Comparison, social media and the reference point problem

Social comparison is not a modern failing. It is a well-documented feature of how people assess themselves, and what has changed is the comparison set.

How comparison works

Social comparison theory proposes that in the absence of objective standards, people evaluate themselves against others.

The direction matters. Upward comparison — with people doing better — can motivate or can diminish, depending on whether the other person seems attainable. Downward comparison generally protects self-esteem and provides no useful information.

Neither is inherently good or bad. The problem arises from who is available to compare against.

What changed

Historically the comparison set was people you knew: colleagues, neighbours, family. Roughly comparable circumstances, and their difficulties were visible along with their successes.

The current comparison set is curated presentations from a global population, selected by algorithms that favour the exceptional.

Two distortions follow.

Selection. You see the best outcomes from an enormous sample, presented as though typical.

Curation. You see the outcome without the process, the failures, the resources or the circumstances.

Which produces a comparison of your complete experience against someone else's edited highlights — a comparison that cannot come out well and that feels like an assessment.

What the evidence actually shows

Worth being careful, because this area attracts strong claims.

Large studies of social media use and wellbeing generally find small associations, with effect sizes that some researchers have characterised as comparable to trivial factors.

Causation is difficult to establish: people who feel worse may use social media more, rather than the reverse.

Where the evidence is somewhat stronger is on the type of use. Passive consumption — scrolling without interacting — is more consistently associated with worse mood than active use involving direct communication.

Appearance-focused content and comparison specifically show stronger associations than general use, particularly among adolescents.

The honest summary is that social media is not established as broadly harmful, and that a particular pattern of use — passive, comparison-heavy, appearance-focused — is more consistently associated with feeling worse.

The reference point mechanism

A more useful frame than harm.

Satisfaction depends heavily on the reference point against which an outcome is judged, which is a well-established finding in decision research.

The same income, the same house, the same career progress can be experienced as success or failure depending entirely on the comparison.

Which means that changing what you see changes how you feel about a life that has not changed at all.

That is worth sitting with, because it locates the lever somewhere unexpected: not in improving circumstances but in choosing the comparison.

What helps

Curate deliberately. Unfollowing accounts that reliably produce comparison is not avoidance; it is choosing your reference group, which people have always done by choosing where to live and who to spend time with.

Shift to active use. Messaging people directly rather than scrolling, which is the pattern associated with better rather than worse outcomes.

Compare to your own past. Temporal comparison is available, informative and generally more accurate, because you have full information about both points.

Ask what you would actually want. Frequently the envy is for a specific element rather than the whole life, and separating them clarifies whether anything is actually wanted.

Remember the base rate. The exceptional outcomes are visible because they are exceptional. The distribution they came from is not shown.

The version that is not fixed by individual effort

Worth acknowledging.

Some comparison reflects real inequality rather than distorted perception. Someone comparing unfavourably because they are genuinely paid less for the same work is perceiving accurately.

The advice to adjust your reference point is unhelpful in that case, and the response is to address the inequality rather than the feeling.

Distinguishing the two is the useful first step, and it is skipped by most advice on this subject.

Curating the input

Since the comparison happens automatically, the available control is over what is compared against.

Unfollowing accounts that reliably produce the feeling is not weakness or avoidance. It is the same environmental logic applied to social information.

People generally report that doing so has a larger effect on mood than any amount of trying to think about it differently, which is consistent with everything else known about resisting automatic processes.

General information about wellbeing, not psychological advice. Persistent low mood or distress warrants assessment by a qualified professional.