GROUP.PROFILES: Group Profile Plots

In DFA.CANCOR: Linear Discriminant Function and Canonical Correlation Analysis

Group Profile Plots

Description

Produces profile plots of group means for one or more continuous outcome variables.

Usage

GROUP.PROFILES(data, groups, variables,  
        plot_type ='bar', bar_type = 'all',
        rescale='standardize',
        CI_level= 95, ylim=NULL,
        verbose=TRUE)

Arguments

data	A dataframe where the rows are cases and the columns are the variables.
groups	The name of the groups variable in data, e.g., groups = 'Group'.
variables	The name of the dependent (outcome) variable(s) in data, e.g., variables = c('esteem','anxiety').
plot_type	The options are 'bar' for bar plot, or 'profile' for a lines profile plot.
bar_type	When plot_type = 'bar', the options for bar_type are 'all', for placing the bar plots for all variables in one plot, or 'separate', for placing the bar plots for the variables in separate plots.
rescale	(optional) Should the variables be rescaled into a common metric? The options are 'no' (the default), 'standardize', or 'data', in which case rescaling will be done using the data variables (see the Details below).
CI_level	(optional) The confidence interval for the input, if provided (in whole numbers). The default is 95.
ylim	(optional) Limits for the y-axis, e.g., ylim = c(0, 5). Not implemented when multiple bar plots are requested.
verbose	(optional) Should detailed results be displayed in console? The options are: TRUE (default) or FALSE.

Details

The continuous 'variables' can be rescaled into the same metric, to facilitate interpretation when the means for multiple variables are placed on one plot. The variables can be standardized, or they can be rescaled using the minimum and maximum values in the data variables as the new range for the rescaled variables.

When plot_type = 'bar' and bar_type = 'separate', a maximum of four plots will be produced, for the first four 'variables'.

Value

If verbose = TRUE, the displayed output includes the means, standard deviations, Ns, and confidence intervals for the groups on the variables.

Author(s)

Brian P. O'Connor

Examples

GROUP.PROFILES(data = data_DFA$Ho_2014,
               groups = 'group_1_fac',
               variables = c("fast_ris", "disresp", "sen_seek", "danger"),
               rescale= 'data', 
               plot_type ='bar', 
               bar_type = 'separate')

#first run DFA
DFA_output <- DFA(data = data_DFA$Field_2012, 
                  groups = 'Group', 
                  variables = c('Actions','Thoughts'),
                  predictive = TRUE, 
                  priorprob = 'EQUAL',   
                  covmat_type='separate',
                  verbose = TRUE)

# then produce a profile plot of the group centroids on the discriminant functions
GROUP.PROFILES(data = DFA_output$dfa_scores,
               groups = 'group',
               variables = c('Function.1','Function.2'),
               rescale= 'no',
               plot_type ='profile',
               bar_type = 'separate')

DFA.CANCOR documentation built on June 8, 2025, 11:12 a.m.