match.partition: Match most similar pairs for trial generation in a subset of...

Description Usage Arguments Details Examples

Description

This function matches pairs of names based on similarity to create a set of trials. Names are matched, such that each name is only used once and the sum of the distances over all names in total is minimized.

Usage

1

Arguments

split

The name of the rating on which the split should be performed

discard

The percentage of names too close to the median which should be discarded (default: 0, i.e. keep all names)

subset

An optional subset on which the split should be done. If this is left out, the split will be created on all names.

...

Arguments passed on to match.split

s

The split of the names. Only the first two groups will be used, if the split contains more than two groups.

Details

Normally the principal components are used for matching names, but other ratings can also be added and emphasized to create sets of names which are more similar along a special set of ratings.

Note: This function is a shortcut to create the split using partition.names and match the pairs using match.split in a single step.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
# Just match names split on Sex rating
m <- match.partition(Sex)
m

# Match names, but discard some which are ambigous in terms of Sex
m <- match.partition(Sex, discard = 0.2)
m

# First filter unfamiliar and foreign names
s <- filter.names(Familiarity >= 0.5, Nationality >= 0.5)
m <- match.partition(Sex, discard = 0.2, subset=s)
m

# Emphasize on competence and intelligence (weighted 10 times)
m <- match.partition(Sex, discard = 0.2, subset=s, Competence=10, Intelligence=10)
m

aggloeckner/GerNameR documentation built on May 20, 2019, 8:01 p.m.