What the data actually shows
A study by Almaatouq and colleagues (2016, PLOS ONE) examined friendship networks by asking people to name their friends and then checking whether those friends named them back. The striking finding was that only about half of the friendships people reported were reciprocated — yet participants overwhelmingly expected their friendships to be mutual. There was a large gap between assumed reciprocity, which was very high, and actual reciprocity, which was close to half.
This gap matters because people generally cannot tell which of their friendships are one-sided. The research suggests we are poor at perceiving non-reciprocity in our own relationships, partly because we assume our regard is returned by default. So the experience of discovering or sensing imbalance is common, and the sense that 'this should be mutual' is exactly the expectation the data shows is often inaccurate.
Friendship investment is also not fixed. Beyond reciprocity at a single moment, the effort and closeness in a friendship naturally vary with circumstance — life stage, distance, competing demands, and how much time each person has. Research on friendship and social networks consistently finds that ties wax and wane, so a friendship feeling one-sided at one point is part of a normal, fluctuating pattern rather than a stable verdict on the relationship.
Why this feels different from how it actually is
One-sidedness feels alarming because we carry a strong default assumption that friendship is symmetrical — that if we count someone as a friend, they count us the same way. The Almaatouq research shows this assumption is wrong about half the time, so the feeling of imbalance collides with an expectation we rarely examine, which makes it land harder than the situation warrants.
It also feels different because effort is asymmetrically visible. You always know how much you reach out, plan, and remember, but you only partially see what the other person does — their constraints, their other relationships, the things they do that you never notice. That visibility gap can make a friendship look more lopsided from the inside than it actually is.
And friendship has no clear scorecard. Unlike many relationships, there is no agreed standard for how often friends should contact each other or who should initiate, so any imbalance is easy to over-read. Without an objective yardstick, people tend to interpret ordinary fluctuation as a meaningful signal about how much they are valued.
What the research says to do about it
The most useful correction the research offers is simply normalising: knowing that only about half of friendships are reciprocated, and that people routinely overestimate reciprocity, reframes an occasional sense of imbalance as common rather than as evidence something is wrong. The feeling is widespread precisely because the underlying expectation of near-total mutuality is inaccurate.
It also helps to treat investment as naturally varying rather than as a fixed measure of worth. Because friendship effort fluctuates with life stage, distance, and competing demands, a quiet stretch in one friendship is usually about circumstance and capacity, not about how much you are valued. Reading it as ordinary variation is closer to what the data describes.
None of this means every friendship that feels one-sided is mutual underneath, and the research does not claim that. It supports a calmer baseline expectation — that imbalance is common and not automatically a problem — rather than either suspicion or forced reassurance about any specific relationship.
What the research says does not help
Assuming that a friendship feeling one-sided means something is wrong does not match the data. Because only about half of friendships are reciprocal and investment naturally fluctuates, treating every imbalance as a red flag misreads a common, ordinary pattern as a personal problem.
Keeping a mental tally of who texts first or who makes more plans tends to backfire, because friendship has no agreed scorecard and effort is only partly visible from the outside. Scorekeeping turns normal, fluctuating investment into a running grievance and usually distorts more than it clarifies.
Forcing blanket reassurance — insisting that all your friendships are obviously mutual — is also unsupported, since the research shows people are poor judges of which of their friendships are reciprocated. The accurate stance is neither suspicion of every friendship nor blanket certainty, but the calmer expectation that some imbalance is normal and not a verdict on your worth.
Real numbers in context
The headline finding is about reciprocity. In the Almaatouq et al. (2016, PLOS ONE) study of friendship networks, only about half of reported friendships were mutual — when one person named another as a friend, the feeling was returned roughly 50% of the time — even though participants expected their friendships to be reciprocated at a far higher rate. The large gap between assumed and actual reciprocity is the core result, and these figures come from specific network samples, so treat the ~50% as an approximate, illustrative pattern rather than a universal constant.
Beyond reciprocity at a single moment, friendship investment varies over time. Research on social networks consistently finds that ties wax and wane with life stage, distance, and competing demands, so a friendship feeling one-sided at one point is part of a normal fluctuating pattern. The takeaway is normalising: occasional imbalance is common and not, by itself, evidence that a friendship is failing.