hilbig2014 {multinomineq}

Data: Multiattribute Decisions (Hilbig & Moshagen, 2014)

Description

Choice frequencies of multiattribute decisions across 3 item types (Hilbig & Moshagen, 2014).

Usage

hilbig2014

Format

A data frame 3 variables:

B1

Frequency of choosing Option B for Item Type 1

B2

Frequency of choosing Option B for Item Type 2

B3

Frequency of choosing Option B for Item Type 3

Details

Each participant made 32 choices for each of 3 item types with four cues (with validities .9, .8, .7, and .6).

The pattern of cue values of Option A and and B was as follows:

Item Type 1:

A = (1, 1, 1, -1) vs. B = (-1, 1, -1, 1)

Item Type 2:

A = (1, -1, -1, -1) vs. B = (-1, 1, 1, -1)

Item Type 3:

A = (1, 1, 1, -1) vs. B = (-1, 1, 1, 1)

References

Hilbig, B. E., & Moshagen, M. (2014). Generalized outcome-based strategy classification: Comparing deterministic and probabilistic choice models. Psychonomic Bulletin & Review, 21(6), 1431-1443. doi:10.3758/s13423-014-0643-0

Examples

data(hilbig2014)
head(hilbig2014)

# validities and cue values
v <- c(.9, .8, .7, .6)
cueA <- matrix(
  c(
    1, 1, 1, -1,
    1, -1, -1, -1,
    1, 1, 1, -1
  ),
  ncol = 4, byrow = TRUE
)
cueB <- matrix(
  c(
    -1, 1, -1, 1,
    -1, 1, 1, -1,
    -1, 1, 1, 1
  ),
  ncol = 4, byrow = TRUE
)

# get strategy predictions
strategies <- c(
  "baseline", "WADDprob", "WADD",
  "TTB", "EQW", "GUESS"
)
preds <- strategy_multiattribute(cueA, cueB, v, strategies)
c <- c(1, rep(.5, 5)) # upper bound of probabilities

# use Bayes factor for strategy classification
n <- rep(32, 3)
strategy_postprob(k = hilbig2014[1:5, ], n, preds)