The PSQI is the most widely used self-report measure of sleep quality in research, and this PSQI scoring interpretation guide explains what the questionnaire asks, how each of its seven components is scored, and what the global score means.
What Is the Pittsburgh Sleep Quality Index?
The Pittsburgh Sleep Quality Index (PSQI) is a 19-item self-report questionnaire that measures sleep quality and sleep disturbances over the past month. It was developed by Buysse and colleagues at the University of Pittsburgh and published in 1989, originally to give clinicians and researchers a simple, standardised way to tell good sleepers from poor sleepers.
The questionnaire asks about the respondent’s usual sleep habits over the previous month: bedtime, how long it takes to fall asleep, wake time, hours actually slept, and how often specific problems disturb sleep. The answers are combined into seven component scores, which are then added into a single global score.
The PSQI also contains five extra questions for a bed partner or roommate — about loud snoring, pauses in breathing, leg twitching, and confusion during the night. These five items are not included in the scoring. One practical note: the PSQI is owned by the University of Pittsburgh and is free for non-commercial research and educational use; clinical and commercial use requires a licence.
The 7 Components of the PSQI at a Glance
Each of the seven components is scored from 0 (no difficulty) to 3 (severe difficulty). The table below shows what each component measures and which items feed into it.
| Component | What it measures | Items used |
|---|---|---|
| C1: Subjective sleep quality | The person’s overall rating of their own sleep | Item 9 |
| C2: Sleep latency | How long it takes to fall asleep, and how often falling asleep within 30 minutes is a problem | Items 2 and 5a |
| C3: Sleep duration | How many hours of actual sleep the person gets per night | Item 4 |
| C4: Habitual sleep efficiency | The percentage of time in bed actually spent sleeping | Items 1, 3, and 4 |
| C5: Sleep disturbances | How often nine different problems disturb sleep | Items 5b to 5j |
| C6: Use of sleeping medication | How often the person takes medicine to help them sleep | Item 6 |
| C7: Daytime dysfunction | Trouble staying awake during the day and keeping up enthusiasm | Items 7 and 8 |
How to Score Each Component
Most items are answered on a four-point scale from 0 to 3. The open-ended items (bedtime, minutes to fall asleep, wake time, hours slept) are converted to a 0–3 score using the rules below.
Component 1: Subjective Sleep Quality
Item 9 asks: “During the past month, how would you rate your sleep quality overall?” The answer is scored directly: very good = 0, fairly good = 1, fairly bad = 2, very bad = 3.
Component 2: Sleep Latency
Sleep latency combines two items. Item 2 asks how many minutes it usually takes to fall asleep, recoded as: 15 minutes or less = 0, 16 to 30 minutes = 1, 31 to 60 minutes = 2, more than 60 minutes = 3. Item 5a asks how often the person could not fall asleep within 30 minutes, scored 0 (not during the past month) to 3 (three or more times a week). Add the two sub-scores and recode the sum: 0 becomes 0, 1–2 becomes 1, 3–4 becomes 2, and 5–6 becomes 3.
Component 3: Sleep Duration
Item 4 asks how many hours of actual sleep the person gets per night: more than 7 hours = 0, 6 to 7 hours = 1, 5 to 6 hours = 2, less than 5 hours = 3. Note that this is hours actually asleep, not hours spent in bed — the difference is the basis of the next component.
Component 4: Habitual Sleep Efficiency
Sleep efficiency is the one component that needs a small calculation: subtract usual bedtime (item 1) from usual getting-up time (item 3) to get hours in bed, then divide hours of actual sleep (item 4) by hours in bed and multiply by 100.
Sleep efficiency (%) = (hours of actual sleep ÷ hours in bed) × 100
The percentage is scored: above 85% = 0, 75 to 84% = 1, 65 to 74% = 2, below 65% = 3.
Component 5: Sleep Disturbances
Items 5b to 5j ask how often nine problems disturbed sleep during the past month: waking up in the night, getting up to use the bathroom, breathing uncomfortably, coughing or snoring loudly, feeling too cold or too hot, having bad dreams, having pain, and any other reason. Each sub-item is scored 0 to 3 by frequency. Add all nine together (maximum 27) and recode: 0 stays 0, 1–9 becomes 1, 10–18 becomes 2, and 19–27 becomes 3.
Component 6: Use of Sleeping Medication
Item 6 asks how often the person has taken medicine to help them sleep — prescribed or over the counter. It is scored directly: not during the past month = 0, less than once a week = 1, once or twice a week = 2, three or more times a week = 3.
Component 7: Daytime Dysfunction
This component combines two items. Item 7 asks how often the person has had trouble staying awake while driving, eating, or engaging in social activity. Item 8 asks how much of a problem it has been to keep up enough enthusiasm to get things done. Both are scored 0 to 3; add them and recode the sum the same way as sleep latency: 0 becomes 0, 1–2 becomes 1, 3–4 becomes 2, and 5–6 becomes 3.
The Global PSQI Score
The global PSQI score is simply the sum of the seven component scores:
Global PSQI = C1 + C2 + C3 + C4 + C5 + C6 + C7
Because each component ranges from 0 to 3, the global score ranges from 0 to 21. Higher scores mean worse sleep quality — the opposite of many wellbeing scales, so say so explicitly when you report it.
PSQI Scoring and Interpretation: What the Numbers Mean
The key number to remember is 5. In the original validation study, a global score greater than 5 identified poor sleepers with 89.6% sensitivity and 86.5% specificity — it correctly picked out about nine in ten people with a sleep problem, while wrongly flagging about one in seven healthy sleepers.
- Global score 0 to 5: good sleep quality. The person is classified as a good sleeper.
- Global score above 5: poor sleep quality. The person is classified as a poor sleeper.
A score above 5 does not diagnose a sleep disorder. The PSQI is a screening and research instrument, not a clinical diagnosis, and a high score should be described as “poor sleep quality” rather than as insomnia or any specific disorder. The component scores are just as useful as the global score: two participants can both score 9 overall for completely different reasons — one because of short sleep duration, the other because of nightly disturbances. Report the component breakdown alongside the global score whenever space allows.
A Worked Example
Imagine a participant, Sana, who answers the PSQI like this for the past month: she goes to bed at 11:00 p.m., takes 45 minutes to fall asleep, and gets up at 7:00 a.m., reporting 5.5 hours of actual sleep. She rates her sleep quality as “fairly bad”, could not fall asleep within 30 minutes twice a week, and her nine disturbance items add up to 12. She takes no sleep medication, has trouble staying awake in lectures once a week, and finds it fairly hard to keep up enthusiasm.
Here is her scoring, component by component:
- C1 (subjective quality): “fairly bad” = 2
- C2 (latency): 45 minutes = 2, plus item 5a twice a week = 2; sum 4 recodes to 2
- C3 (duration): 5.5 hours = 2
- C4 (efficiency): 8 hours in bed, 5.5 hours asleep, so efficiency is (5.5 ÷ 8) × 100 = 69%, which scores 2
- C5 (disturbances): subtotal 12 recodes to 2
- C6 (medication): none = 0
- C7 (daytime dysfunction): item 7 = 1, item 8 = 2; sum 3 recodes to 2
Global PSQI = 2 + 2 + 2 + 2 + 2 + 0 + 2 = 12. A score of 12 is well above the cut-off of 5, so Sana would be classified as a poor sleeper. The component pattern tells a story too: her problem is spread across the board — late sleep onset, short sleep, low efficiency, and frequent disturbances — rather than a single issue.
Reliability and Validity Evidence
Before you use any scale in a thesis, your examiner will want to know that it actually works. In the original validation sample of 148 people — 52 healthy controls, 34 patients with major depressive disorder, and 62 patients with a sleep disorder — the PSQI showed good internal consistency, with a Cronbach’s alpha of 0.83 for the global score, and stability over time, with test–retest reliability of r = 0.85 (Buysse et al., 1989).
Since 1989, the PSQI has been translated into many languages and validated in many populations, from university students to older adults and clinical groups. Remember, though, that it is a self-report measure of the past month: it captures perceived sleep quality, not sleep measured by a device. Say so in your limitations section.
How to Use the PSQI in Your Research
Scoring it. Score each component by hand using the rules above, or use a spreadsheet or statistical software. In SPSS, check every recode against this guide — the latency, disturbance, and daytime-dysfunction recodes are the ones most often done wrong.
Reporting it. In your methods section, name the instrument, its authors, and the recall period. In your results, report the mean and standard deviation of the global score and the proportion of participants scoring above 5. A sentence like “The mean global PSQI score was 7.4 (SD = 3.1), and 58% of participants scored above 5, indicating poor sleep quality” tells your reader far more than the mean alone.
Common mistakes to avoid. Four errors show up again and again: adding the five bed-partner items into the total (they are unscored); reporting raw sums instead of recoded 0–3 scores; using “hours in bed” instead of “hours actually asleep” for sleep efficiency; and calling a score above 5 a diagnosis of insomnia.
How to Cite the PSQI
Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193–213.
If you used a translated or adapted version, cite that validation study as well and state the language version in your methods section.
Frequently Asked Questions
How many items are in the PSQI?
The PSQI has 19 self-rated items that are scored, plus five additional questions answered by a bed partner or roommate. Only the 19 self-rated items contribute to the seven component scores and the global score.
What is a good PSQI score?
A global PSQI score of 5 or below is considered good sleep quality. A score above 5 indicates poor sleep quality. The global score ranges from 0 to 21, and higher scores always mean worse sleep.
Can I use the PSQI for free in my thesis?
Yes, for non-commercial research and educational use. The instrument is owned by the University of Pittsburgh, and commercial or clinical use requires a licence. Download it from an official source so you are using the original wording.
Is the PSQI a diagnostic tool for insomnia?
No. The PSQI screens for poor sleep quality and measures it for research purposes. A score above 5 flags a person as a poor sleeper, but it does not diagnose insomnia or any other sleep disorder. Diagnosis requires a clinical assessment, usually including a sleep history and sometimes an overnight sleep study.
Key Takeaways
- The PSQI is a 19-item self-report measure of sleep quality over the past month, developed by Buysse and colleagues in 1989.
- Its 19 items produce seven component scores, each from 0 to 3, which add up to a global score from 0 to 21.
- A global score above 5 means poor sleep quality, with 89.6% sensitivity and 86.5% specificity in the original validation.
- Watch the recodes: latency, sleep disturbances, and daytime dysfunction all squeeze a raw sum back down to a 0–3 component score. The five bed-partner items are not scored, and a high score is a screening result, not a diagnosis.