Case Study: Big Brother

Research in tracking trends - T/Tracking


The Big Brother reality show was not only Israel’s most-viewed TV program in recent years, but also the program that generated the highest volume of conversation online. Understandably, the focus of this conversation was not on the program itself or its production (except for one highly-publicized scene involving the program's director), but rather on the show’s contestants. In March 2011, the third season of the program reached its final stage, when only five contestants remained in the "Big Brother" house. One of them would leave with NIS 1 million. We wanted to know if it was possible to predict this winner in advance.


"The wisdom of crowds" is a common term in social psychology, sociology, and economics. The concept behind it, simply put, is that even without control from a higher authority, the general public knows what’s good for the community.

Winning "Big Brother" depended on receiving the most public votes by text message or via the program's official site. We hypothesized that if we monitored online conversation during the week before the program's finale and analyzed the discussions about each of the five contestants, we would be able to rank their relative popularity. The guiding parameters during the assessment were (1) the volume of conversation (how many discussions during the last week mentioned the contestant), and (2) the mix of opinions (how many of these conversations had a positive, negative or neutral tone). Based on these principles, we set ourselves an impressive target – not merely choosing the big winner, but also predicting the eviction of the other contestants in the right order.


A preliminary analysis revealed that a high volume of conversation about a contestant was a prerequisite to win, but was not the sole deciding factor. The contestant who attracted the most conversation about Big Brother in its final months was Frida, a dominant character who was repeatedly involved in disagreements. She generated a lot of discussion, but also many negative reactions. And, sure enough, Frida was evicted from the Big Brother house about two weeks before the finale. Shortly prior to this event, Nofar took over her spot as the most talked-about contestant, also stirring up a high percentage of negative sentiment. This is a good indication of lack of popularity, in the same way that consumers who speak negatively about a particular product will generally try to avoid purchasing it.

A weighted analysis of the volume of conversation about each contestant in the last week of the program, together with an opinion analysis about each personality, enabled us to formulate the following diagram:

Conversation volume about each contestant, together with the opinions expressed about him/her, reflected the popularity and preference for that candidate on the web:

These findings can be described on two levels. Jackie, Lihi and Nofar clearly had higher conversation volumes than the other two contestants. Among these three, there were no significant differences in conversation volume, but Jackie stood out in terms of positive conversation (the relative number of positive conversations from all the conversations about him), which was higher than that of Lihi and Nofar. Actually, Nofar generated more negative reactions than any other contestant on the show.


How could we know in advance how the public would vote on the decisive night? We had no personal inclination favoring one candidate or another. In fact, our researcher had never even viewed one episode of Big Brother. We approached the study without preconceptions or influence from TV commentators, and simply monitored the conversation taking place online. And the volume of online conversation was huge – hundreds of new posts about Big Brother were made every day, each attracting thousands of response threads.

Although the focus of this study was on lighthearted entertainment, it revealed something much more significant than the fate of contestants on a primetime TV show. It was proof that monitoring social media conversation enables us to study the target audience, become closely acquainted with them, and understand their mindset – the factors that guide them, the events that provoke strong emotions, the decisions they make, and the way they make them.

This study showed us that online social media could be used to accurately define the public mood prior to the finales of Big Brother, "Kochav Nolad" ("Israeli Idol"), or any other interactive program. However, the same concept can also apply to matters of the utmost importance, such as parliamentary elections, or even major regional developments like those seen during the “Arab Spring” of 2011.