# selectiontask

##### Wason Selection Task (WST) and Helpful Hints

In the Wason selection task, a participant is presented with four cards, each one having a letter side and a number side, e.g., A B 3 4. The task is to select the card(s) that have to be turned around in order to test the rule "If there is an A on the letter side then there is a 3 on the number side." Klauer, Stahl, and Erdfelder (2007) report a series of experiments to test their WST model using the aggregate frequencies of the 16 possible response patterns.

- Keywords
- datasets

##### Usage

`data(selectiontask)`

##### Note

In the original analyses, a constant of one was added to all frequencies.

##### Format

A data frame consisting of four variables:

`group`

factor. The control group (

`CG`

) received standard instructions, the experimental group (`EG`

) got additional helpful hints.`pattern`

character. Response pattern indicating which card(s) were selected (

`1`

) or not selected (`0`

).`exp1`

,`exp2`

the aggregate response frequencies for Experiment 1 and 2, respectively.

##### See Also

`mpt`

.

##### Examples

```
# NOT RUN {
data(selectiontask)
## Inference-guessing model with relaxed assumptions
s <- mptspec("WST", .replicates = 2)
m1 <- mpt(s, data = selectiontask$exp1, method = "EM")
## Inference-guessing model
m2 <- mpt(update(s, .restr = list(sf1=s1, sb1=s1, sfb1=s1,
sf2=s2, sb2=s2, sfb2=s2)),
data = m1$y, method = "EM")
## Effect of hint on i parameter (Exp. 1)
m3 <- mpt(update(m2$spec, .restr = list(i2=i1)), data = m1$y,
method = "EM")
## Independence model
m4 <- mpt(update(m2$spec,
.restr = list(a1=0, c1=0, x1=0, d1=0, s1=0, i1=0,
a2=0, c2=0, x2=0, d2=0, s2=0, i2=0)),
data = m1$y, method = "EM")
anova(m4, m3, m2, m1)
plogis(confint(m2))
AIC(m2)
BIC(m2) # BIC w/number of non-redundant response categories
AIC(m2, k = log(sum(m2$y))) # BIC w/total number of data points
## Effect of hint on c parameter (Exp. 2)
m5 <- mpt(m2$spec, data = selectiontask$exp2, method = "EM")
m6 <- mpt(update(m5$spec, .restr = list(c2=c1)), data = m5$y,
method = "EM")
anova(m6, m5)
idx <- c("P", "p", "Q", "q", "a", "c", "x", "d", "s", "i")
par(mfrow = 1:2)
dotchart(coef(m2)[paste0(idx, 1)], xlim=c(0, 1), labels=idx,
main="Exp. 1", xlab="")
points(coef(m2)[paste0(idx, 2)], 1:10, pch=16)
legend(0, 11, c("standard", "hints"), pch=c(1, 16),
title="Instruction", bty="n")
dotchart(coef(m5)[paste0(idx, 1)], xlim=c(0, 1), labels=idx,
main="Exp. 2", xlab="")
points(coef(m5)[paste0(idx, 2)], 1:10, pch=16)
mtext("Parameter estimate (inference-guessing model)", side=1,
outer=TRUE, line=-2)
mtext("Klauer et al. (2007)", side=3, outer=TRUE, line=-3)
# }
```

*Documentation reproduced from package mpt, version 0.6-2, License: GPL (>= 2)*