heck2017_raw: Data: Multiattribute Decisions (Heck, Hilbig & Moshagen,...

In multinomineq: Bayesian Inference for Multinomial Models with Inequality Constraints

Data: Multiattribute Decisions (Heck, Hilbig & Moshagen, 2017)

Description

Raw data with multiattribute decisions (Heck, Hilbig & Moshagen, 2017).

Usage

heck2017_raw

Format

A data frame with 21 variables:

vp

ID code of participant

trial

Trial index

pattern

Number of cue pattern

ttb

Prediction of take-the-best (TTB)

eqw

Prediction of equal weights (EQW)

wadd

Prediction of weighted additive (WADD)

logoddsdiff

Log-odds difference (WADDprob)

ttbsteps

Number of TTB steps (TTBprob)

itemtype

Item type as in paper

reversedorder

Whether item is reversed

choice

Choice

rt

Response time

choice.rev

Choice (reversed)

a1

Value of Cue 1 for Option A

a2

Value of Cue 2 for Option A

a3

Value of Cue 3 for Option A

a4

Value of Cue 4 for Option A

b1

Value of Cue 1 for Option B

b2

Value of Cue 2 for Option B

b3

Value of Cue 3 for Option B

b4

Value of Cue 4 for Option B

Details

Each participant made 40 choices for each of 4 item types with four cues (with validities .9, .8, .7, and .6). Individual choice freqeuncies are available as heck2017

References

Heck, D. W., Hilbig, B. E., & Moshagen, M. (2017). From information processing to decisions: Formalizing and comparing probabilistic choice models. Cognitive Psychology, 96, 26-40. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1016/j.cogpsych.2017.05.003")}

See Also

heck2017 for the aggregated choice frequencies per item type.

Examples

data(heck2017_raw)
head(heck2017_raw)


# get cue values, validities, and predictions
cueA <- heck2017_raw[, paste0("a", 1:4)]
cueB <- heck2017_raw[, paste0("b", 1:4)]
v <- c(.9, .8, .7, .6)
strat <- strategy_multiattribute(
  cueA, cueB, v,
  c(
    "TTB", "TTBprob", "WADD",
    "WADDprob", "EQW", "GUESS"
  )
)

# get unique item types
types <- strategy_unique(strat)
types$unique

# get table of choice frequencies for analysis
freq <- with(
  heck2017_raw,
  table(vp, types$item_type, choice)
)
freqB <- freq[, 4:1, 1] + # reversed items: Option A
  freq[, 5:8, 2] # non-rev. items: Option B
head(40 - freqB)
data(heck2017)
head(heck2017) # same frequencies (different order)

# strategy classification
pp <- strategy_postprob(
  freqB[1:4, ], rep(40, 4),
  types$strategies
)
round(pp, 3)

multinomineq documentation built on May 29, 2024, 6:23 a.m.