what we use here, we believe our definition to be more useful in this specific context given that Stage 1 estimates always underestimate population effects.

Additional information

Funding

Sarfaraz Serang was supported by funding from the National Institute on Aging, Grant Number 3R37AG007137. Ross Jacobucci was supported by funding through the National Institute on Aging, Grant Number T32AG0037. Kim C. Brimhall was funded by U.S. Department of Health and Human Services Agency for Healthcare Research and Quality Grant Number 1R36HS024650-01. Kevin J. Grimm was funded by National Science Foundation Grant REAL-1252463 awarded to the University of Virginia, David Grissmer (Principal Investigator), and Christopher Hulleman (Co-Principal Investigator).