Hello Everyone
I intend to establish causal pathway using mediation analysis but my data is neither experimenter nor longitudinal, however it is cross-sectional data. The assumption of temporality is violated in my case.
I need your fruitful guidance supported by citations.
Regards
Establishing causality will be difficult if not impossible with cross-sectional, non-experimental data alone. You can only rule out (reject) incorrect models (e.g., if they don't fit and/or if the hypothesized causal paths turn out to be zero or near zero). At least this would tell you which causal processes do not hold.
You can try to account for potential confounding variables statistically (by including them in the model as covariates/statistical control variables). However, this will still not allow you to show causality with the given design with absolute certainty.
Here are two relevant references:
Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12(1), 23-44.
Maxwell, S. E., Cole, D. A., & Mitchell, M. A. (2011). Bias in cross-sectional analyses of longitudinal mediation: Partial and complete mediation under an autoregressive model. Multivariate Behavioral Research, 46(5), 816-841.
- Badges
- Science topic