Designing a conjoint survey

Choice Modeling for Marketing in R

Elea McDonnell Feit

Assistant Professor of Marketing, Drexel University

Conjoint survey

Conjoint question

Attributes and levels

Type
milk, dark, milk with nuts, dark with nuts, white

Brand
Dove, Ghirardelli, Godiva, Hershey's, Lindt

Price
0.5, 0.6, ... steps of 0.1 ..., 3.9, 4.0

Designing a choice survey

choc_survey[choc_survey$Subject == 1 & choc_survey$Trial == 1, ]

     Subject Trial Alt Type Brand Price
1          1     1   1   NA    NA    NA
2          1     1   2   NA    NA    NA
3          1     1   3   NA    NA    NA

Creating a random design part 1

# Setup your attributes and levels list
attribs <- list(Type = c("Milk", "Dark", "White"), 
                Brand = c("Cadbury", "Toblerone", "Kinder"), 
                Price = 5:30 / 10)

# Create all possible combinations of attributes
all_comb <- expand.grid(attribs)
nrow(all_comb)
head(all_comb)

144

Type     Brand Price
1  Milk   Cadbury   0.5
2  Dark   Cadbury   0.5
3 White   Cadbury   0.5
4  Milk Toblerone   0.5
5  Dark Toblerone   0.5
6 White Toblerone   0.5

Creating a random design part 2

for (i in 1:100) {
  rand_rows <- sample(1:nrow(all_comb), size = 12 * 3)
  rand_alts <- all_comb[rand_rows, ]
  choc_survey[choc_survey$Subject == i, 4:6] <- rand_alts
}

Fielding your survey: options

Code up the survey yourself.
Upload the survey design to a survey tool like Google Forms or Survey Monkey.
Use a survey tool with a build-in conjoint design feature like Sawtooth, Conjoint.ly or Qualtrics.

Go field a conjoint survey!

Choice Modeling for Marketing in R