PSS-10 (Perceived Stress Scale): Take It, Score It, Understand Your Results
The PSS-10 is the 10-item form of the Perceived Stress Scale, a self-report questionnaire that measures how unpredictable, uncontrollable, and overloaded you have found your life over the past month. Developed by Sheldon Cohen and colleagues at Carnegie Mellon University in 1983, it is one of the most widely used psychological stress measures in the world and has been translated into more than 40 languages. The PSS-10 takes about 3 minutes to complete and produces a 0-40 score. Cohen's own lab is explicit that the PSS is not a diagnostic instrument and there are no formal cutoff scores; widely cited interpretation bands (0-13 low / 14-26 moderate / 27-40 high) are informal benchmarks used in research and self-reflection, not clinical thresholds.
What is the PSS-10?
The PSS-10 is a brief stress questionnaire developed by Dr. Sheldon Cohen and colleagues at Carnegie Mellon University, first published in 1983 with the 14-item original PSS-14 and later refined into the 10-item PSS-10 short form that is now the most commonly used version. It is free to use, available in more than 40 languages, and remains one of the most cited psychological stress measures in research worldwide. The PSS family was designed to capture how stressful a person finds the events in their life — not how many stressful events they have experienced — which is why it asks about feelings of unpredictability, lack of control, and overload rather than about specific life events.
What the PSS-10 measures
The PSS-10 measures perceived psychological stress over the past month. The items cover feelings of being upset by unexpected events, feeling unable to control important things in life, feeling nervous and stressed, feeling confident about handling personal problems, feeling that things are going your way, feeling unable to cope with everything you need to do, feeling in control of irritations, feeling on top of things, being angered by things outside your control, and feeling that difficulties are piling up too high to overcome. Because four items are worded positively, those items are reverse-scored before summing. The construct is sometimes described as having two underlying factors — perceived helplessness and perceived self-efficacy — that combine into the global score.
How the PSS-10 is administered
The PSS-10 is a self-report questionnaire that takes about 3 minutes to complete. Each of the 10 items is rated on a 0-4 scale (0 = never, 1 = almost never, 2 = sometimes, 3 = fairly often, 4 = very often) over the past month. Items 4, 5, 7, and 8 are positively worded and must be reverse-scored (0 becomes 4, 1 becomes 3, and so on) before the total is calculated. The total score is the simple sum of all 10 items after reverse-scoring, with a possible range of 0 to 40. You can complete the PSS-10 on paper or on a screen; no special preparation is required.
Who uses the PSS-10
The PSS-10 is used extensively in research on stress, health, and well-being, including studies linking perceived stress to outcomes such as cardiovascular disease, immune function, and mental health. In clinical settings, it is sometimes used in primary care, behavioral health, and chronic disease management as a brief, low-burden way to track perceived stress over time. Workplace wellbeing programs, university counseling centers, and public health surveys also use it because it is short, free, and reasonably comparable across populations. Many digital wellness platforms include the PSS-10 as part of self-tracking dashboards, though it was designed for periodic assessment rather than daily measurement.
PSS-10 is a research tool, not a diagnostic instrument
Cohen's own laboratory is explicit on this point: the PSS is not a diagnostic instrument and no formal cutoff scores have been established. Interpretation bands that group scores into low, moderate, and high perceived stress (typically 0-13, 14-26, and 27-40) are informal benchmarks based on population averages — they describe where your score sits relative to others, not whether you have a stress disorder or any other clinical condition. If your PSS-10 score is high or you are concerned about how stress is affecting your physical or mental health, the next step is to talk to a clinician, not to self-diagnose from the score alone. Stress can be a normal response to life events, but persistent high stress is associated with poorer health outcomes and is worth attending to.
How to score and interpret your results
Each of the 10 items is rated on a 0-4 scale (0 = never, 1 = almost never, 2 = sometimes, 3 = fairly often, 4 = very often), and the total is the sum of all 10 responses after reverse-scoring the positive items. The range is 0 to 40; higher numbers mean greater perceived stress over the past month. Cohen’s lab requires 75-85% data completion to compute a score — at least 8 valid responses on a 10-item form.
The single most common scoring mistake is forgetting to reverse the positive items. Items 4, 5, 7, and 8 are reverse-scored before summing. Reverse-scoring flips the value on the 0-4 scale: 0 becomes 4, 1 becomes 3, 2 stays 2, 3 becomes 1, and 4 becomes 0. For example, item 4 asks how often you felt confident handling personal problems — a “very often” answer (4) is recoded to 0 before being added to your total, because feeling confident is the opposite of feeling stressed and a high raw score there should lower your total, not raise it.
Cohen’s own laboratory at Carnegie Mellon is unambiguous on what the resulting number means:
“The Perceived Stress Scale is not a diagnostic instrument; there are no score cut-offs.”
That caveat matters because most online versions of the PSS-10 present the following informal bands:
| Band | Score | Common label |
|---|---|---|
| Low perceived stress | 0-13 | Below-average for the past month |
| Moderate perceived stress | 14-26 | Roughly typical for general-population samples |
| High perceived stress | 27-40 | Notably above-average for the past month |
For scale: the largest representative validation of the PSS-10 (Klein and colleagues, German general population, N = 2,463) reported a sample mean of 12.57, SD 6.42 — so the average person sits at the top of the “low” band, and a one-SD swing in either direction crosses into “moderate”. Read your score relative to a comparison group, not as a clinical verdict.
The 10 PSS-10 items: helplessness and self-efficacy factors
Cohen and colleagues originally framed the PSS-10 as a single global measure, but subsequent psychometric work has consistently shown that the 10 items load on two related latent factors rather than one. Both contribute to the same 0-40 total but describe different aspects of the stress experience.
Factor 1: Perceived Helplessness (items 1, 2, 3, 6, 9, 10)
These six negatively worded items ask how often you felt upset, unable to control important things, nervous and stressed, unable to cope, angered by things outside your control, and overwhelmed by difficulties piling up. High scores here reflect the felt experience of being on the receiving end of stress — that life is unpredictable, demands exceed your resources, and the situation is not yours to fix.
Factor 2: Perceived Self-Efficacy (items 4, 5, 7, 8)
These four positively worded items ask how often you felt confident about handling problems, in control of irritations, on top of things, and that things were going your way. They tap the opposite construct: a sense of agency and effective coping. Because high scores here mean less stress, they are reverse-scored before being added to the total.
The two factors are correlated but distinct. In the Klein 2016 German validation the inter-factor correlation was r = 0.47, and confirmatory factor analysis showed excellent fit for the two-factor model (CFI = 0.96, TLI = 0.95, RMSEA = 0.07). The Milo 2025 systematic review of 20 non-English PSS-10 translations replicated the two-factor structure across seven confirmatory analyses, with variance explained ranging from 41.5% to 67.8%.
This matters clinically. Two people with the same total can have very different profiles: a 20 driven by high helplessness reads differently from a 20 driven by low self-efficacy. Hewitt and colleagues (1992) reported that helplessness items predicted depression in both men and women, while self-efficacy items predicted depression only in women.
What a high PSS-10 score means — and what it doesn’t
A high PSS-10 score tells you one specific thing: in the past month, you appraised your life as unpredictable, uncontrollable, and overloaded more often than most people do. It does not tell you why, what condition is causing it, or what to do about it. As Cohen’s lab states plainly:
“The Perceived Stress Scale is not a diagnostic instrument; there are no score cut-offs.”
The PSS-10 is anchored in Lazarus’s transactional model of stress — the idea that stress emerges from a person’s appraisal that demands exceed coping resources. It quantifies appraisal, not disease.
What elevated scores are associated with at the population level
Higher PSS scores have been linked in research to a range of biological and clinical outcomes:
- Telomere shortening, a marker of cellular aging
- Cortisol elevation
- Immune dysfunction
- Increased susceptibility to infectious illness
- Slower wound healing
- Elevated prostate-specific antigen
- Higher rates of depression
These are population-level associations, not personal predictions. A high PSS-10 score does not mean any of these outcomes will happen to you.
Overlap with depression and anxiety
Perceived stress tracks closely with adjacent mental-health constructs. Klein 2016 reported correlations of r = 0.59 with depressive symptoms (PHQ-2), r = 0.59 with anxiety symptoms (GAD-2), and r = -0.47 with life satisfaction. The Milo 2025 review found convergent correlations above r = 0.50 with depression, anxiety, neuroticism, and other stress instruments. The overlap with depression and anxiety means a high score is a signal to look more carefully — not a diagnosis. For a screen that bundles depression, anxiety, and stress into one short questionnaire, the DASS-21 is designed for that pairing.
How accurate is the PSS-10?
The original Cohen, Kamarck, and Mermelstein 1983 paper introduced the scale in the Journal of Health and Social Behavior, and the Milo 2025 systematic review of 20 non-English translations has documented its properties in detail since. The PSS-10 has held up well across more than four decades as a brief, reliable measure of perceived stress.
Internal consistency (Cronbach’s alpha)
Cronbach’s alpha measures whether items hang together as if tapping the same construct. Values above 0.70 are usually considered acceptable; values above 0.80 are considered good.
| Source | Population | N | Cronbach’s α |
|---|---|---|---|
| Cohen 1983 (English original) | US adults | — | 0.78 |
| Milo 2025 (20 translations) | Various | varied | Weighted 0.82; range 0.63-0.88 |
| Klein 2016 (German community) | Adults 14-95 | 2,463 | 0.84 |
| Roberti 2006 (US university) | Students | — | 0.85 helplessness / 0.82 self-efficacy |
The 0.63 lower bound came from a single Malay-language sample of nurses; 11 of 20 translations scored α ≥ 0.80, and 19 of 20 cleared the 0.70 threshold.
Test-retest stability
Across one- to three-week intervals, test-retest stability coefficients compiled in the Milo 2025 review ranged from 0.74 to 0.93, consistent with the PSS-10’s design as a measure of recent appraisal rather than a fixed trait. Longer intervals produced weaker stability — a Vietnamese sample retested at one month showed a Spearman correlation of only 0.43.
Construct and convergent validity
The PSS-10 correlates with related constructs in expected directions:
- Depressive symptoms: r ≈ 0.59
- Anxiety symptoms: r ≈ 0.59
- Fatigue: r ≈ 0.57
- Procrastination: r ≈ 0.42
- Life satisfaction: r ≈ -0.47
These correlations are large enough to confirm the PSS-10 is measuring its target construct, but small enough that it is not simply a relabeled depression or anxiety scale. The Milo 2025 review flagged that most reliability estimates assume tau-equivalence (rarely tested directly) and that the literature over-represents female and student samples.
Limitations and considerations
Appraisal, not events
The PSS-10 deliberately does not ask about specific stressors — no items about job loss, bereavement, illness, or relationship change. Cohen’s group built the scale on Lazarus’s transactional model, which holds that what matters is not the objective count of stressors but how a person appraises them. Two people facing the same external pressure can produce very different PSS-10 scores depending on coping resources, support, and outlook. The scale cannot tell you what is stressing you out — only how stressed you feel.
One-month recall window
The standard PSS-10 asks about the past month. Cohen’s lab notes that “longer retrospective periods likely reduce accuracy; shorter intervals should pose no problem”. A score completed right after an unusually intense week reflects that week’s appraisal, not a stable trait — re-administering weeks later, when conditions have changed, often yields a different total.
Demographic dependency
Mean PSS-10 scores vary substantially by demographic group. In the Klein 2016 German community sample:
- Higher in women (M = 13.07) than men (M = 11.93)
- Highest in 14-19 year olds (M = 14.05) and lowest in those 60+ (M = 11.94)
- Higher in unemployed adults (M = 15.39) than in employed (M = 12.32) or retired (M = 12.14) adults
- Higher in those without educational qualifications (M = 15.68) than in college-educated adults (M = 11.16)
Klein and colleagues concluded that “differentiated norm values” by sex and age are necessary for appropriate interpretation. These numbers are from a German sample and should not be transplanted onto US readers; representative US norms exist (Cohen & Janicki-Deverts 2012) but are not reproduced here.
Choosing between the PSS-14, PSS-10, and PSS-4
Three versions are validated by Cohen’s lab:
| Version | Items | Range | Trade-off |
|---|---|---|---|
| PSS-14 | 14 | 0-56 | 1983 original; four items perform poorly in factor analysis and were dropped in the PSS-10 |
| PSS-10 | 10 | 0-40 | Better psychometric properties than the PSS-14; most commonly used version |
| PSS-4 | 4 | 0-16 | Brief screener; faster but with weaker reliability and less psychometric data |
For most self-administration purposes the PSS-10 is the right balance.
What to do with your results: when to seek help
A high or rising PSS-10 score is a signal worth paying attention to, not an emergency in itself. The American Psychological Association defines stress as “the physiological or psychological response to internal or external stressors” — a normal reaction that becomes a problem when intense, persistent, or interfering with functioning. The 2025 Stress in America survey found societal-level stressors are widespread in US adults, so a moderately elevated score is not unusual.
Healthy first steps
NIMH-recommended coping strategies for stress and difficult events include:
- Maintaining regular routines for sleep, meals, and exercise
- Staying in contact with supportive people
- Practicing exercise or mindfulness
- Setting realistic, manageable goals
- Avoiding alcohol and drugs as coping tools
Pairing a PSS-10 score with a brief mood or anxiety screen — for example a DASS-21, which measures depression, anxiety, and stress together — can give a clearer picture of whether you are dealing with stress alone or stress alongside other symptoms.
Talk to a healthcare professional if any of the following apply
- Your PSS-10 score is in the conventional “high” range (27-40) and has stayed there across repeat completions weeks apart, recognizing that these bands are an informal convention rather than diagnostic thresholds
- You have persistent worry, anxiety, sadness, or fear that does not let up
- You are experiencing nightmares, insomnia, intrusive thoughts, or flashbacks
- You have physical symptoms of stress that persist — recurrent headaches, stomach pain, a racing heart, or sleep disturbance
- You are withdrawing socially, having difficulty thinking clearly, or struggling to function at work, school, or home
- You have a personal or family history of mental illness, prior trauma, or limited social support
- You are using alcohol or drugs to cope
A primary care provider is a reasonable starting point and can refer to a psychiatrist, psychologist, or clinical social worker.
If you are in crisis
- Call or text 988 (Suicide & Crisis Lifeline) or visit 988lifeline.org
- Call the SAMHSA Disaster Distress Helpline at 1-800-985-5990 for free, 24/7 multilingual crisis counseling
- To find local mental health services, use SAMHSA’s treatment locator at findtreatment.samhsa.gov
Frequently asked questions
How is the PSS-10 scored?
Rate each of the 10 items on a 0-4 scale (0 = never to 4 = very often) for the past month, reverse-score items 4, 5, 7, and 8 (so 0 becomes 4, 1 becomes 3, and so on), then sum all 10 scores. The total ranges from 0 to 40, with higher numbers meaning greater perceived stress.
Which items on the PSS-10 are reverse-scored?
Items 4, 5, 7, and 8 are reverse-scored because they are positively worded — they ask about feeling confident, in control, on top of things, and that things are going your way. High scores there mean less stress, so they are flipped before summing.
How do I cite the PSS-10?
The canonical citation is Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 385-396 (PMID 6668417). For US normative data, cite Cohen & Janicki-Deverts (2012) in the Journal of Applied Social Psychology.
What is the difference between the PSS-4 and the PSS-10?
The PSS-4 has 4 items (range 0-16); the PSS-10 has 10 items (range 0-40). The PSS-4 is faster but has weaker reliability and less psychometric data behind it — the PSS-10 is recommended for most purposes.
Where can I get a PSS-10 PDF?
Permission is handled through the ePROVIDE platform of Mapi Research Trust at eprovide.mapi-trust.org; submission is free and non-binding. Non-English translations are owned by their translators — permission requests for translated versions go to the translator directly, not to Dr. Cohen.
What score is considered “high” on the PSS-10?
There is no official cutoff — Cohen’s lab states the PSS “is not a diagnostic instrument; there are no score cut-offs”. The widely cited informal convention is 0-13 low, 14-26 moderate, 27-40 high, but these are not endorsed by the instrument’s author. Interpret your score relative to a comparison group and to your own previous scores.
Is the PSS-10 free to use?
Yes. The PSS-10 is distributed free through ePROVIDE (Mapi Research Trust) with non-binding submission, and has been translated into more than 40 languages including Spanish, French, German, Mandarin Chinese, Arabic, and Polish.
Does the PSS-10 measure stress in the past month or right now?
The standard PSS-10 asks about the past month. Cohen’s lab notes that “longer retrospective periods likely reduce accuracy; shorter intervals should pose no problem” — the recall window can be shortened if needed, but not lengthened.