Presentation of Self in Social Media

We study social media with the assumption that people reveal “who they are” when they post to Facebook, Twitter or Instagram: that men write more like men than women do, that extraverts look extraverted, depressed people depressed, and happy people happy. But do people present their “true selves” on Facebook and Twitter?

Of course not. Twelve year olds pretend to be thirteen — otherwise they are kicked off. And people don’t always share their embarrassing medical conditions or their illicit drug use–although they do share both surprisingly often.

Of course people want to look good. It is claimed that on dating sites like OKCupid, people on average inflate their height by two inches and their income by 20%. And, of course, they pick attractive — and sometime out of date — photos of themselves. Narcissists, on average post more photos of themselves to Facebook and edit them more often than the rest of us.

In fact, it’s not clear if people ever present their true selves–or have true selves to present. You don’t need to be a psychologist to know that simple questions like: “Are you a racist?” “Do you think I’m fat?” or even “How old are you?” do not always elicit honest answers. Sociologist Irving Goffman in his book The presentation of self in everyday life famously observed that we are always acting: when a waiter comes out of the door from the kitchen into the dining room, he puts on a persona for the diners, but when he goes back into the kitchen he doesn’t become ‘his true self’ — he just shows a different persona for the kitchen staff. We all behave differently with our friends than with our colleagues or parents. We’re always presenting ourselves on some stage.

Self discrepancy theory argues that people have an ‘ought’, ‘ideal’, and ‘actual’ self. The ideal self is who one would like to become, the ought self is who one thinks they should be and the actual self is who one actually thinks one is. All of these are valid selves, so an idealized self isn’t necessarily a self that’s not “true”. Thus people may “use” social media to remind them of their ideal self or ought self, or to explore different roles (e.g being openly gay) in ways that are not revealed in face-to-face communication.

Psychologists try and understand people using survey questions, for example measuring personality using questionnaires designed to elicit the “Five Factor” or OCEAN (open, conscientious, extraverted, agreeable, neurotic) personality dimensions. But questionnaire results are no more a ground truth of personality than predictions from the words on people’s Facebook posts; both try to get at an underlying latent variable that will predict a variety of behaviors. Language-based assessments of personality correlate as highly with estimates by friends as questionnaire-based assessments do. None of these assessments predict behavior as well as most people expect.

People adopt different social roles in different contexts. In spite of our strong intuitions that people have stable personalities, estimates of personality and other traits are surprisingly unstable; predicting behavior based on personality is even harder. Seminary students may be more inclined to help the needy than the rest of us, but to predict if they will stop to help someone in need requires knowing not their morality, but how much of a hurry they are in. Psychology is rife with studies showing the importance of context; what a researcher wears (a white lab coat or a t-shirt with ‘eracism’) on it influences how subjects behave (e.g. following requests to shock people or appearing less racist).

Frustratingly, psychologists have a hard time collecting enough information about people’s daily lives to fine tune and validate these models. Social media offers some promise of help here.

In spite of people ‘shading’ the truth, social media do reveal a vast amount about the posters. People present themselves honestly enough that one can see who they are and what they care about. We can guess gender with over 90% accuracy and age on average within 4 years from Facebook posts. Most women on Facebook don’t pretend to be men, and few 20-year olds pretend to be 40. Extraverts and introverts look systematically different on Facebook, even if everyone wants to appear to be someone you’d want to have dinner with. Depressed people generally write more depressed posts, and psychotics often sound genuinely crazy — they aren’t generally aware of how crazy they sound.

Perhaps more surprisingly, some people do reveal many stigmatized identities. Twitter is being used as a source of language of people who publicly tweet “I was diagnosed with depression” or PTSD. People who tweet about their depression are not “typical” people, but they still use the language of depressed people and thus provide a useful source of data for research on the language of depression.

Facebook users are not a random sample of the population, but in spite of findings that extraverts and insecure people are more active on Facebook, we find that the distribution of Facebook users, while not representative of the age variation in the US, have similar distributions over personality as the population at large.

In the end, we all shade our presentation of self to the positive, but it is impossible to fully hide who we are.

Share this on ...Share on Facebook0Tweet about this on Twitter0Share on Google+0Print this pageEmail this to someone

About Lyle Ungar

Dr. Lyle Ungar is a Professor of Computer and Information Science at the University of Pennsylvania, where he also holds appointments in multiple departments in the Schools of Business, Medicine, Arts and Sciences and Engineering and Applied Science. Lyle received a B.S. from Stanford University and a Ph.D. from M.I.T. He has published over 200 articles and is co-inventor on eleven patents. His current research focuses on developing scalable machine learning methods for data mining and text mining, including spectral methods for NLP, and analysis of social media to better understand what determines physical and mental well-being.

One thought on “Presentation of Self in Social Media

  1. Pingback: anqilianqili

Leave a Reply

Your email address will not be published. Required fields are marked *