Bayesian t-test uses frequentist data reader

[JohnnyDoorn]

Fix jasp-stats/jasp-issues#2102

[vandenman]

The changes look fine but I don't understand why the unit tests fail..., let's see if they also fail in #262, otherwise, I'll try to reproduce this locally. Those tests don't run because no R code changed 🙃

[JohnnyDoorn]

Its just some rng thing no?

[vandenman]

It's the same test on each OS, and the error is the same for macOS and Ubuntu but different on Windows. I just want to make sure it's got nothing to do with the specific stuff for handling ordinal input.

[vandenman]

and fix the test if I can :)

[vandenman]

jaspTTests/tests/testthat/test-ttestbayesianindependentsamples.R

Lines 181 to 187 in efccd74

	# ???: the error statistic differs between osx <-> windows. if anyone can figure out why i'd be interested (especially because the BF is the same)
	# vandenman: The difference appears in integrate, for example:
	# BayesFactor:::meta.bf.interval(-Inf, Inf, t = 0.158315759266202, N = 50000, df = 199998, rscale = 0.707)
	if (identical(.Platform$OS.type, "windows"))
	jaspTools::expect_equal_tables(table, list(0.00511047754408505, 4.61848285988607, "dependent_var"))
	else
	jaspTools::expect_equal_tables(table, list(0.00511048079567079, 0.185588508623007, "dependent_var"))

Always fun to encounter your past self in some obscure comment somewhere...

[JohnnyDoorn]

Haha awesome, then we get it right for at least 1 OS ^^