The Ravages of Time

9027427626_e97027ba18_zMy twenty-five year college reunion is right around the corner.  In advance of the event, my classmates were asked to write a short summary of their post college life.  Next to each write-up was the picture from our graduating yearbook twenty-five years ago.  Many of the entries also include current pictures.

Flipping casually through the book, I noticed two things. First, it is amazing how old we all look.  Time really takes its toll, that’s for sure. Second, men were much more likely than women to submit pictures of what they look like now.

There was a third thing that also seemed to be true.  Many of the people who were really attractive twenty-five years ago don’t look so good now.  And even more interesting, there were a surprising number of people who were unattractive in college, but look great (relative to the rest of us geriatrics) now.  If I had been asked to guess, I would have estimated that the correlation between attractiveness twenty-five years ago and today was zero or even negative for women.  For men I would have guessed a small positive correlation.

I was so struck by the pattern that I decided to do a more systematic data analysis.

For the first 100 people who had pictures from today and twenty-five years ago, I first covered up the current pictures and rated the attractiveness of the graduation picture on a scale of 1 to 10.  I then covered up the graduation picture and rated the current picture from 1 to 10.  Then I computed the correlation between attractiveness then and now.  Here were the results:

Correlation for men: .34         Correlation for women: .41

These correlations are lower than I would have predicted before looking at the data, but way higher than I would have guessed after having looked casually at the pictures.  It is a good lesson in how bad the human brain is at making statistical inferences.  My guess is that I dramatically overweighted a few cases of women turning attractive, while underweighting the bulk of the data which was less notable.

Whether these are indeed the true correlations is a different question all together, since sending in a recent picture was voluntary.  I would imagine that the people who look good today were much more likely than the others to submit pictures.  (In case you are wondering, I did not submit an updated picture!)  Whether this “selection effect” operated equally strongly on people who were or were not attractive in college is an open question.  That selection would affect the correlations I report above.

I guess there is only one way to find out: show up at my reunion, pencil and paper in hand, recording data throughout the weekend.


Great article, especially for us who looked awkward until age 24!


Please take that pencil and paper and do the necessary work to complete this project you started and report back.


Many of the "ugly" kids in my class grew up and did one or more of the following: got a good haircut, lost a little weight, had their teeth fixed, learned to apply makeup, acquired some fashion sense and earned enough money to buy clothes not chosen by their parents. Physically they were still pretty much the same people, but a bit of polishing made them appear much more attractive.


It's unlikely somebody's physically the same (or nearly so), and it would make a big difference. A woman who still looks good in a bikini after 25 years and three children is a lot more impressive than somebody who still has an attractive face.

You need to correct for the positive or negative effect someone's level of "popularity" had on their perceived attractiveness. That effect is huge at the high school level, and while smaller at the college level, still pretty large. 25 years later it's largely irrelevant, though professional success may have a similar effect (money certainly does...). This sort of experiment really has to be done with observers who don't know any of the people they're evaluating. Those would be interesting results.


Looking over 45 years of my high school class reunions, I was mostly struck by what has happened to my classmates compared to what I expected. My sample is likely skewed by the fact that people who have not fared well likely avoided the reunions. Mostly, people have done better than I expected. A few with great promise are already dead. There were a few women (of a class of 200) who were still amazingly good looking at age 63. Other hotties had clearly done too many drugs, spent too much time in the sun and/or too little time a the dentist. Seems like the guys fared better, with a fair number looking like slightly heavier versions of their old self, with grey or missing hair. I would estimate that about 75% could easily be identified based on memory of their high school self.


Isn't there also an age bias? That is, as a 23-year-old, I'm more likely to find the younger pictures more attractive, whereas Dr. Levitt might naturally find the older pictures to be more attractive simply because they are closer to his age.


I think I read an explanation for this in some evolutionary-psychology article. Not that I generally put much stock in those theories, but the basic idea is that in females what is attractive are traits that can't be faked that indicate youth which in turn is a proxy for fertility. So things like large perky breasts/butts and long hair are attractive in young women precisely because they age poorly, are hard to fake (at least until recent times), and so are reliable indicators of age. It follows from this that the things that made a woman look pretty at 18, will make her look aged at 43.

For men, attractive traits are those that indicated an ability to gather lots of resources in a primitive society (more pro-athlete -- less Bill Gates). These traits should vary less with time.


I was thinking of the same selection effect question you cite, thinking of these categories: (1) both pictures are in the book; (2) no 1985 picture, yes 2014 picture; (3) yes 1985 picture, no 2014 picture; (4) neither picture. Given those categories:

... What is the relationship between the 1985 attractiveness of groups (1) and (3)? (Disclosure: I am in the Book, and, like you, I am in category 3)
... What about the 2014 attractiveness of (1) and (2)?
... Is likeliness to include more based on an absolute value or on a delta between then and now? In thinking about the missing pictures -- not that we have this data -- would someone who had been attractive and has been more "ravaged by time" (as compared to his or her own standards of attractiveness) be more likely to omit 2014, and someone who didn't love the way he or she looked in 1985 and now wants to a comfort with his or her appearance be more likely to include 2014? Are people who are successful, or involved in certain fields, or even those who were more popular (or less popular) in college more likely to include a 2014 picture?
... And, of course, how does the assessor's knowledge of a classmate affect his scoring? I have to admit that I perceive attractiveness that may not appear in a still photo, if I know of a person's personality, or maybe if I happen to have seen him or her look better in person.

(I'm not winning any looks contests, now or then. I omitted my new pic because I'm lazy...although I do look older than you do!)


jen cohen zonis

Hi Steve,

As a fellow '89er and Leverett House girl, your post reminds me of the original incarnation for Mark Zuckerberg's facebook algorithm "hot or not" (as depicted in The Social Network). I think you left out the variable for 1980's hair; ALL the women of '89 had BIG do's. So naturally we all look a bit better circa 2014. See you in May!


This raises some interesting meta-questions:

1) How many colleges still have yearbooks (or had them in 1989)?

2) What percentage of graduating students would bother getting a yearbook picture?

3) And how many graduates would bother with a class reunion?


Interesting meta-points.

I graduated from a major university in 2007, and to my knowledge, we did not have a yearbook. But, if we did, I wouldn't have bothered getting a yearbook picture. And I wouldn't bother with a reunion, either. I'm already facebook friends with everyone I care to talk to ;).


Wow, excellent work. Thanks for the insights from your unbiased and well-controlled processes.

Keith Neufeld

I would suggest that in person, and to some extent even in pictures, self-confidence is a huge factor in perceived attractiveness; and that the comparison of which people were confident as students versus which are confident as adults is rather interesting.

Chris Foote

I'm not so sure about this one. Hot or not is fun though. It's my understanding that the most important trait in attractiveness is symmetry. I also believe that as one ages, they would be likely to rate older people as more attractive. I also wonder if us freaks associate signs of aging with wisdom whereas a knucklehead may see things differently.


I went to my 40th class reunion (I missed the Big Thirtieth Reunion) and what classmates did show up looked darned old! And there I was, not having changed one bit since graduation! Very sad!

Voice of Reason

I think that one of the reasons for the poor correlation may have to do with students who were athletes in high school, and the popular kid's propensity to drink and party a little too hard. When you're always exercising for a purpose and in general when you have a high teenager's metabolism, you learn terrible eating habits that are hard to break when you're in college and working a desk job. But the kids who never got into sports may have had to watch what they eat a little more. So, when they hit the real world, it wasn't as much of an adjustment to not eat everything in sight.

And, and of the popular, attractive kids developed early dependencies for alcohol, that can ravage their body.


I'm not at all surprised by the results. I'm guessing you're overlooking a certain important type of selection bias: Not just *whether* people send in photos. (As you've imtimated, those who are self-conscious about the way they currently look are less likely to submit a "now" photo). But more importantly, *what* photo they send in. When asked to submit a photo of their choosing, people will always choose the most flattering one they can find. I expect that people at your college didn't get to *choose* their yearbook photo. I recall high school as a time where I was consistently frustrated by how lousy my yearbook photo seemed to turn out, and how I always wished I could choose a more flattering photo to submit in its stead. I consoled myself by saying everyone would always remember their own frustration at this basic, obvious fact of life, and thus would look back on my photo charitably in years to come, immediately recognizing that I didn't actually choose it and that I'm perfectly capable of keeping my (eyes open/chin at a normal angle/eyebrows groomed/hair from sticking up/collar from sticking out askew) on a given day. But your little experiment has made me lose all hope that people remember with accuracy the travails of student life. Ah, well. A reminder to us all to be less vain.



That's an interesting point, but I've noticed the same effect from my facebook friends from high school (I'm not old enough for my college friends to have gotten ugly yet ;)).

A lot of the high school classmates that I had crushes on or found very good looking then look pretty bad now (and facebook confirms that in many cases my memory of their high school attractiveness was not inaccurate). And there are several people who I distinctly remember being average-looking, or not that great looking, who seem to be much more attractive than they were then.

So I tend to think this kind of self-selection bias is not the primary cause here.

Ian M

I would rather spend a weekend in prison than at a class reunion.