The question isn’t whether you compare. It’s what you’re comparing for.
“Comparison is the thief of joy” is one of the most repeated lines in wellbeing. A new study of partnered adults suggests it’s at least half wrong — and the interesting half is which half.
There’s a sentence in a study published last week that we’ve been turning over ever since.
The researcher who led it, Zachary Reese at the University of San Francisco, was asked about the famous line — comparison is the thief of joy — and said he thinks it’s a myopic perspective. Comparison can sting in the moment, he said, but over the long run it helps us make decisions and build the lives we actually want.
That’s a bold thing to say in a field that has spent a decade telling people to stop measuring themselves against others. So it’s worth looking at what the research actually found.
What they did
Reese and colleagues — Robin Edelstein and Stephen Garcia — were interested in a specific, ordinary kind of comparison: the one you make against the person you live with.
The first study combined three earlier online samples into 808 adults who were currently in a relationship. Everyone answered questions about how often they compared their own abilities to their partner’s, alongside their age, self-esteem, how long they’d been together, how satisfied they were, and how intertwined they felt their identity was with their partner’s.
The second study took 300 partnered adults and added a second kind of comparison — comparing opinions rather than abilities — plus open questions about why people compare at all.
What they found
Ability comparisons — am I as good at this as you are, am I pulling my weight — came up most often among younger people, people with lower self-esteem, people in newer relationships, and people who were less satisfied in them. Closeness itself didn’t predict them either way.
Opinion comparisons went the other direction entirely. People in longer, closer, more satisfied relationships were the ones more likely to compare judgements and attitudes: what would you make of this, how would you read this situation, am I seeing this straight?
When the researchers asked people why they compared, the answers sorted into a handful of recognisable motives. Some were using their partner as a benchmark. Some were watching to learn a skill. A good number were quietly auditing fairness — checking whether the load was even. Others were reading the comparison as a signal about where the relationship was heading. And a smaller group described comparisons they hadn’t chosen at all: the ones handed to them by friends or family pointing out how they measured up. Those involuntary ones showed up most among younger people and newer couples.
Two honest caveats, which the authors state plainly. This is cross-sectional data — a single snapshot — so nobody can say which way the arrows point. Lower self-esteem might drive ability comparisons, or a steady diet of coming up short might wear self-esteem down. Both are plausible; the data can’t separate them. And the measure was self-report: how often people think they compare, not what actually crossed their mind on Thursday afternoon. The researchers name daily diary methods as the obvious next step.
What we keep thinking about
The finding isn’t that comparison is good. It’s that “comparison” was never one thing.
One version is a verdict. You put yourself next to someone and the output is a score — better, worse, behind. The other version is a question. You put yourself next to someone and the output is information: that’s how they’d handle it, that’s a way of seeing this I hadn’t tried.
Same act. Completely different machinery. And the study suggests the verdict kind clusters where things are newer and shakier, while the question kind shows up where people already feel secure. Which is a gentler reading than the usual advice, because it stops treating comparison as a character flaw to be stamped out and starts treating it as something that changes shape as your footing does.
We’d add one thing to that, carefully. If the useful version of comparison is the one that hands you information rather than a verdict, then the most available version of it isn’t against another person at all. It’s against your own earlier weeks.
That comparison has a problem, though: almost nobody has the data for it. We think we remember how the last month went. We mostly don’t. We reconstruct it from the loudest recent thing and whatever mood we happen to be in while reconstructing. Which means the version of ourselves we’re measuring against is largely invented — and, on a bad day, invented unkindly.
That’s the gap Nagi is built for. Mood, sleep, energy — four taps, quick enough to do on a Wednesday when nothing much is happening. Nagi doesn’t score you, rank you, or tell you what any of it means; it isn’t a clinician and it doesn’t diagnose anything. It just keeps the record steady, privately — no name, no email, an anonymous ID rather than a name — so that in a few weeks there’s something real to look back at instead of a guess.
The bit we’d underline
You’re going to compare. That’s not a flaw in you; it’s most of how humans have ever worked out where they stand.
The question worth asking is what you’re doing it for. If the answer is a verdict, it will mostly hurt and mostly teach you nothing. If the answer is information — what changed, what helped, what tends to follow what — then it needs something to work with.
And the one set of evidence you can always gather is your own.
If you want a private place to start noticing your own patterns, Nagi is on Google Play and the App Store — or in your browser at app.nagiandme.app. The first 15 messages are free. No name, no email. Just you, checking in.
— The Nagi team
Source: New psychology study reveals an intriguing shift in romantic partner comparisons over time, PsyPost, 29 August 2026, by Eric W. Dolan. Underlying study: Reese, Z. A., Edelstein, R. S., & Garcia, S. M., “From Insecurity to Inspiration: Predictors of and Motives for Romantic Partner Comparisons,” Journal of Personality and Social Psychology. Study 1 combined three prior online samples (n = 808 partnered adults); Study 2 surveyed 300 partnered adults. Stated limitations: the data are cross-sectional, so the direction of the associations cannot be established; measures were self-reported estimates of comparison frequency rather than momentary tracking; one gender-matching result was based on a subsample of only 37 men and is described by the authors as preliminary.
Nagi is a mental health companion, not a medical device, and nothing here is medical advice. If you’re struggling, please reach out to a professional or someone you trust.