Profile Picture Experiment

Introduction

Recently I decided to test out a theory by my friends Race and Kelly on the Plenty of Fish dating site. The concept is simple. Basically it states that the more quality pictures you have visible on your online profile, the more value you display and thus the more dates you will be able to receive. For example, if you applied this concept to plentyoffish.com then it would mean that a person who displays the maximum 8 pictures in their profile is more likely to create attraction than a person who displays only one picture.

Luckily, POF comes with an easy way to measure a profile’s attraction. If we can measure the attraction of a profile and we know how many pictures those same profiles display, it shouldn’t be too difficult to come up with some type of measure between the quantity of pictures displayed and the level of attraction.

How do we measure attraction on a Plenty of Fish profile? Simply by looking at the number times a profile has been “favorited.” When someone favorites you, that is an indicator of interest or attraction. So, we can assume that profiles with a high number of favorites should be considered highly attractive individuals and profiles with a low number of favorites would be considered less attractive.

Procedure

The first step in this experiment was to gather all the data. I went through the PlentyofFish site and selected random men and women from across the country and recorded both the quantity of pictures and number of times favorited for each profile. In all, I collected data from approximately 200 sets of male profiles and 200 sets of female profiles.

I then sorted each data set (one for each gender) by the quantity of pictures displayed and took an average of the “Favorites” for each quantity group.

I then graphed the correlation between the quantity of profile pictures and the # times favorited.

Data

Click to view all data in spreadsheet format

Results

Looking at the male graph, it is pretty clear that there is a positive correlation between quantity of profile pictures and attraction. It shows that for each picture, attraction level or “favorites” will on average increase by the same amount.

The female graph, however, is a bit more crazy. There is certainly an upward trend in # times favorited but has a few erratic swings with a big dip at 5 pictures, a giant spike at 6 pictures and then dropping again at 7 and 8 pictures.

Discussion

This isn’t the most scientific experiment and there is certainly several opportunities for bias to occur in the data results. A few possible flaws are as follows:

  • It is possible that men and women who are perceived at being better looking may be more comfortable displaying more pictures of themselves. This would produce bias in the data as better looking people who display more pictures would naturally receive a higher number of favorites just off of looks alone.
  • The range of # times favorited in the women’s data is much greater than that of the men’s data. Some women’s profiles swing into the triple digits with the highest having over 500 favorites. These wild upswings for some women will have an impact on the average for those particular picture groups. This would explain for the big spikes in the graph.

Conclusion

There appears to be a positive correlation between number of profile pictures displayed and attraction level though certainly not proven. Also, this only applies to an average profile, as it is clear that some profiles with many pictures may still be viewed as unattractive and profiles with few pictures could be seen as highly attractive.

I’ve added in the original Plenty of Fish data sets above, so if there is anyone else out there with a better statistics background who’d like to take a stab at analyzing the data let me know.

10 thoughts on “Profile Picture Experiment

  1. I’d say the swings in the women’s graph are irrelevant. You’re dealing with a small number of data points (8) so little spikes are just noise. The trend is pretty clear.

    I agree that better looking people are more likely to post more photos. I’d also suggest that as people who post multiple photos often post different looks, or situations, there’s content in multiple images which isn’t in single photos. That’s going to make it more likely those people are favorites, you know more about them (as they say in movies, show – don’t tell).

  2. The data look like they aren’t normally distributed, especially among women. Using the median instead of the mean, or log-transforming the data, might help with the skewedness and tone down the swings in the women’s graph.

  3. @Samuel Agboola,
    Yea, having lots of different content displayed in multiple images is a big one. It’s like that saying, “A picture is worth a thousand words.” So, if one picture is worth a thousand than 8 pictures would be worth 8 thousand words, you can’t beat that!

    @Julia,
    Good point, I’ll have to go back and look at the medians for each picture grouping and re-do the graph.

  4. the female “favorite” data is sort of like the homecoming queen/prettiest girl phenomena…just like in high school, men tend to assess attractiveness in the opposite sex as much according to their own proclivities as to those of their peers–which can be largely influenced by the opinions of their peers (and so on, and so on…). whereas women may ADORE sarah jessica parker or find jude law iuber-sexy, most men find SJP assymetrical, opting instead for the cosmetically-enhanced impossibility of a pamela anderson. and, if pressed, would unanimously concur that george clooney is the sexiest man alive.

  5. You would expect that the “picture quality” of pictures for males to have a positive effect. Mostly that translate to level of effort which women like to see.

    Discounting the males who are classically handsome, (appeals to a larger range of women), the advantage of multiple pictures is to hopefully “catch” a specific look/feature that various women may find attractive.

    There are, of course only so many looks that any one person can portray … I suspect 3-5 pictures should suffice. The straight line effect you’re seeing is more likely to one possibility that you mentioned … better looking people are more likely to post more pictures. As well, males posting many pictures are more likely to pay attention to picture quality.

    From the female side I would suspect the number of pictures to have little effect. She is either “attractive enough or not”. Women have far more attraction cues than men, in terms of what’s attractive to the opposite sex AND men generally only need to see a few attraction cues to create an interest.

    I think what you’re picking up on in your experiment is, women see men as attractive defined by what they do and who they are rather than just physical appearance alone.

  6. The most interesting part of this is the fact that women get MUCH more attention then men do on this (and every other) site.

    Then again, any guy's who's tried online dating knows that.

    I recently read an article claiming that more 9 out of every 10 e-mails that are sent on these website are sent by men.

    The men make an effort while the women ignore them .

    Why the women even bother to join baffles me.

  7. women join dating sites because they are desperate, or realize that they are getting no “bites” in the real world.

    interestingly, once online, they find that they are bombarded by interested men and begin to get an artificial ego boost. where as in the real world she’s lucky if a guy looks at her, or pumps and dumps her, online she’s a queen and has the pick of the litter!

    subsequently she begins to get a unrealistic assessment of her attractiveness, and desirableness.

  8. Okay, I’m with “that guy,” as far as the average women and bites… But what about these women that say, “I’m sick of the guys I meet in the bars, etc.” that are GORGEOUS! I know that many of them must be fake (the very obvious professional photography is one clue) but some of them are real, right? And while they might not be exactly what they look like, they are going to be somewhat atractive. Why are they still on the website?

Comments are closed.