Skip to main content
. Author manuscript; available in PMC: 2018 Feb 1.
Published in final edited form as: Addict Behav. 2016 Oct 19;65:63–67. doi: 10.1016/j.addbeh.2016.10.004

Table 2.

Selecting the most likely destination for posts portraying negative alcohol-related consequences: generalized Bradley-Terry model results for pairwise platform comparisons

Stimuli Platform Parameter Estimate SD Probabilities of winning

vs FB vs IG vs SC
Punched out drunk at
party vignette
Facebook (FB) .08 .01 - .26 .03
Instagram (IG) .21*FB .02 .74 - .07
Snapchat (SC) 2.71**FB, IG .02 .97 .92 -
Throwing up out the
window of an Uber
vignette
Facebook (FB) .04 .01 - .27 .02
Instagram (IG) .10*FB .01. .63 - .03
Snapchat (SC) 2.86**FB, IG .01 .98 .97 -
First to get sick after
drinking for hours
vignette
Facebook (FB) .06 .01 - .29 .02
Instagram (IG) .18*FB .01 .71 - .05
Snapchat (SC) 2.78**FB, IG .01 .98 .95 -
Friend throwing up
after a rough night
photo
Facebook (FB) .09 .02 - .36 .03
Instagram (IG) .18*FB .02 .64 - .05
Snapchat (SC) 2.75**FB, IG .01 .97 .95 -
Passed out in yard with
empty beer cans photo
Facebook (FB) .10 .02 - .42 .04
Instagram (IG) .16 .02 .58 - .06
Snapchat (SC) 2.73**FB, IG .02 .96 .94 -
Asleep on toilet rough
night photo
Facebook (FB) .04 .02 - .25 .02
Instagram (IG) .13*FB .01 .65 - .04
Snapchat (SC) 2.82**FB, IG .01 .98 .96 -
*

p<.01;

**

p<.001