Comparisons are futile, people say, and they usually say it to someone who has just made one and felt worse for it. It is meant kindly. It is also the wrong word, and the right word is more useful.
Futile means pointless, without effect. Sixty years of research says comparison is neither. It has a measurable effect, in a consistent direction, and people keep choosing to do it even when they are given the option not to. The interesting question is not how to stop. It is which comparisons do what.
What people actually choose
James Gerber, Ladd Wheeler and Jerry Suls gathered 145 studies and 286 effect sizes covering six decades of social comparison research, and published the synthesis in Psychological Bulletin.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
One part of that literature, the selection studies, does something simple: it offers people a choice about who to compare themselves with, and records which way they go. Given the option of looking up or looking down, people looked up 75.6 percent of the time.
What makes that figure interesting is the finding attached to it. For decades the theory ran that people compare downward to feel better, especially when they feel threatened. Gerber and colleagues found no evidence that downward comparison dominates even under threat. People choose the version that is more likely to make them feel worse, and they do it most of the time.
What it does to you
The dominant response across the literature was contrast, which is the technical word for the thing everyone recognizes: you measure yourself against them and come away smaller. Overall, the effect came out at d = -0.37.
One detail in that section is worth carrying home. Comparison with people close at hand produced substantially stronger contrast on ability, in the range of d = -0.64 to -0.84, than comparison with distant figures at -0.23 to -0.40. The neighbor with the new extension does more damage than the billionaire. The gap that hurts is the one you could plausibly have closed.
The question the authors leave open
Now the honest part, and the reason “futile” has some intuition behind it.
Gerber and colleagues state the puzzle in their own paper and do not solve it: why do people choose to compare upward when the most likely result is self-deflating contrast? They gathered sixty years of evidence that we reliably pick the option that makes us feel worse, and they cannot say why.
So anyone who has caught themselves scrolling through an old classmate’s career and wondered what on earth they are doing is in good company, including the people who study it professionally.
The variable that flips the sign
The contrast finding is an average, though, and averages hide the interesting cases. A set of experiments from 1997 found a condition under which the same comparison does the opposite.
Penelope Lockwood and Ziva Kunda, writing in the Journal of Personality and Social Psychology, gave accounting students a newspaper profile of an outstanding fourth-year accounting student, and then asked them to rate themselves.
The first-year students rated themselves higher afterward, 8.90 against a control average of 8.19, and 82 percent of them described the profile as inspiring. The fourth-year students read the same profile. Six percent found it inspiring. Half of them criticized the comparison itself, against 6 percent of the first-years.
The difference between the two groups was not the target. It was whether that target’s success was still in front of them or already behind them. A third study found the same pattern using a different lever: participants who believed ability is improvable were lifted by an outstanding peer, while those who believed ability is fixed were not.
What that study does not show
It is worth being exact, because this result is often reported more dramatically than it deserves.
The fourth-year students did rate themselves lower than controls, but the difference was not statistically significant, and the authors say so plainly. Their reading is that those students defended themselves successfully, largely by attacking the comparison, which is what the denigration figures suggest. So the demonstrated finding is that an attainable model lifts people. That an unattainable one damages them is suggested rather than shown.
The meta-analysis has its own caveats. Few of the studies it pooled included no-comparison control groups, which limits how cleanly the effects can be separated, and both of its tests for publication bias came back significant. The samples in the Lockwood and Kunda experiments were small, at 50, 65 and 57 undergraduates.
A better word than futile
The evidence does not support pointless. It supports something more specific and more manageable.
Comparison is close to unavoidable, it runs upward by default, its usual effect is to make you feel smaller, and that effect is strongest when the person is near you rather than famous. But the sign of it turns on whether the thing still looks reachable from where you are standing, and that is a fact about your reading of the situation rather than about the other person at all.
Which suggests the useful question is not whether to compare, since on this evidence you are going to. It is whether the particular comparison in front of you is one you could still act on, or one that is now only a way of hurting yourself about a road not taken.