Loneliness research has had a methodology problem since 1978

In 1978, a small team at UCLA — Daniel Russell, Letitia Anne Peplau, and Mary Lund Ferguson — published a twenty-item questionnaire designed to measure loneliness. The UCLA Loneliness Scale has since become the dominant instrument in the field. By one estimate, it appears in over 80% of empirical loneliness research published in major psychology journals (Hughes et al., 2004).

It has also been quietly measuring something narrower than the popular concept of loneliness.

1. What the scale actually asks

The original UCLA scale asks respondents to rate statements like:

• "I lack companionship"
• "There is no one I can turn to"
• "I am no longer close to anyone"
• "My interests and ideas are not shared by those around me"

These are framed as the experience of social deficit. They do not ask about the quality of relationships, the frequency of interaction, or the number of close confidants. They ask whether you feel alone.

This was an intentional design choice. The scale's authors wanted a measure of loneliness as a subjective state, distinguishable from objective social isolation. Their reasoning held: you can be objectively isolated and not lonely; you can be surrounded by people and intensely lonely.

But the choice has consequences for what the literature has built on top of it.

2. The conflation problem

When public-health writing says loneliness is rising, it almost always cites UCLA-scale data. When the U.S. Surgeon General's 2023 advisory said loneliness was an epidemic, the underlying measure was largely subjective self-report on items like the UCLA scale.

This isn't wrong, but it elides a distinction. Three different things are sometimes labeled "loneliness":

Subjective loneliness — the felt experience of unwanted aloneness. What the UCLA scale measures. Social isolation — objective lack of social contact. Distinct from above. Relationship dissatisfaction — having relationships that aren't meeting needs. Overlaps with both.

These correlate, but not perfectly. Cacioppo and Hawkley's longitudinal work showed that subjective loneliness and objective isolation can move in opposite directions in the same person over time (Cacioppo & Hawkley, 2009).

3. What the alternatives look like

Newer instruments have tried to address the conflation:

The De Jong Gierveld Loneliness Scale distinguishes emotional loneliness (lack of intimate attachment) from social loneliness (lack of broader network). The two subscales correlate but show meaningfully different patterns across demographics (de Jong Gierveld & Van Tilburg, 2010).

The Social Network Index measures objective network size and contact frequency, separate from any subjective measure.

The Lubben Social Network Scale uses six items to assess perceived social network adequacy.

Each measures something the UCLA scale partly covers, partly misses, and partly conflates with other constructs.

4. Why the UCLA scale still dominates

Because it works. It has high internal consistency (alpha typically above 0.90), good test-retest reliability, and predictive validity for outcomes that the field cares about — depression, cardiovascular events, mortality (Holt-Lunstad et al., 2015). The trade-off between conceptual precision and predictive utility has favored the broader instrument.

But the implication for popular reporting is real. When a study says "47% of Americans report feeling lonely," it's saying 47% scored above a cutoff on a subjective-experience scale. That is a feeling. It tells you less than the headlines imply about whether the 47% have anyone to call.

5. The honest version

The "loneliness epidemic" framing rests on real data about a real phenomenon. The exact nature of the phenomenon — and whether the right intervention is more friends, deeper friendships, better community structures, or treatment of underlying depression — depends on which version of loneliness you're measuring. The UCLA scale doesn't fully disambiguate, and the public-health discourse rarely acknowledges that.

The interesting question isn't whether loneliness is rising. It's whether the rising is the same thing we mean when we say the word.