LIWC has also been used extensively for studying gender and age [21]. Many studies have focused on function words (articles, prepositions, conjunctions, and pronouns), finding females use more first-person singular pronouns, males use more articles, and that older Vesnarinone biological activity individuals use more plural pronouns and future tense verbs [30?2]. Other works have found males use more formal, affirmation, and informational words, while females use more social interaction, and deictic language [33?6]. For age, the most salient findings include older individuals using more positive emotion and less negative emotion words [30], older individuals preferring fewer self-references (i.e. `I’, `me’) [30,31], and stylistically there is less use of negation [37]. Similar to our finding of 2000 topics (clusters of semantically-related words), Argamon et al. used factor analysis and identified 20 coherent components of word use to link gender and age, showing male components of language increase with age while female factors decrease [32]. Occasionally, studies find contradictory results. For example, multiple studies report that emoticons (i.e. `:)’ `:-(`) are used more often by females [34,36,38], but Huffaker Calvert found males use them more in a sample of 100 teenage bloggers [39]. This particular discrepancy could be sample-related ?differing demographics or having a non-representative sample (Huffaker Calvert looked at 100 bloggers, while later studies have looked at thousands of twitter users) or it could be due to differences in the domain of the text (blogs versus twitter). One should always be careful generalizing new (R)-K-13675MedChemExpress Pemafibrate results outside of the domain they were found as language is often dependent on context [40]. In our case we explore language in the broad context of Facebook, and do not claim our results would up under other smaller or larger contexts. As a starting point for reviewing more psychologically meaningful language findings, we refer the reader to Tauszczik Pennebaker’s 2010 survey of computerized text analysis [20].Personality, Gender, Age in Social Media LanguageEisenstein et al. presented a sophisticated open-vocabulary language analysis of demographics [41]. Their method views language analysis as a multi-predictor to multi-output regression problem, and uses an L1 norm to select the most useful predictors (i.e. words). Part of their motivation was finding interpretable relationships between individual language features and sets of outcomes (demographics), and unlike the many predictive works we discuss in the next section, they test for significance of relationships between individual language features and outcomes. To contrast with our approach, we consider features and outcomes individually (i.e. an “L0 norm”), which we think is more ideal for our goals of explaining psychological variables (i.e. understanding openness by the words that correlate with it). For example, their method may throwout a word which is strongly predictive for only one outcome or which is collinear with other words, while we want to know all the words most-predictive for a given outcome. We also explore other types of open-vocabulary language features such as phrases and topics. Similar language analyses also occurred in many fields outside of psychology or demographics [42,43]. For example, Monroe et al. explored a variety of techniques that compare two frequencies of words ?one number for each of two groups [44]. In particular, they explored frequencies.LIWC has also been used extensively for studying gender and age [21]. Many studies have focused on function words (articles, prepositions, conjunctions, and pronouns), finding females use more first-person singular pronouns, males use more articles, and that older individuals use more plural pronouns and future tense verbs [30?2]. Other works have found males use more formal, affirmation, and informational words, while females use more social interaction, and deictic language [33?6]. For age, the most salient findings include older individuals using more positive emotion and less negative emotion words [30], older individuals preferring fewer self-references (i.e. `I’, `me’) [30,31], and stylistically there is less use of negation [37]. Similar to our finding of 2000 topics (clusters of semantically-related words), Argamon et al. used factor analysis and identified 20 coherent components of word use to link gender and age, showing male components of language increase with age while female factors decrease [32]. Occasionally, studies find contradictory results. For example, multiple studies report that emoticons (i.e. `:)’ `:-(`) are used more often by females [34,36,38], but Huffaker Calvert found males use them more in a sample of 100 teenage bloggers [39]. This particular discrepancy could be sample-related ?differing demographics or having a non-representative sample (Huffaker Calvert looked at 100 bloggers, while later studies have looked at thousands of twitter users) or it could be due to differences in the domain of the text (blogs versus twitter). One should always be careful generalizing new results outside of the domain they were found as language is often dependent on context [40]. In our case we explore language in the broad context of Facebook, and do not claim our results would up under other smaller or larger contexts. As a starting point for reviewing more psychologically meaningful language findings, we refer the reader to Tauszczik Pennebaker’s 2010 survey of computerized text analysis [20].Personality, Gender, Age in Social Media LanguageEisenstein et al. presented a sophisticated open-vocabulary language analysis of demographics [41]. Their method views language analysis as a multi-predictor to multi-output regression problem, and uses an L1 norm to select the most useful predictors (i.e. words). Part of their motivation was finding interpretable relationships between individual language features and sets of outcomes (demographics), and unlike the many predictive works we discuss in the next section, they test for significance of relationships between individual language features and outcomes. To contrast with our approach, we consider features and outcomes individually (i.e. an “L0 norm”), which we think is more ideal for our goals of explaining psychological variables (i.e. understanding openness by the words that correlate with it). For example, their method may throwout a word which is strongly predictive for only one outcome or which is collinear with other words, while we want to know all the words most-predictive for a given outcome. We also explore other types of open-vocabulary language features such as phrases and topics. Similar language analyses also occurred in many fields outside of psychology or demographics [42,43]. For example, Monroe et al. explored a variety of techniques that compare two frequencies of words ?one number for each of two groups [44]. In particular, they explored frequencies.
