How to score the UCLA 3-item loneliness scale

Search for the UCLA loneliness scale and you quickly hit a problem: there are two of them. One has 20 questions and a score that can climb to 80. The other has just three questions and tops out at 9. They share a name and a research heritage, but they are different tools with different scoring, and the internet mixes them up constantly. People paste the long scale’s 20-80 range onto the short form, or score the short form’s three items on a 1-to-4 scale that belongs to the other version. The result is a number that looks official and means nothing.

This article fixes that. It is a clean reference for the 3-item form specifically: the three questions, the correct 1-to-3 coding, the 3-to-9 total, and one caveat that most pages quietly skip. If you just want your number now, the UCLA 3-item loneliness test scores it for you in under a minute and runs entirely in your browser.

Three questions, three points each, 3 to 9 total

The short form comes from a 2004 study by Hughes, Waite, Hawkley, and Cacioppo in Research on Aging (PMID 18504506). They needed a loneliness measure short enough for a telephone survey, so they distilled the long UCLA scale down to three items that still captured most of what it measured.

Here are the three questions, each answered the same way:

1. How often do you feel that you lack companionship?
2. How often do you feel left out?
3. How often do you feel isolated from others?

Each answer is scored:

• Hardly ever = 1
• Some of the time = 2
• Often = 3

Add the three together. The total runs from 3 (hardly ever to everything) to 9 (often to everything). That is the entire scoring system. Higher means lonelier, more often. Notice what the questions do not contain: the word lonely never appears. That is deliberate, because people under-report loneliness when asked about it head-on, so the scale gets at the feeling indirectly.

Why it is not the 20-item scale

The confusion has a real source, and it is worth understanding so you never make the mistake. The parent instrument is the Revised UCLA Loneliness Scale, published by Russell, Peplau, and Cutrona in 1980 (PMID 7431205). It has 20 statements, each rated on a 1-to-4 scale, for a total that ranges from 20 to 80. It is a thorough, well-validated measure, and it is the one most psychology textbooks describe.

The 3-item version is its compact descendant, not a different copy of the same test. The differences that matter for scoring:

If a website tells you a score of 6 puts you in a certain category without telling you which version it scored, be suspicious. A 6 sits near the high end on the 3-item scale but near the rock bottom on the 20-item one. The number is only meaningful once you know which test produced it. The UCLA 3-item loneliness test on this site uses the short-form 1-to-3 coding throughout, so a result there is always on the 3-to-9 scale.

The caveat nobody mentions: there is no official cutoff

Here is the part that gets left out of almost every quick summary. There is no official, validated clinical threshold for loneliness on this scale. Hughes and colleagues built the short form to measure loneliness as a continuum across a large population, not to sort people into lonely and not-lonely boxes. They never published a diagnostic cut score, because loneliness is not a diagnosis.

What you will see, in research papers, is a convention: some studies group scores of 6 or higher as lonely for the purpose of analysis. That is a practical line drawn to split a sample into groups, not a medically meaningful boundary. Crossing it does not mean you have a condition, and staying under it does not mean you are fine. The honest reading of a higher score is simply this: you report feeling lonely more of the time, and that is worth paying attention to.

So treat any single number, including your own, as a flag rather than a label. A 7 is not a diagnosis. It is a prompt to look at your connections and ask whether they match what you want.

What a high score actually means for your health

If loneliness has no diagnostic cutoff, why measure it at all? Because the experience, tracked over time, is one of the better-supported risk signals in public health. A 2015 meta-analysis by Holt-Lunstad and colleagues (PMID 25910392) pooled data on hundreds of thousands of people and found that loneliness, social isolation, and living alone each raised the risk of early death by roughly a quarter to a third, an effect size comparable to well-known physical risk factors.

That finding helped push loneliness onto the public-health agenda. In 2023, the U.S. Surgeon General issued a formal advisory, Our Epidemic of Loneliness and Isolation (HHS, 2023), framing chronic loneliness as a serious, addressable health issue rather than a personal failing. The point of a short scale is not to brand you, but to make a usually invisible state visible enough to act on.

One important distinction sits underneath all of this. Loneliness is not the same as social isolation. Isolation is objective, the number of people you actually interact with. Loneliness is subjective, how you feel about your connections regardless of the count. You can be surrounded and still lonely, or quietly alone and perfectly content. This scale measures the feeling, not the headcount, which is exactly why it can flag something a contact diary would miss.

How reliable can three questions be?

It is fair to be skeptical that three questions can measure something as layered as loneliness. The reassuring answer is that the short form was validated against the long one and holds up well. Hughes (2004) showed the 3-item total tracks the full 20-item scale closely, and a 2023 study by Daniel and colleagues in Heliyon (PMID 37215896) confirmed solid internal consistency and reliability in community-dwelling adults.

Three items will never capture every shade of the experience, and that is the trade-off you accept for speed. What the short form is good at is screening and tracking: a quick, repeatable reading you can take now and again in a month to see which way things are heading. For that job, the evidence says it works.

How to read your own result calmly

A score is a starting point, not a conclusion. A few ways to keep it in proportion:

1. Know your range. Your number is on the 3-to-9 scale. If a source quotes you anything outside that, it scored the wrong test.
2. Read the trend, not the snapshot. One reading is a moment; two or three over time tell you whether things are improving. Loneliness rises and falls with life events.
3. Separate the feeling from the count. A high score reflects how you feel, not a verdict on your social life or your worth.
4. Act small and often. Research on connection points to repeated, low-effort contact, one regular call, a shared activity, reconnecting with someone you drifted from, over grand gestures.
5. Watch for overlap with mood. Loneliness often travels with low mood. If sadness, hopelessness, or loss of interest are also present, a depression screener like the PHQ-9 or a conversation with a clinician can help sort out what is going on.

If loneliness is persistent and weighing on you, that is a good reason to talk to a doctor or mental-health professional, not because the scale diagnosed anything, but because chronic loneliness is treatable and worth addressing. And if you ever have thoughts of harming yourself, contact a local crisis line or emergency services straight away. This is a screening tool, not a substitute for care.

The takeaway

The UCLA short loneliness scale is three questions, scored 1 to 3 each, summed to a total between 3 and 9. It is not the 20-item version, and the two scores never mean the same thing. There is no official clinical cutoff, so read your number as a continuous signal rather than a label: higher means you feel lonely more often, which is worth noticing. The feeling is common, it is not a moral failing, and the evidence that it matters for health is exactly why a one-minute check is worth taking seriously.