UCLA Loneliness Scale: Version 3 Items, Scoring, and Interpretation

If you are measuring loneliness in a thesis or a research project, UCLA Loneliness Scale scoring is one of the first things you need to get right. The scale is short, free to use, and trusted by researchers around the world — but its scoring has one trap that catches many students: nine of its twenty items must be reverse-scored before you add anything up. Skip that step, and your results will be wrong in a way that is hard to spot later.

This guide walks you through Version 3 of the scale, the version almost everyone uses today. You will find all 20 items, the exact scoring rules, a worked example, the interpretation bands researchers commonly use, and the reliability evidence you need to justify the scale in your methodology chapter.

What Does the UCLA Loneliness Scale Measure?

The UCLA Loneliness Scale measures subjective loneliness: the distressing feeling that your social relationships are not giving you what you need. That is different from simply being alone. A person can live alone and feel perfectly connected, while another can be surrounded by people and still feel lonely. The scale captures that feeling, not the size of a person’s social network.

It is a self-report questionnaire for adults. Respondents rate how often they feel a certain way, and their answers are combined into a single total score. The whole thing takes about three to five minutes to complete, which is one reason it is so popular in student research — it adds almost no burden to a survey, yet it produces a clean, well-validated measure.

Researchers use it in studies of mental health, social support, ageing, student wellbeing, and the effects of interventions such as befriending programmes. If loneliness is one of your variables, this is very likely the scale your supervisor will expect you to use.

A Short History of the Scale

The scale has been revised twice since it first appeared, and knowing which version you are using matters, because the items and the scoring changed along the way.

The original version was published in 1978 by Daniel Russell, Letitia Peplau and Ferguson at the University of California, Los Angeles. All twenty of its items were worded negatively — every statement described feeling lonely. That created a problem: respondents who tended to agree with everything scored as lonelier than they really were.

The revised version, published in 1980 by Russell, Peplau and Cutrona, fixed this by adding ten positively worded items, so that agreeing with everything no longer pushed the score in one direction. Then, in 1996, Russell published Version 3. He simplified the wording of several items, removed double negatives, and put the phrase “How often do you feel” at the start of every item so the scale would work for a wider range of respondents, including older adults, and in mail and interview surveys as well as in person.

Version 3 is the standard version today. When a paper says “the UCLA Loneliness Scale” without specifying a version, it almost always means Version 3. Use Version 3 in your own research, and say so explicitly in your methodology chapter — it tells your examiner exactly which instrument you used.

The 20 Items of Version 3

Version 3 has 20 items. Each one begins with “How often do you feel” and is answered on a four-point scale: 1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Always. There is no neutral middle option, so every respondent has to lean one way or the other.

The items marked with an asterisk (*) are positively worded — they describe feeling connected rather than lonely — and they are the ones you will need to reverse-score. Here is the full list, exactly as published:

  1. How often do you feel that you are “in tune” with the people around you? *
  2. How often do you feel that you lack companionship?
  3. How often do you feel that there is no one you can turn to?
  4. How often do you feel alone?
  5. How often do you feel part of a group of friends? *
  6. How often do you feel that you have a lot in common with the people around you? *
  7. How often do you feel that you are no longer close to anyone?
  8. How often do you feel that your interests and ideas are not shared by those around you?
  9. How often do you feel outgoing and friendly? *
  10. How often do you feel close to people? *
  11. How often do you feel left out?
  12. How often do you feel that your relationships with others are not meaningful?
  13. How often do you feel that no one really knows you well?
  14. How often do you feel isolated from others?
  15. How often do you feel that you can find companionship when you want it? *
  16. How often do you feel that there are people who really understand you? *
  17. How often do you feel shy?
  18. How often do you feel that people are around you but not with you?
  19. How often do you feel that there are people you can talk to? *
  20. How often do you feel that there are people you can turn to? *

Notice the balance: eleven items describe loneliness, and nine describe connection. That mix is deliberate. It stops the score from being distorted by people who simply agree with every statement they read.

How Scoring Works

Scoring has two steps, and the order matters.

Step 1: Reverse-score the nine positively worded items. These are items 1, 5, 6, 9, 10, 15, 16, 19 and 20 — the ones marked with an asterisk above. For each of these, flip the answer using the formula: reversed score = 5 minus the respondent’s answer. So an answer of 1 (Never) becomes 4, 2 becomes 3, 3 becomes 2, and 4 (Always) becomes 1. The logic is simple: saying you “always” feel part of a group of friends means you are not lonely, so it should contribute a low score, not a high one.

Step 2: Add up all 20 items. The eleven negatively worded items are scored exactly as answered (1 to 4). Add them to the nine reversed items, and you have the total score.

The total ranges from 20 to 80. A score of 20 means the respondent answered “Never” to every loneliness item and “Always” to every connection item — the lowest possible loneliness. A score of 80 is the opposite extreme. Higher scores always mean greater loneliness.

A Worked Example

Imagine a respondent, Sara, who answers the first five items as follows: item 1 = 2, item 2 = 3, item 3 = 2, item 4 = 3, item 5 = 2.

First, reverse-score items 1 and 5: item 1 becomes 5 − 2 = 3, and item 5 becomes 5 − 2 = 3. Items 2, 3 and 4 are scored as answered: 3, 2 and 3. So Sara’s subtotal for the first five items is 3 + 3 + 2 + 3 + 3 = 14.

Now suppose that after reverse-scoring the remaining items, her answers to items 6 to 20 add up to 33. Her total score is 14 + 33 = 47. As you will see in the next section, a score of 47 falls in the moderate loneliness band.

Work through one or two questionnaires by hand like this before you score your full dataset. It is the fastest way to catch a reverse-scoring error, and it gives you a concrete example to describe in your methodology chapter.

How to Interpret the Total Score

The scale has no official clinical cut-off point, and it was not designed to diagnose anyone. The bands below are descriptive ranges used in the research literature (Deckx et al., 2014), and they are a useful guide for describing your sample — not a diagnosis.

Total scoreLevelWhat it suggests
20–34Low lonelinessThe person generally feels connected and content with their social relationships.
35–49Moderate lonelinessSome feelings of disconnection, but not a dominant part of daily life.
50–64Moderately high lonelinessSignificant social isolation; the range where researchers start to pay close attention.
65–80High lonelinessPronounced, persistent feelings of loneliness and social alienation.

Two points matter for your analysis. First, treat the score as a continuous variable wherever you can — report means and standard deviations, and use the total score in correlations and regressions rather than splitting people into groups. Splitting a continuous score into bands throws away information and statistical power. Second, always report your own sample’s mean and spread alongside these bands, because average scores differ between populations: students, older adults and clinical samples do not score the same.

Finally, remember that a high score is a research finding, not a label. If a participant’s score worries you, the right response is to make sure they know where support is available — not to tell them they have a condition based on one questionnaire.

Reliability and Validity Evidence

Before you use any scale in a thesis, your examiner will want to know that it measures what it claims to measure, and that it does so consistently. Version 3 holds up well on both counts, and the evidence comes from the original publication itself.

In Russell’s 1996 validation study, the scale showed internal consistency (Cronbach’s alpha) between .89 and .94 across samples — comfortably above the .70 threshold most textbooks recommend. Test-retest reliability was .73 over a one-year period, which is what you would expect for loneliness: it is a fairly stable feeling, but it does respond to changes in a person’s life, so a perfect test-retest score would actually be suspicious.

On validity, factor analysis supported a single global loneliness factor, which is why the scale is scored as one total rather than split into subscales. The scale also behaves as theory predicts: it correlates positively with depression and negatively with social support and self-esteem, which is exactly the pattern you would expect from a genuine measure of loneliness. It has since been translated into many languages and used in thousands of studies, which is why reviewers rarely question it — provided you cite the original source and report the reliability in your own sample.

One honest limitation to mention in your thesis: because the scale gives a single total, it cannot separate social loneliness (feeling you lack a network) from emotional loneliness (feeling you lack a close bond). If that distinction is central to your research question, you may need a second measure alongside it.

Version 3 vs the Short Forms

If you search for the UCLA Loneliness Scale, you will also meet shorter versions, most commonly the 3-item UCLA-3 and the 8-item UCLA-8. These were created for large surveys where every question costs money or attention — for example, national health surveys that measure dozens of things at once.

The trade-off is precision. A three-item scale can track loneliness in a population, but it is a blunt tool for an individual study. If loneliness is one of your main variables — especially if it is your dependent variable — use the full 20-item Version 3. It gives you a wider score range, better reliability, and more room to detect real differences between groups. Save the short forms for studies where loneliness is a background variable and questionnaire length is a genuine constraint.

How to Use the Scale in Your Research

Using the UCLA Loneliness Scale in a student project is straightforward, but a few practical points will save you trouble later.

Permission: The scale is freely available for research use, and there is no licence fee. You do not need to write to the author for permission to use it in a thesis, but you must cite the original publication. If you are translating it into another language, follow a proper forward-backward translation procedure and report it.

Administration: It is a self-report questionnaire, so it works in person, on paper, or in an online survey tool. The standard instruction is: “The following statements describe how people sometimes feel. For each statement, please indicate how often you feel the way described.” Keep the instruction exactly as published so your results stay comparable with other studies.

Ethics: Loneliness questions can stir up difficult feelings. Mention this in your ethics application, and make sure participants know they can skip questions or withdraw, and where they can find support if they need it.

How to cite it (APA 7th): Russell, D. W. (1996). UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. Journal of Personality Assessment, 66(1), 20–40. https://doi.org/10.1207/s15327752jpa6601_2

In the text of your thesis, cite it as (Russell, 1996) for a parenthetical citation or Russell (1996) for a narrative citation. In your methodology chapter, state the number of items, the response format, the reverse-scored items, and the Cronbach’s alpha you found in your own sample — examiners look for exactly those details.

Common Mistakes to Avoid

1. Forgetting to reverse-score. This is the classic error. If you add up the raw answers, the nine positive items cancel out the eleven negative ones and your scores will cluster meaninglessly in the middle. Always reverse-score items 1, 5, 6, 9, 10, 15, 16, 19 and 20 first.

2. Treating 50 as a clinical cut-off. The bands above are descriptive, not diagnostic. There is no validated threshold that separates “lonely” from “not lonely,” so do not report your results that way.

3. Mixing versions. The 1978, 1980 and 1996 versions have different items. If you copy items from an old questionnaire or a random website, check them against the Version 3 list above.

4. Averaging instead of summing without saying so. Some researchers report the mean item score (total divided by 20) instead of the total. Either is fine, but say which one you used, because a mean of 2.35 and a total of 47 describe the same person and look completely different.

5. Using it with children. The scale was designed for adults. It can be used with older teenagers in some studies, but check the validation evidence for your age group before you assume it applies.

Frequently Asked Questions

How many items does the UCLA Loneliness Scale Version 3 have?
It has 20 items, answered on a four-point scale from Never (1) to Always (4). Nine items are reverse-scored, and the total score ranges from 20 to 80.

Is the UCLA Loneliness Scale free to use?
Yes. The scale is freely available for research use at no cost. You do not need a licence, but you must cite Russell (1996) as the source.

What is a good score on the UCLA Loneliness Scale?
There is no pass or fail. Lower scores mean less loneliness: 20–34 is generally described as low, 35–49 as moderate, 50–64 as moderately high, and 65–80 as high. Compare your sample’s average with published norms for a similar population rather than judging a single score in isolation.

Can I use the UCLA Loneliness Scale in an online survey?
Yes. It is a self-report questionnaire, so it works well in online survey tools. Keep the original wording and the four response options unchanged so your results stay comparable with other studies.

Does the scale diagnose loneliness or depression?
No. It is a research measure, not a diagnostic tool. A high score means the person reports feeling lonely often; it does not diagnose loneliness as a condition or depression as a disorder. Any concerns raised by a score should be followed up by a qualified professional.

Conclusion

The UCLA Loneliness Scale Version 3 earns its place as the standard loneliness measure for a simple reason: twenty clear items, five minutes of a respondent’s time, and scoring rules that anyone can follow once they know about the nine reverse-scored items. Get the reverse scoring right, keep the score continuous in your analysis, and cite Russell (1996) properly, and you will have a measure your examiner can trust.

If loneliness is at the heart of your research, this scale gives you a solid foundation. And if your own score ever lands higher than you expected while you are testing the questionnaire, take it as a useful reminder of why this research matters — then go and talk to someone. Good research starts with honest measurement, and honest measurement starts with understanding your instrument.

Leave a Comment