We looked at prospective differences of the webpages, geographical part, and you will ethnicity using t-tests and you will analysis from difference (ANOVA) into LIWC classification percent. With the a couple websites, six of your a dozen t-testing was significant in the adopting the kinds: first-people only one [t(3998) = ?5.61, p Supplementary Dining table 2 getting setting, important deviations, and you may contrasts anywhere between ethnic groups). Contrasts revealed extreme differences between White and all sorts of most other cultural teams into the five of one’s half a dozen tall ANOVAs. Thus, we provided ethnicity as the a dummy-coded covariate for the analyses (0 = Light, step one = Another ethnic teams).
Of your twelve ANOVA examination connected with geographic area, simply two was in fact tall (household members and you can self-confident feelings). Once the differences were not officially meaningful, i failed to imagine geographic part within the after that analyses.
Frequency regarding term use is obvious into the detailed analytics (get a hold of Table step 1) and you will thru term-clouds. The term-cloud techniques portrays one particular widely used words over the whole attempt and in each of the a long time. The expression-affect system immediately excludes particular conditions, and content (a, and you will, the) and you may prepositions (in order to, which have, on). The remainder articles words was scaled in dimensions in line with their volume, undertaking an intuitive portrait of the very most prevalent blogs terms and conditions across the this new test ( Wordle, 2014).
Profile 1 reveals the new 20 most commonly known blogs conditions utilized in the complete take to. As well as get noticed, one particular frequently employed words had been love (appearing in the 67% away from users), such as for example (looking when you look at the 62% regarding users), appearing (lookin in the 55% regarding users), and some one (looking when you look at the 50% regarding profiles).