Assessing Omitted Confounder Bias in Multilevel Mediation Models
1a School of Psychology , Georgia Institute of Technology.
Abstract:
To draw valid inference about an indirect effect in a mediation model, there must be no omitted confounders. No omitted confounders means that there are no common causes of hypothesized causal relationships. When the no-omitted-confounder assumption is violated, inference about indirect effects can be severely biased and the results potentially misleading. Despite the increasing attention to address confounder bias in single-level mediation, this topic has received little attention in the growing area of multilevel mediation analysis. A formidable challenge is that the no-omitted-confounder assumption is untestable. To address this challenge, we first analytically examined the biasing effects of potential violations of this critical assumption in a two-level mediation model with random intercepts and slopes, in which all the variables are measured at Level 1. Our analytic results show that omitting a Level 1 confounder can yield misleading results about key quantities of interest, such as Level 1 and Level 2 indirect effects. Second, we proposed a sensitivity analysis technique to assess the extent to which potential violation of the no-omitted-confounder assumption might invalidate or alter the conclusions about the indirect effects observed. We illustrated the methods using an empirical study and provided computer code so that researchers can implement the methods discussed.
Insights
Ensuring no omitted confounders is crucial for accurate mediation analysis. This study examines how unaddressed confounders bias multilevel mediation and offers a sensitivity analysis to check results.
Area of Science:
- Multilevel mediation analysis
- Causal inference
- Statistical modeling
Background:
- Valid inference in mediation models requires the absence of omitted confounders, which are common causes of hypothesized relationships.
- Violating the no-omitted-confounder assumption can lead to biased indirect effect estimates and misleading conclusions in mediation analysis.
- While confounder bias is studied in single-level mediation, it remains under-addressed in multilevel mediation.
Purpose of the Study:
- To analytically examine the biasing effects of omitted Level 1 confounders in two-level mediation models with random intercepts and slopes.
- To propose a sensitivity analysis technique for assessing the impact of potential no-omitted-confounder assumption violations on indirect effect conclusions.
- To provide practical tools and illustrations for researchers conducting multilevel mediation analysis.
Main Methods:
- Analytical examination of bias due to omitted Level 1 confounders in a two-level mediation model.
- Development and application of a sensitivity analysis technique to evaluate assumption violations.
- Illustration with an empirical study and provision of accompanying computer code.
Main Results:
- Omitting a Level 1 confounder can significantly bias estimates of both Level 1 and Level 2 indirect effects in multilevel mediation.
- The proposed sensitivity analysis quantifies the potential impact of omitted confounders on observed indirect effects.
- The methods are demonstrated to be applicable in real-world research scenarios.
Conclusions:
- The no-omitted-confounder assumption is critical and untestable in multilevel mediation, necessitating sensitivity analyses.
- Researchers should be aware of potential confounding at Level 1 and its impact on indirect effects.
- The proposed sensitivity analysis technique offers a valuable tool for enhancing the robustness of multilevel mediation findings.
More Related Videos
Related Concept Videos
01:27Confounding in Epidemiological Studies
01:25Strategies for Assessing and Addressing Confounding
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
01:29Bias in Epidemiological Studies
01:21Friedman Two-way Analysis of Variance by Ranks
01:31Confirmation Biases
01:09Truncation in Survival Analysis
Left truncation occurs when individuals who experienced the event of interest before a certain time are not included in the study. This is often due to a "delayed entry" into the study where only those who survive until a certain entry point are...