Predicting individual-level income from Facebook profiles.

Information about a person's income can be useful in several business-related contexts, such as personalized advertising or salary negotiations. However, many people consider this information private and are reluctant to share it. In this paper, we show that income is predictable from the digital footprints people leave on Facebook. Applying an established machine learning method to an income-representative sample of 2,623 U.S. Americans, we found that (i) Facebook Likes and Status Updates alone predicted a person's income with an accuracy of up to r = 0.43, and (ii) Facebook Likes and Status Updates added incremental predictive power above and beyond a range of socio-demographic variables (ΔR2 = 6-16%, with a correlation of up to r = 0.49). Our findings highlight both opportunities for businesses and legitimate privacy concerns that such prediction models pose to individuals and society when applied without individual consent.

Living in the Past, Present, and Future: Measuring Temporal Orientation With Language.

Temporal orientation refers to individual differences in the relative emphasis one places on the past, present, or future, and it is related to academic, financial, and health outcomes. We propose and evaluate a method for automatically measuring temporal orientation through language expressed on social media. Judges rated the temporal orientation of 4,302 social media messages. We trained a classifier based on these ratings, which could accurately predict the temporal orientation of new messages in a separate validation set (accuracy/mean sensitivity = .72; mean specificity = .77). We used the classifier to automatically classify 1.3 million messages written by 5,372 participants (50% female; ages 13-48). Finally, we tested whether individual differences in past, present, and future orientation differentially related to gender, age, Big Five personality, satisfaction with life, and depressive symptoms. Temporal orientations exhibit several expected correlations with age, gender, and Big Five personality. More future-oriented people were older, more likely to be female, more conscientious, less impulsive, less depressed, and more satisfied with life; present orientation showed the opposite pattern. Language-based assessments can complement and extend existing measures of temporal orientation, providing an alternative approach and additional insights into language and personality relationships.

Musical Preferences are Linked to Cognitive Styles.

Why do we like the music we do? Research has shown that musical preferences and personality are linked, yet little is known about other influences on preferences such as cognitive styles. To address this gap, we investigated how individual differences in musical preferences are explained by the empathizing-systemizing (E-S) theory. Study 1 examined the links between empathy and musical preferences across four samples. By reporting their preferential reactions to musical stimuli, samples 1 and 2 (Ns = 2,178 and 891) indicated their preferences for music from 26 different genres, and samples 3 and 4 (Ns = 747 and 320) indicated their preferences for music from only a single genre (rock or jazz). Results across samples showed that empathy levels are linked to preferences even within genres and account for significant proportions of variance in preferences over and above personality traits for various music-preference dimensions. Study 2 (N = 353) replicated and extended these findings by investigating how musical preferences are differentiated by E-S cognitive styles (i.e., 'brain types'). Those who are type E (bias towards empathizing) preferred music on the Mellow dimension (R&B/soul, adult contemporary, soft rock genres) compared to type S (bias towards systemizing) who preferred music on the Intense dimension (punk, heavy metal, and hard rock). Analyses of fine-grained psychological and sonic attributes in the music revealed that type E individuals preferred music that featured low arousal (gentle, warm, and sensual attributes), negative valence (depressing and sad), and emotional depth (poetic, relaxing, and thoughtful), while type S preferred music that featured high arousal (strong, tense, and thrilling), and aspects of positive valence (animated) and cerebral depth (complexity). The application of these findings for clinicians, interventions, and those on the autism spectrum (largely type S or extreme type S) are discussed.

Do Facebook Status Updates Reflect Subjective Well-Being?

Nowadays, millions of people around the world use social networking sites to express everyday thoughts and feelings. Many researchers have tried to make use of social media to study users' online behaviors and psychological states. However, previous studies show mixed results about whether self-generated contents on Facebook reflect users' subjective well-being (SWB). This study analyzed Facebook status updates to determine the extent to which users' emotional expression predicted their SWB-specifically their self-reported satisfaction with life. It was found that positive emotional expressions on Facebook did not correlate with life satisfaction, whereas negative emotional expressions within the past 9-10 months (but not beyond) were significantly related to life satisfaction. These findings suggest that both the type of emotional expressions and the time frame of status updates determine whether emotional expressions in Facebook status updates can effectively reflect users' SWB. The findings shed light on the characteristics of online social media and improve the understanding of how user-generated contents reflect users' psychological states.

Automatic personality assessment through social media language.

Language use is a psychologically rich, stable individual difference with well-established correlations to personality. We describe a method for assessing personality using an open-vocabulary analysis of language from social media. We compiled the written language from 66,732 Facebook users and their questionnaire-based self-reported Big Five personality traits, and then we built a predictive model of personality based on their language. We used this model to predict the 5 personality factors in a separate sample of 4,824 Facebook users, examining (a) convergence with self-reports of personality at the domain- and facet-level; (b) discriminant validity between predictions of distinct traits; (c) agreement with informant reports of personality; (d) patterns of correlations with external criteria (e.g., number of friends, political attitudes, impulsiveness); and (e) test-retest reliability over 6-month intervals. Results indicated that language-based assessments can constitute valid personality measures: they agreed with self-reports and informant reports of personality, added incremental validity over informant reports, adequately discriminated between traits, exhibited patterns of correlations with external criteria similar to those found with self-reported personality, and were stable over 6-month intervals. Analysis of predictive language can provide rich portraits of the mental life associated with traits. This approach can complement and extend traditional methods, providing researchers with an additional measure that can quickly and cheaply assess large groups of participants with minimal burden.

From "Sooo excited!!!" to "So proud": using language to study development.

We introduce a new method, differential language analysis (DLA), for studying human development in which computational linguistics are used to analyze the big data available through online social media in light of psychological theory. Our open vocabulary DLA approach finds words, phrases, and topics that distinguish groups of people based on 1 or more characteristics. Using a data set of over 70,000 Facebook users, we identify how word and topic use vary as a function of age and compile cohort specific words and phrases into visual summaries that are face valid and intuitively meaningful. We demonstrate how this methodology can be used to test developmental hypotheses, using the aging positivity effect (Carstensen & Mikels, 2005) as an example. While in this study we focused primarily on common trends across age-related cohorts, the same methodology can be used to explore heterogeneity within developmental stages or to explore other characteristics that differentiate groups of people. Our comprehensive list of words and topics is available on our web site for deeper exploration by the research community.

The online social self: an open vocabulary approach to personality.

OBJECTIVE: We present a new open language analysis approach that identifies and visually summarizes the dominant naturally occurring words and phrases that most distinguished each Big Five personality trait. METHOD: Using millions of posts from 69,792 Facebook users, we examined the correlation of personality traits with online word usage. Our analysis method consists of feature extraction, correlational analysis, and visualization. RESULTS: The distinguishing words and phrases were face valid and provide insight into processes that underlie the Big Five traits. CONCLUSION: Open-ended data driven exploration of large datasets combined with established psychological theory and measures offers new tools to further understand the human psyche.

Divided we stand: three psychological regions of the United States and their political, economic, social, and health correlates.

There is overwhelming evidence for regional variation across the United States on a range of key political, economic, social, and health indicators. However, a substantial body of research suggests that activities in each of these domains are typically influenced by psychological variables, raising the possibility that psychological forces might be the mediating or causal factors responsible for regional variation in the key indicators. Thus, the present article examined whether configurations of psychological variables, in this case personality traits, can usefully be used to segment the country. Do regions emerge that can be defined in terms of their characteristic personality profiles? How are those regions distributed geographically? And are they associated with particular patterns of key political, economic, social, and health indicators? Results from cluster analyses of 5 independent samples totaling over 1.5 million individuals identified 3 robust psychological profiles: Friendly & Conventional, Relaxed & Creative, and Temperamental & Uninhibited. The psychological profiles were found to cluster geographically and displayed unique patterns of associations with key geographical indicators. The findings demonstrate the value of a geographical perspective in unpacking the connections between microlevel processes and consequential macrolevel outcomes.

BIS impulsivity and acute nicotine exposure are associated with discounting global consequences in the Harvard game.

OBJECTIVE: The Barratt Impulsivity Scale (BIS) provides a transdiagnostic marker for a number of psychiatric conditions and drug abuse, but the precise psychological trait(s) tapped by this questionnaire remain obscure. METHOD: To address this, 51 smokers completed in counterbalanced order the BIS, a delay discounting task and a Harvard game that measured choice between a response that yielded a high immediate monetary payoff but decreased opportunity to earn money overall (local choice) versus a response that yielded a lower immediate payoff but afforded a greater opportunity to earn overall (global choice). RESULTS: Individual level of BIS impulsivity and self-elected smoking prior to the study were independently associated with increased preference for the local over the global choice in the Harvard game, but not delay discounting. CONCLUSIONS: BIS impulsivity and acute nicotine exposure reflect a bias in the governance of choice by immediate reward contingencies over global consequences, consistent with contemporary dual-process instrumental learning theories.

Effects of measurement methods on the relationship between smoking and delay reward discounting.

AIMS: Delay reward discounting (DRD) measures the degree to which a person prefers smaller rewards soon or larger rewards later. People who smoke have been shown to have higher DRD. There are several ways of measuring DRD, and the method used might influence the association between smoking and DRD. The key differences are the order in which the items are presented, the delays used and the magnitude of the delayed amount. SETTING: An international online study running from September 2010 to June 2011. PARTICIPANTS: A total of 9454 individuals; 38% male, mean age = 23.1 years. DESIGN AND MEASUREMENTS: Users completed a multi-item DRD task. They were randomly presented the immediate rewards in an ascending, descending or randomized order. The delays were between 1 week and 5 years. The delayed amounts were $1000 for all delays, and $100 for 1 month. Users also self-reported their smoking status. FINDINGS: A hyperbolic DRD function fitted better than an exponential function. There were differences in the derived DRD function based on methodology used; items presented in a randomized order, longer delays and smaller rewards showed steeper discounting. However, these did not interact with smoking status, as for all methodologies used daily smokers showed the steepest discounting, followed by non-daily smokers, then non-smokers. CONCLUSIONS: Smokers discount future consequences more than non-smokers, irrespective of which measurement is used, but variations in method lead to different estimates that are not comparable between experiments.

The Song Remains the Same: A Replication and Extension of the MUSIC Model.

There is overwhelming anecdotal and empirical evidence for individual differences in musical preferences. However, little is known about what drives those preferences. Are people drawn to particular musical genres (e.g., rap, jazz) or to certain musical properties (e.g., lively, loud)? Recent findings suggest that musical preferences can be conceptualized in terms of five orthogonal dimensions: Mellow, Unpretentious, Sophisticated, Intense, and Contemporary (conveniently, MUSIC). The aim of the present research is to replicate and extend that work by empirically examining the hypothesis that musical preferences are based on preferences for particular musical properties and psychological attributes as opposed to musical genres. Findings from Study 1 replicated the five-factor MUSIC structure using musical excerpts from a variety of genres and subgenres and revealed musical attributes that differentiate each factor. Results from Studies 2 and 3 show that the MUSIC structure is recoverable using musical pieces from only the jazz and rock genres, respectively. Taken together, the current work provides strong evidence that preferences for music are determined by specific musical attributes and that the MUSIC model is a robust framework for conceptualizing and measuring such preferences.

Melioration behaviour in the Harvard game is reduced by simplifying decision outcomes.

Self-control experiments have previously been highlighted as examples of suboptimal decision making. In one such experiment, the Harvard game, participants make repeated choices between two alternatives. One alternative has a higher immediate pay-off than the other, but with repeated choices results in a lower overall pay-off. Preference for the alternative with the higher immediate pay-off seems to be impulsive and will result in a failure to maximize pay-offs. We report an experiment that modifies the Harvard game, dividing the pay-off from each choice into two separate consequences-the immediate and the historic components. Choosing the alternative with the higher immediate pay-off ends the session prematurely, leading to a loss of opportunities to earn further pay-offs and ultimately to a reduced overall pay-off. This makes it easier for participants to learn the outcomes of their actions. It also provides the opportunity for a further test of normative decision making by means of one of its most specific and paradoxical predictions-that the truly rational agent should switch from self-control to impulsivity toward the end of the experimental sessions. The finding that participants maximize their expected utility by both overcoming impulsivity and learning to switch implies that melioration behaviour is not due to the lure of impulsivity, but due to the difficulty of learning which components are included in the pay-off schedules.