fsQCApermTest: A Simple Permutation Test for Type I Error in fsQCA

In QCAfalsePositive: Tests for Type I Error in Qualitative Comparative Analysis (QCA)

Description

A permutation test for fuzzy-set qualitative comparative analysis (fsQCA), designed to calculate the probability of a false positive given the number of hypotheses implicitly tested and the distribution of the data.

Usage

fsQCApermTest(y, configs, total.configs, num.iter = 10000, my.seed = 123,
  adj.method = "holm")

Arguments

y	The outcome variable of interest.
configs	A list of configurations to be tested against y.
total.configs	The total number of configurations used in the original fsQCA analysis. This will generally equal the number of lines in the truth table used for Boolean minimization.
num.iter	The number of iterations to use for the permutation test. Larger numbers of iterations result in more precise p-values.
my.seed	The seed used to generate random numbers.
adj.method	The method used to calculate adjusted p-values (see p.adjust for details).

Value

An object containing the aggregate results of the permutation test as well as the individual permutations.

Examples

data(social.revolutions)
attach(social.revolutions)

intersect <- pmin(breakdown, pop.ins)
intersect2 <- pmin(breakdown, (1-pop.ins))
intersect3 <- pmin((1-breakdown), pop.ins)
intersect4 <- pmin((1-breakdown), (1-pop.ins))

test <- fsQCApermTest(y=soc.rev, configs=list(BI=intersect, Bi=intersect2,
   bI=intersect3, bi=intersect4), total.configs=4)
summary(test)
plot(test)

Example output

Generating permutations for BI...
  Lower-triangular configuration detected...
Generating permutations for Bi...
  Lower-triangular configuration detected...
Generating permutations for bI...
   Upper-triangular configuration detected...
Generating permutations for bi...
   Upper-triangular configuration detected...
Call:
fsQCApermTest(y = soc.rev, configs = list(BI = intersect, Bi = intersect2, 
    bI = intersect3, bi = intersect4), total.configs = 4)

Counterexamples
   Observed Upper Bound Lower c.i.   p-adj se(p-adj)
BI   0.0000     11.0000     4.0000  0.0000    0.0000
Bi   8.0000     15.0000     8.0000  0.1851    0.0868
bI   9.0000     12.0000     6.0000  0.9132    0.0630
bi   7.0000     12.0000     5.0000  0.9132    0.0630

Consistency
   Observed Lower Bound Upper c.i.   p-adj se(p-adj)
BI  1.00000     0.62267    0.86267 0.00000         0
Bi  0.52933     0.45200    0.70400 1.00000         0
bI  0.67336     0.54380    0.82482 1.00000         0
bi  0.61029     0.54044    0.83272 1.00000         0

Total number of configurations: 4 
Number of permutations: 10000 
p-value adjustment method: holm

QCAfalsePositive documentation built on May 2, 2019, 6:40 a.m.