designmatch (version 0.1.1)

meantab: Tabulate means of covariates after matching

Description

Function for tabulating the means and other basic statistics useful to assess covariate balance after matching.

Usage

meantab(X_mat, t_ind, t_id, c_id, exact = NULL, digits = 2)

Arguments

X_mat

matrix of covariates: a matrix of covariates used to assess balance.

t_ind

treatment indicator: a vector of zeros and ones indicating treatment (1 = treated; 0 = control).

t_id

a vector of indexes of the treated units.

c_id

a vector of indexes of the matched controls.

exact

a vector of characters equal to "f" or "w" indicating whether Fisher's exact test or Wilcoxon rank-sum test should be used for binary (or categorical) and continous covariates, respectively. Otherwise, if exact exact = NULL, simple t-tests are used. The default is exact = NULL. If exact != NULL, the length of exact has to be equal to the number of columns of X_mat.

digits

a scalar indicating the number of digits to display in the columns of the table.

Value

A table with the following columns:

Mis

proportion of missing values for each covariate;

Min

minimum value for each covariate;

Max

maximum value for each covariate;

Mean T

mean of the treated units for each covariate;

Mean C

mean of the matched controls for each covariate;

Std Dif

standardized differences in means after matching for each covariate;

P-val

P-values for t-tests for differences in means between treated units and matched controls for each covariate.

Details

meantab is a function for tabulating the means and other basic statistics useful to assess covariate balance after matching.

Examples

	# Load data
	data(germancities)

	# Sort and attach data
	germancities = germancities[order(germancities$treat, decreasing = TRUE), ]
	attach(germancities)

	# Treatment indicator
	t_ind = treat
	
	# Indexes of the treated units
	t_id = which(t_ind == 1)

	# Matrix of covariates
	X_mat = cbind(log2pop, popgrowth1939, popgrowth3339, emprate, indrate, rubble, 
	rubblemiss, flats, flatsmiss, refugees)

	# Indices of the matched controls (obtained using bmatch in designmatch)
	c_id = c(67, 75, 39, 104, 38, 93, 79, 59, 64, 55, 106, 99, 97, 61, 82, 57, 
	76, 47, 46, 49)
 
 	# meantab
	meantab(X_mat, t_ind, t_id, c_id)
	
	# meantab
	meantab(X_mat, t_ind, t_id, c_id, exact = c(rep("w", 6), "f", "w", "f", "w"), digits = 3)