Poirier et al. (2024), Study 3 — PCSCFL24

Data for Study 3, which assigned every participant to a pairwise comparison. The variable condition indicates which brand was in the no/medium/high, depending on the pairwise comparison. Participants had to select one treadmill from two models (counterbalanced), with images showing either condition with brand. The choice is recorded in choice. In a contingency table of say no vs high for social presence, the high are the sum of the diagonal entries. The authors performed Pearson chi-square test for testing the presence of an interaction in two by two tables.

Usage

PCSCFL24_S3

Format

A data frame with 302 rows and 3 variables:

subset: [factor] experimental comparison
condition: [factor] social presence dummy index: e.g., in the no vs high subset, brand A is high and B is no if condition=1, and brand A is no, B is high when condition=0
choice: [integer] choice between two products, either brand A or B

Source

Sara-Maude Poirier, personal communication, distributed under CC BY-NC-SA 4.0

Note

The authors seemingly compared the interaction in a two-way contingency table using Pearson chi-square statistic.

References

Poirier, S.-M., Cosby, S., Sénécal, S., Coursaris, C. K., Fredette, M. et Léger, P.-M. (2024). The impact of social presence cues in social media product photos on consumers’ purchase intentions. Journal of Business Research, 185, 114932. doi:10.1016/j.jbusres.2024.114932 .

Examples

data_novshigh <- subset(PCSCFL24_S3, subset == "no vs high", select = c(choice, condition))
(tab_novshigh <- table(data_novshigh))
#>       condition
#> choice  0  1
#>      A 27 19
#>      B 23 32
# Order is counterbalanced, so roughly the same number with each condition
colSums(tab_novshigh)
#>  0  1 
#> 50 51 
# Pearson chi-square test
chisq.test(tab_novshigh, correct = FALSE)
#> 
#> 	Pearson's Chi-squared test
#> 
#> data:  tab_novshigh
#> X-squared = 2.8544, df = 1, p-value = 0.09112
#>