We examined prospective variations from the website, geographical region, and you may ethnicity playing with t-assessment and you may investigation from variance (ANOVA) to the LIWC class percent. Into two websites, half a dozen of one’s a dozen t-testing was high on the pursuing the categories: first-individual only one [t(3998) = ?5.61, p Secondary Table 2 getting setting, standard deviations, and you will contrasts between ethnic communities). Contrasts shown significant differences between White and all of other cultural communities into the four of your half a dozen high ANOVAs. Thus, i integrated ethnicity since a great dummy-coded covariate in analyses (0 = Light, step one = Another ethnic teams).
Of your own several ANOVA screening linked to geographic area, just several was indeed tall (family and self-confident feelings). Just like the variations weren’t commercially significant, we didn’t thought geographic part into the after that analyses.
Performance
Frequency out of term have fun with goes without saying from inside the detailed analytics sexy La paz women (pick Table step one) and you will thru keyword-clouds. The term-cloud method illustrates by far the most widely used conditions across the whole shot plus each of the a long time. The definition of-affect program immediately excludes certain terminology, also stuff (a beneficial, and you will, the) and you will prepositions (to help you, that have, on). The remainder content words is actually scaled sizes relative to their volume, doing an user-friendly portrait really prevalent content terminology round the new take to ( Wordle, 2014).
Shape 1 shows the new 20 most typical content terms and conditions used in the whole shot. As well as rise above the crowd, the most frequently used terms was in fact love (lookin when you look at the 67% out of users), such as for example (lookin during the 62% from users), looking (lookin when you look at the 55% away from pages), and you may somebody (looking inside the 50% out-of profiles). Ergo, the most popular words had been comparable all over a long time.
Contour dos shows next 30 most typical stuff conditions in the fresh youngest and you can eldest age range. By eliminating the first 20 well-known posts words across the shot, i instruct heterogeneity on the matchmaking users. Next 30 terms towards the youngest generation, raised percentage conditions included rating (36% away from users regarding youngest age bracket), go (33% out of profiles about youngest age group), and you may really works (28% off pages in the youngest age bracket). In contrast, the newest eldest generation had large percentages out of terms and conditions for example traveling (31% out of pages regarding eldest generation), higher (24% regarding users regarding the earliest generation), and relationship (19% from users from the oldest age group).
Next 31 most common terms and conditions in the youngest and you will earliest ages teams (immediately following subtracting the newest 20 most frequent terminology regarding Figure 1).
Hypothesis Evaluation old Variations in Vocabulary into the Dating Pages
To test hypotheses, this new part of conditions on relationships reputation that suit per LIWC category offered while the founded variables inside the regressions. I checked age and gender as independent parameters and adjusting having web site and you can ethnicity.
Theory step one: Elderly decades will be with the a high part of terminology in the following the categories: first-person plural pronouns, friends, members of the family, fitness, and you may positive emotion.
Conclusions mainly served Hypothesis 1 (come across Dining table dos). Five of the five regressions shown a critical main impact to possess many years, in a way that since chronilogical age of the fresh new profile copywriter increased, the brand new portion of terms and conditions throughout the category enhanced from the following categories: first-individual plural, relatives, wellness, and you can self-confident emotion. We found no significant ages impression on the proportion away from terminology regarding family group.
a Gender: 0 (female) and you can step 1 (male). b Webpages: The 2 websites have been dictomously coded while the step one and you will 0. c Ethnicity: 0 (White) and step one (Ethnic or racial fraction).
good Gender: 0 (female) and you can 1 (male). b Website: Both other sites was dictomously coded because 1 and you can 0. c Ethnicity: 0 (White) and you can 1 (Ethnic or racial fraction).