Sleep Quality (PSQI)

What the PSQI measures

Sleep architecture is the foundational bedrock of all psychological health. Severe mental illnesses (e.g., Bipolar, Schizophrenia, MDD) are almost universally preceded or accompanied by severe sleep disruption. The PSQI is the clinical gold standard for measuring sleep because it does not just ask "how tired are you?" (which measures fatigue). Instead, it calculates 7 distinct, quantitative components of sleep mechanics to pinpoint exactly why the user's sleep is failing.

The dimensions it assesses

• Subjective Sleep Quality: How the user rates their overall rest.
• Sleep Latency: How long it takes to fall asleep.
• Sleep Duration: Total hours of actual sleep.
• Habitual Sleep Efficiency: The ratio of time spent asleep vs. time spent in bed. (The metric that matters most for insomnia).
• Sleep Disturbances: Waking up due to bathroom needs, breathing issues, cold/heat, or nightmares.
• Use of Sleeping Medication: Reliance on chemical induction (OTC or prescribed).
• Daytime Dysfunction: Trouble staying awake while driving/eating, and lack of enthusiasm.

What to do with your result

1. Sleep Efficiency Protocol (C4): If Sleep Efficiency is low (< 85%), the user has classic Insomnia. The gold-standard treatment is Cognitive Behavioral Therapy for Insomnia (CBT-I) and Sleep Restriction Therapy. The user must stop reading, eating, or worrying in bed. The bed is for sleep and sex only. If awake for 20 minutes, they must leave the bed.
2. Apnea Screening (C5d, C5e): If the user scores high on "cannot breathe comfortably" or "snore loudly", they must immediately be referred for an overnight Polysomnography (Sleep Study) to rule out Obstructive Sleep Apnea (OSA), which destroys cardiovascular health.
3. Latency Protocols (C2): If latency is high, target pre-sleep arousal. Mandate a 60-minute digital sunset (no screens), drop the room temperature to 65°F (18°C), and eliminate caffeine after 2 PM.

Common questions about the PSQI

What does the PSQI measure?

Measurement of subjective sleep quality, sleep latency, and specific sleep disturbances over a 1-month time interval.

Can the PSQI diagnose me?

No. Sleep Quality (PSQI) is a screening questionnaire, not a diagnostic tool. It can flag that something may need a closer look. It cannot tell you what is causing it. Grief, burnout, a thyroid problem, medication side effects, chronic pain and months of bad sleep are not among its questions, and any of them can move a score. Telling those apart is what a professional assessment is for.

Is this PSQI test really free?

Yes. No account, no email address, no payment. Psychometrics Today is funded by donations rather than advertising, which is also what allows us to publish instruments whose authors permit free but non-commercial use.

Are my answers stored or shared?

No. Your answers are scored inside your browser and never sent to us. There is no database of responses, because we never receive any. Closing the tab erases them, which is why results cannot be recovered afterwards.

How often should I retake the PSQI?

Most of these questionnaires ask about the past one to two weeks, so retaking sooner than that mostly measures the same period twice. Every two to four weeks is a reasonable rhythm if you are tracking change. A direction of travel over several weeks tells a doctor far more than any single score.

How accurate is the PSQI?

Screening questionnaires are good at flagging that something merits attention and poor at establishing what it is. This one is sensitive to how you feel on the day you take it, and many instruments were validated on narrow populations whose cut-offs may not transfer cleanly to everyone. The references at the foot of this page are the primary sources.

Why can I not take the PSQI here?

Free for academic and non-commercial research, but commercial use requires a paid licence. We will not reproduce a questionnaire we have no right to reproduce, so this page explains what it measures instead. Use original sleep-quality items, or the free Insomnia Severity Index for research contexts.