describe.r: Descriptive Statistics for Correlations

In multicon: Multivariate Constructs

Description

Returns the descriptive statistics for a vector, matrix, or data.frame of correlation coefficients stored in x by using fisher's r to z transformation, computing the values, and then back tranforming the values using fisher's z to r transofrmation.

Usage

describe.r(x, na.rm = TRUE, tr = 0.2, type = 3)

Arguments

x	A vector, matrix, or data.frame object of correlation coefficients.
na.rm	A logical indicating whether NA values should be removed before calculations are done.
tr	A numeric element between .00 and .50 specifying the amount of trimming to be done for the calculation of the trimmed mean.
type	Which estimate of kurtosis should be used? See the describe function in the 'psych' package.

Details

The psych package function describe computes a number of descriptive statistics for ordinary data. However, correlation coefficients are typically r-to-z transformed before computing such statistics. This function makes getting the descriptive statistics for correlation coefficients easy by doing such transformations.

Value

A data.frame of descriptive statistics: item name, item number, number of valid cases, mean, standard deviation, median, trimmed mean, mad: median absolute deviation (from the median), minimum, maximum, skew, kurtosis, standard error.

Author(s)

Ryne A. Sherman

See Also

describe

Examples

data(caq)
data(acq.comp)
mycors <- Profile.r(caq, acq.comp) # Get profile agreement correlations
describe.r(mycors)

Example output

Loading required package: psych
Loading required package: abind
Loading required package: foreach
  var   n miss      mean        sd    median   trimmed       mad         min
1   1 205    0 0.4661666 0.2244845 0.4733151 0.4683826 0.2321864 -0.04131409
        max     range        skew   kurtosis         se
1 0.8106831 0.8243863 -0.01745555 -0.3504475 0.01594895

multicon documentation built on May 2, 2019, 3:18 a.m.