# correlate: Correlation In uwIntroStats: Descriptive Statistics, Inference, Regression, and Plotting in an Introductory Statistics Course

## Description

Computes correlation matrix for an arbitrary number of numeric variables, optionally within strata.

## Usage

 ```1 2 3``` ```correlate(..., strata = NULL, subset = NULL, conf.level = 0.95, use = "pairwise.complete.obs", method = "pearson", stat = "cor", byStratum = TRUE, version = FALSE) ```

## Arguments

 `...` an arbitrary number of variables for which a correlation matrix is desired. The arguments can be vectors, matrices, or lists. Individual columns of a matrix or elements of a list that are not of class `numeric`, `factor`, or `Date` will be omitted. Factor and Date variables are converted to integers. Character vectors will be coerced to numeric. Variables must all be of the same lengths. `strata` vector, matrix, or list of stratification variables. Descriptive statistics will be computed within strata defined by each unique combination of the stratification variables, as well as in the combined sample. If `strata` is supplied, all variables must be of that same length. `subset` vector indicating a subset to be used for all descriptive statistics. If `subset` is supplied, all variables must be of that same length. `conf.level` a numeric scalar between 0 and 1 denoting the confidence level to be used in constructing confidence intervals for the correlation. `use` character string denoting the cases to use: `"everything"` uses all cases (and causes NA when any needed variable is missing), `"complete.obs"` uses only those rows with no missing data for any variable, and `"pairwise.complete.obs"` computes pairwise correlations using all cases that are not missing data for the relevant variables. `method` character string denoting the correlation method to use: `"pearson"` denotes Pearson's correlation coefficient and `"spearman"` denotes Spearman's rank correlation. `stat` a vector of character strings indicating the descriptive statistic(s) to be tabulated. Possibilities include any statistic as specified by one or more of `"cor"`, `"n"`, `"t.stat"`, `"pval"`, `"loCI"`, or `"hiCI"`. Only enough of the string needs to be specified to disambiguate the choice. Alternatively (and more usefully), a single special format character string can be specified as described in the Details below. `byStratum` a logical scalar indicating whether statistics should be grouped by pair of variables. If `TRUE`, the results will be displayed in a series of tables where each table correspond to a single variable, with rows corresponding to different strata and columns reflecting all other variables. If `FALSE`, the results will be displayed in a series of tables where each table corresponds to a single stratum and rows and columns reflect the variables. `version` if `TRUE`, the version of the function will be returned. No other computations will be performed.

## Value

An object of class uCorrelate is returned, which consists of a list of correlation estimates and inference for each specified stratum and for the combined dataset. Each element of the list has six arrays:

 `cormtx` the correlation matrix, printed. `n` matrix of sample sizes used to compute each correlation `t.stat` matrix of t-statistics, testing a correlation of 0. `pval` matrix of two-sided p-values for the t-test. `lo95%CI` lower bound of the 95% confidence interval. `hi95%CI` upper bound of the 95% confidence interval.

## Author(s)

Scott S. Emerson, M.D., Ph.D., Andrew J. Spieker, Brian D. Williamson, Travis Y. Hee Wai, and Solomon Lim

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23``` ```# Load required libraries library(survival) # Reading in a dataset mri <- read.table("http://www.emersonstatistics.com/datasets/mri.txt",header=TRUE) # Estimated correlation matrix using all data, complete cases, or pairwise complete (the default) with (mri, correlate(age,weight,ldl,use="everything")) with (mri, correlate(age,weight,ldl,use="complete")) with (mri, correlate(age,weight,ldl)) # Correlation matrices for each stratum with (mri, correlate(age,weight,ldl,strata=male)) # Correlations grouped by variable with (mri, correlate(age,weight,ldl,strata=male,byStratum=FALSE)) # Special formatting of inference for correlations within strata with (mri, correlate(age,weight,ldl,strata=male,stat="@[email protected] (@[email protected], @[email protected]); P @[email protected]; n= @[email protected]")) # Special formatting of inference for correlations grouped by variable with (mri, correlate(age,weight,ldl,strata=male,stat="@[email protected] (@[email protected], @[email protected]); P @[email protected]; n= @[email protected]", byStratum=FALSE)) ```

uwIntroStats documentation built on Oct. 10, 2018, 5:04 p.m.