fpAnova {fixedpointproperty}

Function to perform ANOVA on fp objects

Description

This function computes Bayes Factors and p-values for within-subjects ANOVA designs, encoded as fp objects.

Usage

fpAnova(object, stat = "BF", na.rm = TRUE, check = TRUE)

Arguments

object

a list of objects from class fpp.

stat

Either "BF" (default), "p", or "both", specificying what statistic to report.

na.rm

Are NAs removed?

check

Should the data be checked for suitability? A warning will be provided if a check is failed.

Details

The function expects the output of fpGet, but in a list.

Value

A list containing the results of either the Bayesian or frequentist analysis, or both:

BF

The output of anovaBF

p

The output of summary.aov

Warning

If check=TRUE, then warnings will be provided if the data are not suitable for correct inferences.

Author(s)

Leendert van Maanen ([e-mail])

References

Van Maanen, L., De Jong, R., Van Rijn, H (2014). How to assess the existence of competing strategies in cognitive tasks: A primer on the fixed-point property. PLOS One, 9, e106113

Van Maanen, L. Couto, J. & Lebetron, M. (2016). Three boundary conditions for computing the fixed-point property in binary mixture data. PLOS One, 11, e0167377.

See Also

fpGet, anovaBF, fpConditionCheck, summary.aov.

Examples

## generate data
p <- c(.1,.5,.9)
rt <- sapply(1:3, function(i) {rnormMix(10000, c(1,2), c(1,1), p[i])})
dat <- data.frame(rt=c(rt), cond=rep(1:3, each=10000), pp=rep(1:50, each=200, times=3))

## compute the list of fpp objects
res <- tapply(1:nrow(dat), dat$pp, function(X) {fpGet(dat[X,], 1000, bw=.75)})

## call fpAnova, with stat="both" to do both a Bayesian and a frequentist test
fpAnova(res, stat="both")