rowChisq2Class {scrime}

Rowwise Pearson's ChiSquare Test Based on Tables

Description

Given a set of matrices, each of which represents one group of subjects, and summarizes rowwise the numbers of these observations showing the levels of the categorical variables represented by the rows of the matrices, the value of Pearson's ChiSquare statistic for testing whether the distribution of the variable differs between the different groups is computed for each variable.

Using this function instead of rowChisqStats is recommended when the total number of observations is very large.

Usage

rowChisq2Class(cases, controls, add.pval = TRUE, sameNull = FALSE)

rowChisqMultiClass(..., listTables = NULL, add.pval = TRUE,
   sameNull = FALSE)

Arguments

cases

a numeric matrix in which each row represents one categorical variable and each column one of the levels that the variables exhibit. The entries of this matrix are the numbers of observations from one group (e.g., the cases in a case-control study) showing a particular level at the different variables. Such a matrix can, e.g., be generated by rowTables.The rowwise sums of cases are allowed to differ between variables (which might happen when some of the observations are missing for some of the variables).

controls

a numeric matrix of the same dimensions as cases comprising the numbers of observations from the second group (e.g., the controls in a case-control study) that show the respective level at the different variables. The rows of controls must represent the same variables in the same order as cases, and the columns must represent the same levels in the same order. This matrix can also be generated by employing rowTables. The rowwise sums of controls are allowed to differ between variables (which might happen when some of the observations are missing for some of the variables).

...

numeric matrices (such as cases and controls) each of which comprises the numbers of observations showing the respective levels at the different variables. The dimensions of all matrices must be the same, and the rows and columns must represent the same variables and levels, respectively, in the same order in all matrices.

listTables

instead of inputting the matrices directly into rowChisqMultiClass a list can generated, where each entry of this list is one of matrices, and this list can be used in rowChisqMultiClass by specifying listTables.

add.pval

should p-values be added to the output? If FALSE, only the rowwise values of Pearson's ChiSquare statistic will be returned. If TRUE, additionally the degrees of freedom and the (raw) p-values are computed by assuming approximation to the ChiSquare-distribution, and added to the output.

sameNull

should all variables follow the same null distribution? If TRUE, then all variables must show the same number of levels such that their null distribution is approximated by the same ChiSquare-distribution.

Value

Either a vector containing the rowwise values of Pearson's ChiSquare statistic (if add.pval = FALSE) or a list containing these values (stats), the degrees of freedom (df), and the p-values (rawp) not adjusted for multiple comparisons (if add.pval = TRUE).

Note

In the case of a 2 x 2 table, no continuity correction is applied. In this case, the results of rowChisq2Class and rowChisMultiClass are only equal to the results of chisq.test if in the latter correct = FALSE is used.

The usual contingency table for a variable can be obtained from the matrices by forming a variable-specific matrix in which each row consists of the row of one of these matrices.

Author(s)

Holger Schwender, [e-mail]

See Also

rowChisqStats, rowTables, rowCATTs, rowMsquares

Examples

## Not run: 
# Generate a matrix containing data for 10 categorical 
# variables with levels 1, 2, 3.

mat <- matrix(sample(3, 500, TRUE), 10)

# Now assume that the first 25 columns correspond to
# cases and the remaining 25 columns to cases. Then
# a vector containing the class labels is given by

cl <- rep(1:2, e=25)

# and the matrices summarizing the numbers of subjects
# showing the respective levels at the different variables
# are computed by

cases <- rowTables(mat[, cl==1])
controls <- rowTables(mat[,cl==2])

# To obtain the ChiSquare values call

rowChisq2Class(cases, controls)

# This leads to the same results as

rowChisqStats(mat, cl)

# or

rowChisqMultiClass(cases, controls)

# or

listTab <- list(cases, controls)
rowChisqMultiClass(listTables = listTab)

  
## End(Not run)