Learn R Programming

psychmeta (version 2.7.0)

get_stuff: Extract results from a psychmeta meta-analysis object

Description

Functions to extract specific results from a meta-analysis tibble. This family of functions harvests information from meta-analysis objects and returns it as lists or tibbles that are easily navigable.

Available functions include:

  • get_stuff: Wrapper function for all other "get_" functions.

  • get_metatab: Retrieve list of meta-analytic tables.

  • get_ad: Retrieve list of artifact-distribution objects or a summary table of artifact descriptive statistics.

  • get_plots: Retrieve list of meta-analytic plots.

  • get_escalc: Retrieve list of escalc objects (i.e., effect-size data) for use with metafor.

  • get_metafor: Alias for get_escalc.

  • get_followup: Retrieve list of follow-up analyses.

  • get_leave1out: Retrieve list of leave-one-out meta-analyses (special case of get_followup).

  • get_cumulative: Retrieve list of cumulative meta-analyses (special case of get_followup).

  • get_bootstrap: Retrieve list of bootstrap meta-analyses (special case of get_followup).

  • get_metareg: Retrieve list of meta-regression analyses (special case of get_followup).

  • get_heterogeneity: Retrieve list of heterogeneity analyses (special case of get_followup).

  • get_matrix: Retrieve a tibble of matrices summarizing the relationships among constructs (only applicable to meta-analyses with multiple constructs).

Usage

get_stuff(
  ma_obj,
  what = c("metatab", "escalc", "metafor", "ad", "followup", "heterogeneity",
    "leave1out", "cumulative", "bootstrap", "metareg", "matrix", "plots"),
  analyses = "all",
  match = c("all", "any"),
  case_sensitive = TRUE,
  ma_method = c("bb", "ic", "ad"),
  correction_type = c("ts", "vgx", "vgy"),
  moderators = FALSE,
  as_ad_obj = TRUE,
  inputs_only = FALSE,
  ad_type = c("tsa", "int"),
  follow_up = c("heterogeneity", "leave1out", "cumulative", "bootstrap", "metareg"),
  plot_types = c("funnel", "forest", "leave1out", "cumulative"),
  ...
)

get_escalc( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, moderators = TRUE, ... )

get_metafor( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, moderators = TRUE, ... )

get_metatab( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ma_method = c("bb", "ic", "ad"), correction_type = c("ts", "vgx", "vgy"), ... )

get_ad( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ma_method = c("ad", "ic"), ad_type = c("tsa", "int"), as_ad_obj = FALSE, inputs_only = FALSE, ... )

get_followup( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, follow_up = c("heterogeneity", "leave1out", "cumulative", "bootstrap", "metareg"), ... )

get_heterogeneity( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ... )

get_leave1out( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ... )

get_cumulative( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ... )

get_bootstrap( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ... )

get_metareg( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ... )

get_matrix( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, ... )

get_plots( ma_obj, analyses = "all", match = c("all", "any"), case_sensitive = TRUE, plot_types = c("funnel", "forest", "leave1out", "cumulative"), ... )

Value

Selected set of results.

Arguments

ma_obj

A psychmeta meta-analysis object.

what

For the get_stuff() function only: Character scalar telling get_stuff() what to get. All suffixes from functions in the "get_" family can be passed as arguments to what: "metatab", "escalc", "metafor", "ad", "followup", "heterogeneity", "leave1out", "cumulative", "bootstrap", "metareg", "matrix", "plots"

analyses

Which analyses to extract? Can be either "all" to extract references for all meta-analyses in the object (default) or a list containing one or more of the following arguments:

  • construct: A list or vector of construct names to search for.

  • construct_pair: A list of vectors of construct pairs to search for.
    (e.g., list(c("X", "Y"), c("X", "Z"))).

  • pair_id: A list or vector of numeric construct pair IDs (unique construct-pair indices).

  • analysis_id: A list or vector of numeric analysis IDs (unique analysis indexes).

  • k_min: A numeric value specifying the minimum k for extracted meta-analyses.

  • N_min: A numeric value specifying the minimum N for extracted meta-analyses.

match

Should extracted meta-analyses match all (default) or any of the criteria given in analyses?

case_sensitive

Logical scalar that determines whether character values supplied in analyses should be treated as case sensitive (TRUE, default) or not (FALSE).

ma_method

Meta-analytic methods to be included. Valid options are: "bb", "ic", and "ad"

correction_type

Types of meta-analytic corrections to be included. Valid options are: "ts", "vgx", and "vgy"

moderators

Logical scalar that determines whether moderator variables should be included in escalc objects (TRUE; default) or not (FALSE).

as_ad_obj

Logical scalar that determines whether artifact information should be returned as artifact-distribution objects (TRUE) or a summary table of artifact-distribution descriptive statistics (FALSE; default).

inputs_only

Used only if as_ad_obj = TRUE: Logical scalar that determines whether artifact information should be returned as summaries of the raw input values (TRUE) or artifact values that may have been cross-corrected for range restriction and measurement error (FALSE; default).

ad_type

Used only if ma_method = "ic": Character value(s) indicating whether Taylor-series approximation artifact distributions ("tsa") and/or interactive artifact distributions ("int") should be retrieved.

follow_up

Vector of follow-up analysis names (options are: "heterogeneity", "leave1out", "cumulative", "bootstrap", "metareg").

plot_types

Vector of plot types (options are: "funnel", "forest", "leave1out", "cumulative"; multiple allowed).

...

Additional arguments.

Examples

Run this code
if (FALSE) {
## Run meta-analysis:
ma_obj <- ma_r(ma_method = "ic", rxyi = rxyi, n = n, rxx = rxxi, ryy = ryyi,
               construct_x = x_name, construct_y = y_name,
               sample_id = sample_id, citekey = NULL,
               moderators = moderator, data = data_r_meas_multi,
               impute_artifacts = FALSE, clean_artifacts = FALSE)
ma_obj <- ma_r_ad(ma_obj, correct_rr_x = FALSE, correct_rr_y = FALSE)

## Run additional analyses:
ma_obj <- heterogeneity(ma_obj)
ma_obj <- sensitivity(ma_obj, boot_iter = 10, boot_ci_type = "norm")
ma_obj <- metareg(ma_obj)
ma_obj <- plot_funnel(ma_obj)
ma_obj <- plot_forest(ma_obj)

## View summary:
summary(ma_obj)

## Extract selected analyses:
get_metatab(ma_obj)
get_matrix(ma_obj)
get_escalc(ma_obj)
get_bootstrap(ma_obj)
get_cumulative(ma_obj)
get_leave1out(ma_obj)
get_heterogeneity(ma_obj)
get_metareg(ma_obj)
get_plots(ma_obj)
get_ad(ma_obj, ma_method = "ic", as_ad_obj = TRUE)
get_ad(ma_obj, ma_method = "ic", as_ad_obj = FALSE)

## Same extractions as above, but using get_stuff() and the "what" argument:
get_stuff(ma_obj, what = "metatab")
get_stuff(ma_obj, what = "matrix")
get_stuff(ma_obj, what = "escalc")
get_stuff(ma_obj, what = "bootstrap")
get_stuff(ma_obj, what = "cumulative")
get_stuff(ma_obj, what = "leave1out")
get_stuff(ma_obj, what = "heterogeneity")
get_stuff(ma_obj, what = "metareg")
get_stuff(ma_obj, what = "plots")
get_stuff(ma_obj, what = "ad", ma_method = "ic", as_ad_obj = TRUE)
get_stuff(ma_obj, what = "ad", ma_method = "ic", as_ad_obj = FALSE)
}

Run the code above in your browser using DataLab