As part of a research module in college we had to come up with some sort of IT based experiment. We were given a few minutes to come up with something and i actually couldn’t think of anything. Blank. Not for the want of trying, I just didn’t like any ideas i came up with, so i was prepared to admit i couldn’t come up with anything but just as the lecturer went around the room asking what our ideas were, lightbulb moment…
I came up with the idea of setting up two fake facebook profiles (male and female). Each profile would have a realistic photo (of a male or female) on it, nothing else. Each of my fake users would then add the same 20 ‘friends’; 10 male, 10 female.
The aim of the experiment was to try and calculate;
- what percentage of people would accept complete strangers as friends.
- to see if females were more likely to accept strange males or vice versa.
It was simple, it was doable, it was pretty interesting (at least i thought so) so that’s the idea i ran with and the lecturer liked it too, encouraging me to actually carry it out which is exactly what i’ve done over the past week 🙂
The Fake Profiles
The two fake users i came up with were both Irish. One male, one female. The photos i got on a stock photography site were both of amateur models. Try to visit their profiles and this is all you could see – just a single profile photo of ‘Rob’ or ‘Katie’ (I tried to give them fairly ‘normal’ Irish names, that matched their profile picture and that wouldn’t raise any suspicion.)
I wanted to add as many friends as possible. I thought 100 was a nice round target. 50 male, 50 female. But then i thought that in one week, i was being totally unrealistic purely because of facebook spam filters. I thought people would report the profiles as spam if i was adding 100 complete strangers as friends. So i decided 20 friends was enough (10 male, 10 female), but just to make sure i wasn’t caught by the spam police, i’d stagger the friend requests and send them over a few days from different IP’s just to throw the Facebook police off any scent they may have been following.
I decided to target specific people, they had to be;
- male or female (sounds obvious, but a lot of companies actually have ‘profiles’)
- aged between 20-30
- have between 50-300 friends (average friend count is about 130 on facebook but i didn’t have the time to find 20 people with a friend count in and around that number)
- be reasonably active on the site and have a lot of recent activity (to ensure they’d respond quickly)
I added 10 females friends to Rob Mc Kenna’s profile at 1am on Wednesday 6th October. At 6pm the same day, i added 10 male friends. At 3pm on Thursday, 7th October, i added the same 10 female friends as i’d added to Rob’s profile to Katie McKenna’s profile. At 1pm on October 8th, i added the same 10 male friends as i’d added to Rob’s profile to Katie McKenna’s profile.
The experiment was fascinating. In total, Rob was accepted by 11 friends of the 20 he’d sent requests to. Katie was accepted by the exact same number. However, they weren’t all accepted by the same people.
- 7 males accepted Rob, 4 females.
- 6 males accepted Katie, 5 females.
- One person sent a private message to Katie asking how she knew them.
- Two people sent private messages to Rob asking how he knew them.
- Rob RECEIVED 1 friend request from a mutual friend of one the females that accepted him.
What can we learn?
- ‘real’ friend counts are totally inflated in a lot of cases if (in this experiment at least) 55% of users are willing to accept strangers as friends.
- people are very open online and aren’t too concerned about privacy issues (younger generations at least), certainly not as much as the media make out – something i’ve blogged about & suggested before.
- males are more likely to accept strange males as friends over strange females.
- very few people (in this case 15%) ask questions before accepting strange friends.
At the start of this, i was expecting more males to accept Katie as a friend, rather than Rob. The opposite happened. I thought that was very surprising. One possible reason for this is that males have become used to getting friend requests from females on other social networks (i’m talking about bebo here).
Females on bebo who approach males often look too good to be true and want to chat with them on web cams. So perhaps most males are used to getting spam on other social networks and could see that Katie was too good to be true. Rob is perhaps seen as less spammy and more genuine because males aren’t used to getting ‘fake’ friend requests from other males.
Most people who accepted Rob, also accepted Katie but that wasn’t always the case. For example despite sending friend requests from Rob & Katie to the exact same people, 1 male accepted Rob but not Katie. Also, one female accepted Katie but not Rob.
Think i could have improved on this or spot any flaws in my experiment? Let me know, i’m interested to see what you guys think and how you would have gone about this or whether you feel it’s an area worth doing further research on….