Meta-Analysis of Single-Case Experimental Design using Multilevel Modeling.

Baek et al. (2023) wrote a how-to guide. They show step-by-step ways to pool single-case graphs with multilevel models.

The paper gives code, software tips, and worked examples. No new data were collected.

The authors did not report new effect sizes. They simply show that multilevel modeling can handle the nested shape of SCED data.

Jamshidi et al. (2018) checked 30 years of SCED meta-analyses. They found most still skip bias tests and misuse old synthesis rules. The new guide answers that gap by giving clearer, modern rules.

Moeyaert et al. (2023) listed the moderators most meta-analyses use: age, diagnosis, gender, with lots of missing values. Their list helps you pick predictors when you follow the new modeling steps.

Vassos et al. (2023) warn that about 1 in 14 SCED graphs starts with a slow change that can hide real effects. The multilevel approach can model that transition phase instead of ignoring it.

If you review single-case work, you now have a ready script. Plug your graphs into the multilevel code, add the moderators Mariola lists, and run bias checks that Laleh says are missing. You get a tighter pooled effect and stronger evidence for your next treatment brief.