Automated Data Collection Strategies on Groups

I am a researcher in [a country] working on a Facebook-based, agenda-setting study. Ideally, I would like to perform an analysis of participation in a specific, public FB group but I have run into a snag. I am trying to capture all of the content on the FB group's wall without resorting to tedious copy and paste methods. Do you know of anyone who has successfully worked with this type of data set or how one would go about automating the copying process?

As far as I know it’s not possible to use webcrawlers or automated data collection programs with Facebook – they have mechanisms built into the site to block or ban users who exhibit inhuman tendencies (like being able to follow a hundred links at a time). Offhand I can think of two workarounds, though. I believe you can set Facebook Page (as in the Facebook Pages platform application, a front/service used by companies and other stakeholders - http://www.facebook.com/advertising/?pages) to push out to an RSS feed. I think that’ll only capture wall content put up by the page operators, however. The other way you might go about it could be to write a client-side script with Greasemonkey and Firefox or something similar. I think ideally you’d have to visit the page to download it and then perform some kind of programmatic operation on it to sort or refine the data. It might be possible to write a script that could be used in conjunction with a user account that’s always logged on – it would refresh the page say every hour, download it, parse through the code to record the posts, stopping if it encounters an identical one, and appending the text into a file. Lots of work but it might be possible to do without flagging Facebook’s bot-catching mechanisms.

Honestly I haven’t done anything like this so I’d suggest you try contacting Eric Gilbert (at UIUC) or Fred Stutzman (UNC Chapel Hill) and asking them, they’re quite a bit more familiar with the security and programming side of Facebook research.

Compliments and Comments on Facebook

I am currently writing, or trying to write a paper on compliment responses in the facebook context, which is for my Masters programme. i am practically going bonkers trying to figure out and collect the data for my paper. so, i am hoping that u cud help me. is there any researchers doing a paper on communication in facebook or compliment responses in facebook? and it would b great if u cud give me some tips or points on this matter. thank u very very much

I’m not sure if I understand what you mean when you say ‘compliment responses.’ Are you referring to the comments people can leave in response to posted statements, links and media?

Regardless, I think the question of the best way to collect data is a good one. You should probably pick a research method that fits your research questions and approach. If you have a hypothesis you’d like to test you might think about laying it out in explicit statement form and then determining the criteria or variables that would apply. For instance, you might copy and paste several pages of Facebook wall posts from your newsfeed into a text file and then count the number of times people respond to each other, anticipating that, say, the more a person posts the more likely they are to get responses. You could deepen the analysis by codifying certain words as tied to a type of emotion and then look for changes in the amount of type of responses between individuals. This sort of thing would be done programmatically, and be considered to be a quantitative content analysis. You’d have to be careful about what you say about your results, in this case there would be sample limitations.

Alternatively you might not worry about getting a lot of data, but instead looking closely at a small amount of data to see what it presents. I suspect this angle might work better for you since you seem to be unsure of the best way to go about your project. If you work inductively, and start by making a lot of observations of something you can eventually start to find consistencies and ongoing themes. With enough observation you can begin to construct theories of the greater narratives afoot. For instance, you might take a selection of wall post exchanges that involve comments that happen to be compliments and see what they say. Are they usually initiated by people of one gender? Are the compliments mostly about one type of thing, like a person’s appearance? What might these exchanges say about the discourse of compliments, or the way people think about complimenting one another? The goal here would not be to make global statements based on generalizable data, but instead to help shed some light on the way some people make sense of things or what compliments implicitly mean to them. Or how people are able to give and receive compliments on Facebook.