1% Generated by roxygen2: do not edit by hand
2% Please edit documentation in R/rma-tidiers.R
3\name{glance.rma}
4\alias{glance.rma}
5\title{Glance at a(n) rma object}
6\usage{
7\method{glance}{rma}(x, ...)
8}
9\arguments{
10\item{x}{An \code{rma} object such as those created by \code{\link[metafor:rma.uni]{metafor::rma()}},
11\code{\link[metafor:rma.uni]{metafor::rma.uni()}}, \code{\link[metafor:rma.glmm]{metafor::rma.glmm()}}, \code{\link[metafor:rma.mh]{metafor::rma.mh()}},
12\code{\link[metafor:rma.mv]{metafor::rma.mv()}}, or \code{\link[metafor:rma.peto]{metafor::rma.peto()}}.}
13
14\item{...}{Additional arguments. Not used. Needed to match generic
15signature only. \strong{Cautionary note:} Misspelled arguments will be
16absorbed in \code{...}, where they will be ignored. If the misspelled
17argument has a default value, the default value will be used.
18For example, if you pass \code{conf.lvel = 0.9}, all computation will
19proceed using \code{conf.level = 0.95}. Additionally, if you pass
20\code{newdata = my_tibble} to an \code{\link[=augment]{augment()}} method that does not
21accept a \code{newdata} argument, it will use the default value for
22the \code{data} argument.}
23}
24\description{
25Glance accepts a model object and returns a \code{\link[tibble:tibble]{tibble::tibble()}}
26with exactly one row of model summaries. The summaries are typically
27goodness of fit measures, p-values for hypothesis tests on residuals,
28or model convergence information.
29
30Glance never returns information from the original call to the modeling
31function. This includes the name of the modeling function or any
32arguments passed to the modeling function.
33
34Glance does not calculate summary measures. Rather, it farms out these
35computations to appropriate methods and gathers the results together.
36Sometimes a goodness of fit measure will be undefined. In these cases
37the measure will be reported as \code{NA}.
38
39Glance returns the same number of columns regardless of whether the
40model matrix is rank-deficient or not. If so, entries in columns
41that no longer have a well-defined value are filled in with an \code{NA}
42of the appropriate type.
43}
44\examples{
45
46library(metafor)
47
48df <-
49  escalc(
50    measure = "RR",
51    ai = tpos,
52    bi = tneg,
53    ci = cpos,
54    di = cneg,
55    data = dat.bcg
56  )
57
58meta_analysis <- rma(yi, vi, data = df, method = "EB")
59
60glance(meta_analysis)
61}
62\value{
63A \code{\link[tibble:tibble]{tibble::tibble()}} with exactly one row and columns:
64  \item{cochran.qe}{In meta-analysis, test statistic for the Cochran's Q_e test of residual heterogeneity.}
65  \item{cochran.qm}{In meta-analysis, test statistic for the Cochran's Q_m omnibus test of coefficients.}
66  \item{df.residual}{Residual degrees of freedom.}
67  \item{h.squared}{Value of the H-Squared statistic.}
68  \item{i.squared}{Value of the I-Squared statistic.}
69  \item{measure}{The measure used in the meta-analysis.}
70  \item{method}{Which method was used.}
71  \item{nobs}{Number of observations used.}
72  \item{p.value.cochran.qe}{In meta-analysis, p-value for the Cochran's Q_e test of residual heterogeneity.}
73  \item{p.value.cochran.qm}{In meta-analysis, p-value for the Cochran's Q_m omnibus test of coefficients.}
74  \item{tau.squared}{In meta-analysis, estimated amount of residual heterogeneity.}
75  \item{tau.squared.se}{In meta-analysis, standard error of residual heterogeneity.}
76
77}
78