Meta-Analysis Regression: Inference
The Inference dialog provides settings for specifying the estimation methods for meta-analysis regression.
- Include intercept in regression
- The setting is enabled be default.
- Estimate statistics based on t-distribution
- Controls the distribution used in the regression-based tests. The setting is enabled by default, which estimates the statistics based on the t-distribution. When the setting is not selected, the statistics are estimated based on the normal distribution.
- Estimator
- Provides settings for specifying the estimator.
- Restricted maximum likelihood (REML)
- The default setting applies the iterative method and calculates the restricted maximum likelihood estimator.
- Maximum likelihood (ML)
- Applies the iterative method and calculates the maximum likelihood estimator.
- Empirical Bayes
- Applies the iterative method and calculates the empirical Bayes estimator.
- Hedges
- Applies the non-iterative method and calculates the Hedges estimator.
- Hunter-Schmidt
- Applies the non-iterative method and calculates the Hunter-Schmidt estimator.
- DerSimonian-Laird
- Applies the non-iterative method and calculates the DerSimonian-Laird estimator.
- Sidik-Jonkman
- Applies the non-iterative method and calculates the Sidik-Jonkman estimator.
- Standard Error Adjustment
- Provides settings that control whether to apply the Knapp-Hartung
standard-error adjustment.
- No adjustment
- The default setting does not apply the adjustment.
- Apply the Knapp-Hartung adjustment
- Applies the Knapp-Hartung adjustment method.
- Apply the truncated Knapp-Hartung adjustment
- Applies the Knapp-Hartung adjustment method and truncates the value if less than 1 when estimating the variance-covariance matrix.
Defining Meta-Analysis Regression inference settings
- From the menus choose:
- In the Meta-Analysis Regression dialog, click Inference.
- Select and define the appropriate inference settings.
- Click Continue.