Depression, Anxiety & Stress Test (DASS-21)

The Depression Anxiety Stress Scales (DASS) came out of the School of Psychology at the University of New South Wales in Sydney. The authors, Lovibond and Lovibond, published the manual in 1995 through the Psychology Foundation of Australia, and the developers ask people to cite it. The full DASS has 42 items, and the DASS-21 is a short form with seven items on each scale.

The authors treat depression, anxiety and stress as dimensions of severity and don't sort people into diagnostic categories. They picked the 21 items for strong factor loadings, coverage of each scale's content and item means that keep each score close to half the full-scale score. That's why each sum gets doubled before it's compared with the full DASS.

The 1995 paper by Lovibond and Lovibond tested the original DASS in a non-clinical sample of 717 people. Exploratory and confirmatory factor analyses supported the three-scale structure. The DASS Anxiety scale correlated 0.81 with the Beck Anxiety Inventory, and the Depression scale correlated 0.74 with the Beck Depression Inventory.

For the 21-item version, Henry and Crawford (2005) studied 1,794 people in a non-clinical sample broadly representative of UK adults. The best-fitting model had a general distress factor plus separate depression, anxiety and stress factors. The authors concluded that the subscales can validly measure depression, anxiety and stress. Each one also reflects general distress. The developers add that the three scales are moderately correlated, typically .5 to .7.

Each scale's seven items are added and the sum is doubled first, so scores run from 0 to 42 and can be compared with the full 42-item DASS. The developers publish the severity labels only in their paid manual and call every cut-off arbitrary, so we couldn't check these ranges against the manual itself. They match the ranges printed by two other sources that credit the manual: NovoPsych and an Australian clinic's scoring sheet.

The severity labels describe where a score falls in the general population. The developers say "mild" means above the population average but still well below the typical level of people seeking help, so it doesn't mean a mild disorder. They call the cut-offs arbitrary, and the scales don't map onto DSM or ICD diagnoses.

The DASS has no item about suicidal thoughts, because such items didn't load on any scale. It leaves out symptoms shared by these states, such as sleep, appetite and sexual problems, and it asks only about the past week. The development work used non-clinical samples, and the youngest people in them were 17. The developers say decisions based on score profiles should come from experienced clinicians who've done a clinical examination.

Read the three scores separately, since each one points to a different kind of difficulty. The DASS is a state measure, so your scores should move as your circumstances do, and retaking it after a few weeks shows whether they have. If a score stays high, or gets in the way of work, sleep or relationships, bring it to a doctor or therapist. The developers say people with high levels of distress should have their results interpreted by a qualified health professional.

If you're having thoughts of suicide, call or text 988 (the 988 Suicide & Crisis Lifeline), which is free and open 24/7, or call 911 if you're in immediate danger. For other views of the same symptoms, see the anxiety test, the depression test and the stress test.

More on this topic: Anxiety test (GAD-7), Depression test (PHQ-9), Stress test (PSS-10), CBT for anxiety: a beginner's guide.

Your answers are scored in your browser and aren't sent or saved anywhere. Close the page and they're gone.

• DASS home page, University of New South Wales: the developers' description of the 42-item DASS, the public domain statement and the paid manual.
• DASS overview, University of New South Wales: scale content, the past-week time frame, development in non-clinical samples, no suicide item, and the dimensional view of diagnosis.
• DASS frequently asked questions, University of New South Wales: how the 21 items were chosen, doubling the DASS-21 scores, the citation, what the severity labels mean, and the developers' terms for websites and apps.
• DASS21 response form and scoring template: official item wording and answer labels. The scoring template at the same site (Dass_template.pdf) shows which items belong to which scale.
• Lovibond and Lovibond, Behaviour Research and Therapy, 1995 (PubMed): abstract only: the 42-item DASS in 717 people, correlations with the Beck inventories. It doesn't cover the 21-item form or the severity labels.
• Henry and Crawford, British Journal of Clinical Psychology, 2005 (PubMed): abstract only: the DASS-21 in 1,794 UK adults, factor structure.
• NovoPsych, DASS-21 scoring: a commercial page, not the manual: its severity table credits Table 9 of the manual and doubles DASS-21 scores. Used only to cross-check the ranges.
• Signature Psychology, DASS 21 scoring instructions: a clinic's scoring sheet, not the manual: the same ranges, except that it prints 37+ for extremely severe stress, which looks like a typo for 34+.
• 988 Suicide and Crisis Lifeline: call or text 988, free, confidential, 24/7.