BIAS Subcommand (META BINARY command)

The META BINARY command calls the meta-analysis procedure for binary outcomes when the raw data are provided in the active dataset for the estimation of the effect size. The optional BIAS subcommand accesses the publication bias by conducting the regression-based test.

TEST

The required keyword specifies the tests when /BIAS is specified. At least one of EGGER, HARBORD, or PETERS must be specified.

EGGER

Conducts the Egger’s test.

HARBORD

Conducts the Harbord’s test, which is available when /DATA ESTYPE = LOGOR or LOGRR, and is otherwise ignored.

PETERS

Conducts the Peters’ test, which is available when /DATA ESTYPE = LOGOR, and is otherwise ignored.

COVARIATES

The optional keyword specifies the variables that are treated as covariates. Multiple variables are allowed.

FACTORS

The optional keyword specifies the variables that are treated as factors. Multiple variables are allowed.

INTERCEPT

The optional keyword controls the intercept in the regression based test. TRUE is the default setting by default, which includes the intercept term. FALSE excludes the intercept term.

MULTIPLICATIVE

The optional keyword controls the multiplicative model setting. The process is available when /INFERENCE MODEL = FIXED, and is otherwise ignored. FALSE is the default setting, which does not include the dispersion parameter. TRUE introduces the multiplicative dispersion parameter to the analysis.

DISTRIBUTION

The optional keyword controls the distribution used in the regression-based tests. T is the default setting, which estimates the statistics based on the t distribution. When NORMAL is specified, the statistics are estimated based on the normal distribution. The process is available when /INFERENCE ADJUSTSE = NONE, and is otherwise ignored.