Show/hide main menu


News Highlights

Social media users adapt personas specifically for platforms

Posted on 01/06/2017
Typing at keyboard

Researchers at King’s College London, working in collaboration with Penn State University, have found that social media users adapt their behaviour to individual social media platforms in a way that is clearly identifiable and learnable when tested on a model.

Using the webpages of 116,998 users, the research team extracted matched user profiles on several major social networks including Facebook, Twitter, LinkedIn and Instagram and found that different genders and age groups adapt their behaviour differently from each other. Such behaviour includes the language they use in profile descriptions and the type of images they choose for their profile pictures.

Women were found to smile more than men in the profile pictures across all five of the social networks, while those under 25 were less likely to smile and this was suggested to be a result of the popularity of selfies amongst young people.

On LinkedIn, 90% of profile pictures were of a single person, while less than 60% of profile pictures on Facebook and Instagram were of only one person. Interestingly, on Facebook and Instagram, up to 40% of users used a picture that didn’t include their face.

In contrast to statistics which show that women have a slightly higher myopia rate than men, more men were found to wear glasses in their profile pictures, suggesting that the two genders choose to fit in with gender-specific norms or social pressures; men may want to appear more intelligent while women appear to give in to social pressure to not wear glasses.

The first author, Dr Changtao Zhong, now working for Twitter, says that the model developed by the researchers is able to automatically identify the network when given a profile picture or self-description.

The differences in behaviour were generally consistent across different platforms and these behaviour differences correspond to how informal or formal the network is. Networks that are more formal such as LinkedIn are easier to tell apart from networks that are less formal such as Facebook.

Dr Nishanth Sastry, Department of Informatics, said: ‘The results of our research have shown that different social networks do have different conventions, and users adapt their profiles to suit these conventions. Our findings have implications for advertising strategies. For example, brands that find a large audience on one particular social media platform may not be popular on another, because the expected norms and core demographic could be different.’

For further information please contact the Public Relations Department at King’s College London on 0207 848 3202 or

The full paper, Wearing Many (Social) Hats: How Different are Your Different Social Network Personae?, is available to read here.

Rss Feed Atom Feed

News Highlights:

News Highlights...RSS FeedAtom Feed

Fake online profiles easier to fish out with new software tool

Fake online profiles easier to fish out with new software tool

Computer scientists from the Department of Informatics and the University of Edinburgh have trained computer models to spot social media users who make up information about themselves – known as catfishes.
Schools explore horticultural robotics with King's researchers

Schools explore horticultural robotics with King's researchers

A group of 16-year-olds from local schools recently got the chance to explore horticultural robotics in a series of workshops run by Department of Informatics researcher Aran Sena.
PhD student wins innovation award

PhD student wins innovation award

A PhD student who has created a medical app to help patients and their carers keep a track of medication has won a £20,000 prize for innovation.

Share this story:


Follow Us


Live Twitter feed...

Join the conversation
Sitemap Site help Terms and conditions Privacy policy Accessibility Modern slavery statement Contact us

© 2017 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454