Meta-Analysis Binary: Criteria
The Criteria dialog provides settings for specifying criteria for meta-analysis with binary outcomes on raw data that are provided in the active dataset for the estimation of the effect size.
- Confidence Interval
- The optional setting specifies the confidence level. The value must be a numeric value between 0 and 100. The default setting is 95.
- Missing Data Scope
- The optional settings control how the procedure handles missing data.
- User Missing Values
- The optional settings control how user missing values are treated.
- Iterations
-
- Maximum iterations
- The optional setting specifies the maximum number of iterations in the iterative methods. The setting is available when iterative methods are used. The value must be a single positive integer. The default value is 100. A value of 0 means that no iterations are performed.
- Maximum step-halvings
- The optional setting specifies the maximum step-halving in the iterative methods. The setting is available when iterative methods are used. The value must be a single positive integer. The default value is 5. A value of 0 means that step-halving is not applied.
- Convergence
- The optional setting specifies the convergence tolerance. The value must be
a single positive value. The setting is available when iterative methods are used. The default value
is
1E-6
.
- Adjust Zero Counts
- The optional settings control how to adjust zero-count data, if applicable,
in the effect size estimation.
- Adjust each value of only those studies containing at least one zero
- The default setting adjusts only zero-count data.
- Adjust each value of all studies only if there exists at least one study containing a zero value
- Adjusts all frequencies when at least one zero count exists in a study.
- Adjust each value of all studies
- Adjusts all data regardless of the existence of zero-counts.
- Make no adjustments and keep zero(s)
- Makes no data adjustment.
- Added Value
- The optional setting specifies the value added to the zero-count data. The default value is 0.5. The specified value must be a single numeric value greater than 0 and less than or equal to 1.
Defining Meta-Analysis Binary criteria
- From the menus choose:
- In the Meta-Analysis Binary dialog, click Criteria.
- Select and define the appropriate criteria settings.
- Click Continue.