Example of a raw, “wide” conjoint dataset to demonstrate functionality of cj_tidy

A simulated dataset containing 100 respondents' responses to four decision tasks (a,b,c,d) involving a forced choice between two alternative profiles, described by three features (1,2,3), as well as a secondary rating-scale outcome and a response time measure, along with two respondent-varying covariates. This is used in testing and examples within the package.

data(wide_conjoint)

Format

A data frame with 100 observations on the following variables:

respondent

a numeric vector indicating the respondent identifier

feature1a1

Feature 1 for task A left profile, a factor

feature1b1

Feature 1 for task B left profile, a factor

feature1c1

Feature 1 for task C left profile, a factor

feature1d1

Feature 1 for task D left profile, a factor

feature1a2

Feature 1 for task A right profile, a factor

feature1b2

Feature 1 for task B right profile, a factor

feature1c2

Feature 1 for task C right profile, a factor

feature1d2

Feature 1 for task D right profile, a factor

feature2a1

Feature 2 for task A left profile, a factor

feature2b1

Feature 2 for task B left profile, a factor

feature2c1

Feature 2 for task C left profile, a factor

feature2d1

Feature 2 for task D left profile, a factor

feature2a2

Feature 2 for task A right profile, a factor

feature2b2

Feature 2 for task B right profile, a factor

feature2c2

Feature 2 for task C right profile, a factor

feature2d2

Feature 2 for task D right profile, a factor

feature3a1

Feature 3 for task A left profile, a factor

feature3b1

Feature 3 for task B left profile, a factor

feature3c1

Feature 3 for task C left profile, a factor

feature3d1

Feature 3 for task D left profile, a factor

feature3a2

Feature 3 for task A right profile, a factor

feature3b2

Feature 3 for task B right profile, a factor

feature3c2

Feature 3 for task C right profile, a factor

feature3d2

Feature 3 for task D right profile, a factor

choice_a

outcome for task A indicating which profile was chosen, randomly 1 or 2, each equally probable

choice_b

outcome for task B indicating which profile was chosen, randomly 1 or 2, each equally probable

choice_c

outcome for task C indicating which profile was chosen, randomly 1 or 2, each equally probable

choice_d

outcome for task D indicating which profile was chosen, randomly 1 or 2, each equally probable

rating_a1

rating for task A left profile, random variable between 1 and 7, uniformly distributed

rating_a2

rating for task A right profile, random variable between 1 and 7, uniformly distributed

rating_b1

rating for task B left profile, random variable between 1 and 7, uniformly distributed

rating_b2

rating for task B right profile, random variable between 1 and 7, uniformly distributed

rating_c1

rating for task C left profile, random variable between 1 and 7, uniformly distributed

rating_c2

rating for task C right profile, random variable between 1 and 7, uniformly distributed

rating_d1

rating for task D left profile, random variable between 1 and 7, uniformly distributed

rating_d2

rating for task D right profile, random variable between 1 and 7, uniformly distributed

timing_a

timing for task A in seconds, random draws from a beta distribution (2,5) times 10

timing_b

timing for task A in seconds, random draws from a beta distribution (2,5) times 10

timing_c

timing for task A in seconds, random draws from a beta distribution (2,5) times 10

timing_d

timing for task A in seconds, random draws from a beta distribution (2,5) times 10

covariate1

random draws from a uniform distribution between -1 and 1

covariate2

random draws from the set of 1 and 2

See also

cj_tidy cj

Examples

if (FALSE) {
data("wide_conjoint")
# feature_variables
list1 <- list(
 feature1 = list(
     names(wide_conjoint)[grep("^feature1.{1}1", names(wide_conjoint))],
     names(wide_conjoint)[grep("^feature1.{1}2", names(wide_conjoint))]
 ),
 feature2 = list(
     names(wide_conjoint)[grep("^feature2.{1}1", names(wide_conjoint))],
     names(wide_conjoint)[grep("^feature2.{1}2", names(wide_conjoint))]
 ),
 feature3 = list(
     names(wide_conjoint)[grep("^feature3.{1}1", names(wide_conjoint))],
     names(wide_conjoint)[grep("^feature3.{1}2", names(wide_conjoint))]
 ),
 rating = list(
     names(wide_conjoint)[grep("^rating.+1", names(wide_conjoint))],
     names(wide_conjoint)[grep("^rating.+2", names(wide_conjoint))]
 )
)
# task variables
list2 <- list(choice = paste0("choice_", letters[1:4]),
              timing = paste0("timing_", letters[1:4]))
str(cj_tidy(wide_conjoint, profile_variables = list1, task_variables = list2, id = ~ respondent))
}