The conclusions of a wide research carried out by a team from the University of Pennsylvania under the direction of H. Andrew Schwartz have recently been published. Under the title “Personality, Gender and Age in the Language of social Media: The Open-Vocabulary Approach”, the results of the work carried out with 75,000 volunteers have been published and analysed. Not bad. For nearly 3 years, researchers have studied over 15 million messages posted in Facebook, analysing approximately 700 million words.
Conclusions about gender and age, among other things, have been drawn in this study. And this will lead to interesting reflections.
The vocabulary and references most repeated in social media by men are related to sex, government, girlfriend, Playstation and football.
The vocabulary and expressions used by women focus more frequently on emotional issues, including constant references to love, family, shopping, birthdays and hair.
We must admit this study seems to reproduce any and all the stereotypes we may think of, but it is in fact an empirical and carefully carried out study covering an impressing sample.
At the same time, a progression has been observed with regards to the topics that become more important depending on age. Thus, the most frequent topics among teenagers under 18 years old are school and homework. Among young people between 18 and 22, words such as sex, university and drunkenness become prominent. Between 23 and 29 years of age, words such as beer or drinking appear more frequently, and this seems to point towards more moderate drinking habits. Between 30 and 65 years old, words such as son/daughter or work are more frequent.
Some other conclusions refer to introversion and extraversion, for example. It has been concluded that introverted people use shorter words than extraverted people (of course). Older people tend to use plural pronouns and younger people singular pronouns like “I” (just as we would expect according to the different developing stages…)
At this point, I am convinced that you are probably thinking about the very same question that came to my mind after reading the whole study… But we will not ask it! We will be cautious… So I decided to seek help from several primary school and high school teachers who work with teenagers everyday. I have asked them about the topics their students talk more frequently about, and if they see differences between girls and boys. And in fact the conclusions they reached in a few minutes are exactly the same that are mentioned in this study. And when I then referred to the research I was reading, they asked me the very same question you and I didn’t dare ask before…: “And they needed to do a 3-year study with 75,000 volunteers to come to THAT CONCLUSION…?”
Taking into account I totally respect all researchers, I believe this excellent study gives me the opportunity to draw a conclusion I believe to be very clear and obvious. I do not think there exists any wider or better developed study to show that teenagers, just like adults, talk on the Internet about the very same things they talk about offline. The conversations we hold with our social media “online” are the same we hold with our daily social environment “offline”. No more, no less. And the reason for this is very simple: in our everyday life and in our environment, just like on the Internet, we relate to people. And what’s more, we relate mainly with the same people. Internet is just a means, a channel that may determine certain things, but only to a certain extent. In fact, for teenagers there is no difference between their online life and their offline life. They constantly insist on that. And in fact, most people with whom they talk to using Facebook or Tuenti are the same people they spend the morning with in class: their friends.
H. Andrew Schwartz, Johannes Eichstaedt, Margaret L. Kern, Lukasz Dziurzynski, Stephanie Ramones, M. Agrawal, A. Shah, A., M. Kosinski, D. Stillwell, Martin Seligman, and Lyle Ungar. 2013. “Personality, Gender and Age in the Language of Social Media: The Open-Vocabulary Approach”. Published in PLoS ONE.