Fig 1.
Examples of the original Dutch dating profiles used for the experiment (a, c) and their translated English versions (b, d). Profiles (a) and (b) are male profiles with a high TF-IDF score (bin seven), and (c) and (d) are female profiles with a low TF-IDF score (bin one).
Fig 2.
Distribution of the TF-IDF scores of all 31,163 profile texts in the sample.
Table 1.
Distribution of texts written by men and women in seven bins based on percentiles of the TF-IDF scores.
Table 2.
Means, standard deviations, and correlations among perceived text originality scores and all impression formation variables.
Fig 3.
Results of the four mediation analyses displaying that the effect of perceived profile text originality on (a) physical attractiveness is mediated by perceived intelligence and humor, and that the effect of profile text originality is mediated by perceived intelligence, humor, and oddness for (b) social attractiveness, (c) romantic attractiveness, and (d) dating intention, but for oddness in the opposite direction than hypothesized. The coefficients represent the unstandardized coefficients. Note. ** p < .001, * p < .01. All variables were measured on a seven-point Likert scale, except for ‘intention to date’ which was measured using a dichotomous yes (1)/no (0) question.
Table 3.
Means, standard deviations, and correlations among text originality scores and all coded features.
Table 4.
Summary of Hierarchical regression analysis for text features predicting perceived profile text originality.