The Forest Plot tab provides settings for controlling forest plot charts that display in the output for meta-analysis with continuous and binary outcomes when raw data and pre-calculated effect size data are provided in the active data set.
- Forest plot
- The optional setting enables and disables forest plot output. The setting is disabled by default.
- Display Columns
- The optional settings specify the estimated statistics that are
- Effect size
- Includes the estimated effect size.
- Standard Error
- Includes the standard error of the estimated effect size.
- Confidence interval limits
- Includes the confidence interval lower and upper limits of the estimated effect size.
- Includes the p-value of the estimated effect size for each individual study.
- Includes the weight of the primary studies.
- Display exponentiated form
- Note: This setting applies only to meta-analysis with binary outcomes.Controls the plot column and the display of effect size and confidence interval. When the setting is selected, plots the odds ratio, Peto’s odds ratio, or risk ratio, and tabulates the exponential-form statistics. When the setting is not selected, plots the log odds ratio, Peto’s log odds ratio, or log risk ratio, and tabulates the log-transformed statistics.
- Lists the available dataset variables.
- Additional Columns
- Optionally, select variables to display as additional columns. The variable order determines the column display order.
- Optionally, specify a variable upon which the forest plot is sorted. When a variable is specified, the forest plot is sorted in Ascending order (the default setting). When Descending is selected, the forest plot is sorted in descending order.
- Position of plot column
- Controls the placement of the plot column.
- The default setting places the plot column to the right of the other table columns.
- Places the plot column to the left of the other table columns.
- Reference Lines
- Controls the reference lines that are added to the forest plot. Overall effect size adds the line to denote the estimated overall effect size. Null effect size adds the line to denote the null effect size.
- Controls the display of the annotations.
- Prints the homogeneity test statistics.
- Prints the heterogeneity test statistics.
- Prints the test of the overall effect size and the test of between-subgroup homogeneity in the presence of the subgroup analysis.
- Crop Range
- Specifies the crop range. When the setting is selected, two numeric values
must be specified and satisfy
Upper bound > Lower bound.
- For meta-analysis with binary outcomes, when the Effect
Size setting is set to Risk Difference, it must also satisfy
Upper bound ≤ 1and
Lower bound ≥ −1. When the Display exponentiated form setting is selected, it must also satisfy
Lower bound ≥ 0.
Defining forest plot settings
- From the menus choose:
Pre-Calculated Effect Size...or
- In the Meta-Analysis dialog, click Plot and then click the Forest Plot tab.
- Select and define the appropriate forest plot settings.
- Click Continue.