We first examined descriptive statistics when it comes to proportions of terms folks of different ages found in their dating pages. We additionally produced illustrative numbers portraying the absolute most words that are common.
We then looked to theory evaluation utilizing ordinary minimum squares regression. The result variables in this scholarly research had been the proportion of terms suitable each one of the 12 groups within the LIWC analyses. The LIWC categories had been all favorably skewed because of the amount of zero values (in other words., participant failed to utilize any terms within the category). We went each analysis with a square-root transformation (used to deal with non-normality in previous studies utilising the LIWC; DeAndrea, Shaw & Levine, 2010; Hirsh & Peterson, 2009). The pattern of findings had been comparable after using the transformations. For simplicity of interpretation, findings are presented utilizing the untransformed LIWC category data. The separate variable had been age, treated as a variable that is continuous. We additionally included sex.
Initially, we ran the regressions such as the Age Г— Gender discussion term. One interaction that is significant based in the group of good feeling, so that ladies had greater mean proportions of good feeling terms than guys after all many years, with females showing a somewhat steeper linear enhance as we grow older than males. Hence, we didn’t range from the relationship term for Age Г— Gender into the models reported right right here.
We examined prospective differences by internet site, geographic area, and ethnicity utilizing t-tests and analysis of variance (ANOVA) for the LIWC category percentages. When it comes to two internet sites, six for the twelve t-tests had been significant within the following categories: first-person single [t(3998) = в€’5.61, p Supplementary Table 2 for means, standard deviations, and contrasts between cultural teams). Contrasts unveiled significant differences when considering White and all sorts of other cultural teams in four associated with the six ANOVAs that are significant. Because of this, we included ethnicity as a covariate that is dummy-coded analyses (0 = White, 1 = all the cultural groups).
Associated with 12 ANOVA tests related to geographical area, only two had been significant (family members and good emotion). Due to the fact distinctions weren’t theoretically significant, we would not give consideration to geographical region in subsequent analyses.
Descriptive Statistics and Illustrations of Widely Used Terms
Frequency of word usage is evident in descriptive data (see dining dining Table 1) and via word-clouds. The word-cloud method illustrates probably the most widely used terms over the whole test and in each one of the age ranges. The word-cloud system automatically excludes specific terms, including articles (a, and, the) and prepositions (to, with, on). The rest of the content terms are scaled in dimensions in accordance with their regularity, producing an intuitive portrait of the most extremely content that is prevalent throughout the sample ( Wordle, 2014).
Figure 1 shows the 20 most frequent content terms utilized in catholicmatch the sample that is entire. As can be observed, the essential frequently employed terms were love (showing up in 67percent of pages), like (appearing in 62per cent of pages), looking (showing up in 55% of pages), and some body (showing up in 50per cent of pages). Hence, the absolute most words that are common comparable across age ranges.
Twenty most frequent content terms over the sample that is entire.
Twenty most frequent content terms throughout the sample that is entire.
Figure 2 shows the following 30 most frequent content terms within the youngest and earliest age brackets. By eliminating 1st 20 content that is common over the test, we illustrate heterogeneity when you look at the dating pages. Within the next 30 words for the youngest age bracket, high level percentage words included get (36% of pages within the youngest age bracket), get (33% of pages into the youngest age bracket), and work (28% of pages within the youngest age bracket). In comparison, the age group that is oldest had greater percentages of terms such as for example travel (31% of pages into the earliest age bracket), great (24% of pages into the earliest age bracket), and relationship (19% of pages into the earliest age bracket).
Next 30 most typical terms within the youngest and age groups that are oldest (after subtracting the 20 most typical terms from Figure 1).
Next 30 most frequent terms into the youngest and age groups that are oldest (after subtracting the 20 most frequent terms from Figure 1).