Fig. 2: Distributions of average persuasiveness scores across conditions in studies 1, 2a, and 2b.

Distributions of average persuasiveness scores within subjects across conditions in Study 1 (left, N = 666), Study 2a (center, N = 432), and Study 2b (right, N = 669). In each study, each participant provided ratings for three groups of ads: non-targeted ads (red), targeted ads without a warning popup (yellow), and targeted ads with a warning popup (blue). The point ranges within the violin plots depict the 95% confidence intervals for the estimated marginal means of persuasiveness for each ad type. These values were extracted from the regression models and represent the average estimated persuasiveness for each level of the “ad type” predictor. In Study 1, the marginal means of persuasiveness are M_{non−targeted} = 2.47, M_popup = 2.66, and M_targeted = 2.68. The patterns are similar in Study 2a, with M_{non−targeted} = 2.59, M_popup = 2.71, and M_targeted = 2.74, as well as in Study 2b, with M_{non−targeted} = 2.65, M_popup = 2.76, and M_targeted = 2.84. In each panel, the red dotted line represents the marginal mean of reference (i.e., the targeted ads without a popup).