About this study
Half the adults and three-quarters of the teenagers in America use social networking sites (SNS) and Facebook by far is the most popular of these sites.
The Pew Research Center’s Internet & American Life Project fielded a nationally representative phone survey about the social and civic lives of SNS users and reported the findings in June 2011 in a report entitled “Social networking sites and our lives.”1 During the phone survey, 269 of 877 original respondents who were Facebook users gave us permission to access data on their use of Facebook so that it could be matched with their survey responses. We partnered with Facebook to match individual responses from the survey with profile information and computer logs of how those same people used Facebook services over a one-month period in November 2010 that overlapped when the survey was in the field.
The results of that special analysis of 269 Facebook users identified in and recruited from a random, representative telephone survey are reported here.
The average Facebook user gets more from their friends on Facebook than they give to their friends. Why? Because of a segment of “power users,” who specialize in different Facebook activities and contribute much more than the typical user does.
The typical Facebook user in our sample was moderately active over our month of observation, in their tendency to send friend requests, add content, and “like” the content of their friends. However, a proportion of Facebook participants – ranging between 20% and 30% of users depending on the type of activity – were power users who performed these same activities at a much higher rate; daily or more than weekly. As a result of these power users, the average Facebook user receives friend requests, receives personal messages, is tagged in photos, and receives feedback in terms of “likes” at a higher frequency than they contribute. What’s more, power users tend to specialize. Some 43% of those in our sample were power users in at least one Facebook activity: sending friend requests, pressing the like button, sending private messages, or tagging friends in photos. Only 5% of Facebook users were power users on all of these activities, 9% on three, and 11% on two. Because of these power users, and their tendency to specialize on specific Facebook activities, there is a consistent pattern in our sample where Facebook users across activities tend to receive more from friends than they give to others.
- On average, Facebook users in our sample get more friend requests than they make: 63% received at least one friend request during the period we studied, but only 40% made a friend request.
- It is more common to be “liked” than to like others. The postings, uploads, and updates of Facebook users are liked – through the use of the “like” button – more often than these users like the contributions of others. Users in the sample pressed the like button next to friends’ content an average of 14 times per month and received feedback from friends in the form of a “like” 20 times per month.
- On average, users receive more messages than they send. In the month of our analysis, users received an average of nearly 12 private messages, and sent nine.
- People comment more often than they update their status. Users in our sample made an average of nine status updates or wall posts per month and contributed 21 comments.
- People are tagged more in photos than they tag others. Some 35% of those in our sample were tagged in a photo, compared with just 12% who tagged a friend in a photo.
Women make more status updates than men
Women are more intense contributors of content on Facebook than are men. In our sample, the average female user made 21 updates to their Facebook status in the month of observation, while the average male made six.
Facebook users average seven new friends a month
While most users did not initiate a friend request during the month we looked at their activities, and most received only one, an active 19% of users initiated friendship requests at least once per week. Because of the prolific friending activity of this top 19%, the average (mean) number of friend requests accepted was three and the average number accepted from others was four. Overall, some 80% of friend requests that were initiated were reciprocated.
Few unsubscribe from friends’ feeds
Facebook users have the ability to unsubscribe from seeing the content contributed by some friends on their newsfeed. Less than 5% of users in our sample hid another user’s content from their feed in the month of our observation.
There is little evidence of Facebook fatigue
We found no evidence among our sample that length of time using Facebook is associated with a decline in Facebook activity. On the contrary, the more time that has passed since a user started using Facebook, the more frequently he/she makes status updates, uses the “like” button, comments on friends’ content, and tags friends in photos. Similarly, the more Facebook friends someone has, the more frequently they contribute all forms of Facebook content and the more friend requests they tend to send and accept.
Friends of Friends
Your friends on Facebook have more friends than you do
In this sample of Facebook users, the average person has 245 friends. However, the average friend of a person in this sample has 359 Facebook friends. The finding, that people’s friends have more friends than they do, was nearly universal (as it is for friendship networks off of Facebook). Only those in our sample who had among the 10% largest friends lists (over 780 friends) had friends who on average had smaller networks than their own.
Facebook friends are sparsely interconnected
It is commonly the case in people’s offline social networks that a friend of a friend is your friend, too. But on Facebook this is the exception, not the rule. A fully connected list of friends on Facebook would have a density of 1 (everyone knows everyone else). The average Facebook user in our sample had a friends list that is sparsely connected. As an example, if you were the average Facebook user from our sample with 245 friends, there are 29,890 possible friendship ties among those in your network. For the average user with 245 friends, 12% of the maximum 29,890 friendship linkages exist between friends.
Facebook users can reach an average of more than 150,000 Facebook users through their Facebook friends; the median user can reach about 31,000 others
At two degrees of separation (friends-of-friends), Facebook users in our sample can on average reach 156,569 other Facebook users. However, the relatively small number of users with very large friends lists, who also tended to have lists that are less interconnected, overstates the reach of the typical Facebook user. In our sample, the maximum reach was 7,821,772 other Facebook users. The median user (the middle user from our sample) can reach 31,170 people through their friends-of-friends.
Making friends on Facebook is associated with higher levels of social support. Those who made the most frequent status updates also received more emotional support.
In our phone survey, we asked SNS users a variety of questions about their close friends on and offline, the kind of support they received from their friends, the level of diversity of their social circles, and their civic and political activity. We matched the answers to those survey questions to data in these users’ Facebook logs and then analyzed the relationship between certain activities on Facebook and the social lives of these users.
One key finding is that Facebook users who received more friend requests and those that accepted more of those friend requests tended to report that they received more social support/assistance from friends (on and offline). There was also a weak, but positive relationship between receiving and approving friendship requests, as well as posting status updates, and higher levels of emotional support, such as help with a personal problem.
Tagging Facebook friends in photos is associated with knowing more people from diverse backgrounds and having more close relationships – off of Facebook
There is a statistically positive correlation between frequency of tagging Facebook friends in photos, as well as being added to a Facebook group, and knowing people with more diverse backgrounds off of Facebook. These are relatively weak relationships, but they still are statistically significant. Similarly, from our sample, those who tagged Facebook friends in photos more frequently also reported that they had a larger number of people with whom they could discuss important matters (on or off of Facebook).
A wide range of activities on Facebook are associated with attending political meetings
Those users from our sample who are intensive Facebook users are more likely to report that they attended a political meeting or rally. The Facebook activities associated with attending a meeting/rally included: having more Facebook friends, having more friends-of-friends, being added to a Facebook group or adding someone else to a group, sending more personal messages, receiving more wall posts, tagging more friends in photos, and being tagged themselves in photos.
Those who participate in Facebook groups are more likely to try to persuade someone to vote for a specific candidate
Among these users, participation in Facebook groups, either by being added to a group or adding someone else, is associated with trying to influence someone to vote in a specific way.
Survey answers and Facebook logs line up pretty well
Facebook users underestimate the number of their Facebook friends
On average, users in our sample reported in our phone survey that they have 18 fewer friends than is actually the case in their accounts. They reported an average of 227 friends. They actually have an average of 245 friends.
Self-reported survey responses are close to logs of actual Facebook activity
Comparing self-reported survey data to logs of people’s actual Facebook activity, we found that survey data is close to actual use. There is a strong positive relationship between actual and reported use of the “like” button and for commenting on other users’ content. The relationship is slightly more moderate, but still positively correlated for activities that are performed on Facebook less frequently, such as private messaging and status updates or wall posts. Self-report data is generally consistent with actual use, especially for the most popular Facebook activities.