PHQ-9 vs GAD-7 vs DASS-21 vs WEMWBS: Which Validated Instrument Should Your Psychology PhD Use? (2026)

Why the instrument you pick changes what your thesis can claim

A doctoral thesis using a self-report psychological instrument is making an implicit claim every time it reports a score: that the instrument measures what the thesis says it measures, in the population the thesis studies, in a way comparable to the wider literature the discussion chapter will cite. Choosing PHQ-9 over DASS-21, or WEMWBS over either, is not a minor operational decision — it changes which body of comparator studies your results chapter can credibly sit alongside, and examiners in psychology and health-related fields notice when an instrument is a poor conceptual fit for the stated research question. This is the same underlying discipline as choosing the right theoretical framework before building an argument on it — see how an education doctorate selects a theoretical framework that does explanatory work rather than one chosen for familiarity, since an instrument that does not match your construct fails an examiner’s scrutiny for the same reason a mismatched theory does.

PHQ-9: the depression-screening standard, built for repeat administration

The nine-item Patient Health Questionnaire, developed by Spitzer, Kroenke and Williams and introduced in 2001, maps directly onto DSM depressive symptom criteria and is short enough for repeated administration across a longitudinal design without excessive respondent burden. This makes it the natural choice for a thesis tracking depressive symptoms over time — a treatment trial, an intervention study, or a longitudinal cohort — where the same nine items need to be completed multiple times without fatiguing participants. Its widespread use across primary care and mental health research also means a PhD using it can draw on a very large comparator literature when interpreting scores, which strengthens a discussion chapter considerably. A caution worth stating explicitly in a methodology chapter: PHQ-9 was designed and validated as a screening and severity-tracking tool, not a diagnostic instrument on its own, and a thesis that treats a PHQ-9 cut-off score as equivalent to a clinical diagnosis of depression is overstating what the instrument was built to establish — state the score as symptom severity, and reserve diagnostic language for where a clinical interview or structured diagnostic assessment was also used.

GAD-7: the anxiety counterpart, almost always paired with PHQ-9

The seven-item Generalized Anxiety Disorder scale, developed by Spitzer, Kroenke, Williams and Löwe and published in 2006, is the anxiety-specific counterpart most often administered alongside PHQ-9 in mental-health research, because the two conditions frequently co-occur and a thesis measuring one usually needs to account for the other as a covariate or comorbidity. If your research question is specifically about anxiety rather than depression, GAD-7 alone may suffice; if your population is likely to show comorbid depression, administering both from the outset avoids having to justify a post-hoc addition to your ethics approval later in the candidature. Adding an instrument mid-study is rarely a trivial paperwork exercise — it typically requires a formal ethics amendment, a revised participant information sheet, and, for a longitudinal design, a defensible decision about whether earlier waves of data collection can be retrospectively compared to later ones that include the new measure.

DASS-21: the tripartite instrument for distinguishing three related states

The Depression Anxiety Stress Scales, in the shorter 21-item form used in most doctoral research, splits its items into three seven-item subscales — depression, anxiety and stress — built on the tripartite model that treats these as related but empirically separable negative emotional states, rather than the two-condition framing PHQ-9 and GAD-7 assume. It is in the public domain and freely available to download, developed by the Psychology Foundation of Australia. DASS-21’s specific advantage for a thesis is the stress subscale, which PHQ-9 and GAD-7 do not measure at all — a project examining stress as a distinct construct, rather than folding it into anxiety, needs an instrument built to separate the two, and DASS-21 is the standard choice in the UK literature for that purpose. Its public-domain status also matters practically for a self-funded or tightly budgeted candidature: unlike a licensed instrument, there is no per-participant fee, no institutional licence to negotiate, and no delay while a licensing agreement is signed before data collection can begin — a genuine operational advantage on top of its measurement properties.

WEMWBS: the positively framed wellbeing scale, licensed rather than free

The Warwick-Edinburgh Mental Wellbeing Scale, developed jointly by the Universities of Warwick and Edinburgh, measures positive mental wellbeing directly rather than the absence of symptoms — a genuinely different measurement target from the three symptom-focused instruments above, and the right choice where a thesis’s research question is about flourishing or wellbeing promotion rather than disorder or distress. A 14-item full version and a 7-item short form are both available. Unlike DASS-21, WEMWBS is licensed rather than public domain: non-commercial licences are available to public-sector organisations, registered charities and social enterprises, with separate NHS and commercial tiers, so a doctoral researcher based at a university should confirm which licence category applies to their specific project and institution before assuming free use, rather than treating it as equivalent to a public-domain instrument.

Translation, adaptation and administering electronically

None of the four instruments above should be translated informally for a non-English-speaking population — each has its own process for validated translation, and using an unofficial translation undermines the very comparability with the wider literature that is the main reason for choosing a named, validated instrument in the first place. Check the instrument’s own website or publisher for an existing validated translation in your target language before commissioning or attempting your own; where none exists, a formal translation and back-translation process, reported in your methodology chapter, is expected. Administering any of these instruments electronically, through a university-approved survey platform rather than paper, raises the same UK GDPR and data storage questions any other personal data collection does — confirm your survey platform is covered by your ethics approval and your university’s approved-tools list before switching from a paper administration your ethics committee already approved.

A worked illustration of instrument choice in practice

An invented but representative example: a doctoral project in occupational psychology investigating whether a workplace wellbeing intervention reduces burnout and improves flourishing among NHS staff would plausibly need two instruments, not one — DASS-21 to capture whether stress and anxiety symptoms actually reduce post-intervention, and WEMWBS to capture whether the intervention also increases positive wellbeing, since a reduction in distress and an increase in flourishing are conceptually and empirically distinguishable outcomes that a single symptom-focused instrument cannot both detect. Confirming the WEMWBS licence category for an NHS-based project specifically — rather than assuming a university’s non-commercial status extends automatically to an NHS partner site — would be one of the practical steps this candidate’s ethics application needed to resolve before data collection could begin.

Which instrument should your thesis actually use?

For a thesis measuring depressive symptoms specifically, especially with repeat administration across a longitudinal or intervention design, PHQ-9 is the standard choice with the largest comparator literature. Where anxiety is equally central to the research question, pair it with GAD-7 from the outset rather than adding it later. Where your research question requires distinguishing stress from anxiety and depression as separable constructs — common in occupational, educational or nursing doctoral research — DASS-21’s tripartite structure and public-domain status make it the more defensible choice. Where the thesis is about positive wellbeing or flourishing rather than distress or disorder, none of the symptom-focused instruments above are a good conceptual fit, and WEMWBS is the field’s standard alternative, provided the licensing terms are confirmed for your specific institution and use case first. Whichever instrument you land on, the same sample-size logic that governs any quantitative doctoral design still applies before you can interpret a change in score as meaningful — see how sample size is reasoned for postgraduate health research for the underlying power calculation, since an instrument choice does not substitute for a properly powered design.

The recommendation above assumes a UK doctoral candidature choosing among general-purpose, widely used instruments; a professional or clinical doctorate may have its own field-mandated instrument set entirely — see how a DClinPsy thesis structures its empirical component for how a clinical training programme’s own requirements can override the general considerations discussed here.

FAQ

Can I use PHQ-9 and GAD-7 together in the same study?

Yes, and this is common practice given how often depression and anxiety co-occur — many studies administer both instruments together to capture comorbidity.

Is DASS-21 free to use in a PhD thesis?

Yes. DASS-21 is in the public domain and can be downloaded and used without a licensing fee, developed by the Psychology Foundation of Australia.

Is WEMWBS free for a university-based PhD project?

Not automatically. WEMWBS operates a licensed model with NHS, commercial and non-commercial tiers; check which tier applies to your specific institution and project before assuming free use.

Why would I choose DASS-21 over PHQ-9 and GAD-7?

DASS-21 measures stress as a construct distinct from depression and anxiety, which PHQ-9 and GAD-7 do not measure at all — choose it when your research question specifically requires separating stress from the other two states.

What is the main difference between WEMWBS and the other three instruments?

WEMWBS measures positive mental wellbeing directly, rather than the severity of depressive, anxiety or stress symptoms — it is built to detect flourishing, not distress.

Are there short forms of these instruments for a time-limited study?

Yes for WEMWBS, which has a validated 7-item short form alongside the full 14-item version. PHQ-9, GAD-7 and DASS-21 are already relatively short and are typically used in full.

Do I need ethics committee approval to use any of these instruments?

Yes. Using a validated psychological instrument does not bypass your university’s ethics approval process — the instrument choice is part of what your ethics application must justify.

Can I translate one of these instruments myself for a non-English population?

Not informally. Check for an existing validated translation first; where none exists, a formal translation and back-translation process, reported in your methodology chapter, is the expected standard.

Is a PHQ-9 score alone enough to claim a participant has clinical depression?

No. PHQ-9 is a screening and severity-tracking tool, not a standalone diagnostic instrument — report it as symptom severity, and reserve diagnostic claims for cases where a clinical interview or structured diagnostic assessment was also used.

Build the instrument choice into your methodology chapter early

The instrument comparison above is also, in miniature, the argument your methodology chapter needs to make explicitly — why this instrument, for this construct, in this population. Tesify’s thesis writing workspace helps you draft that justification against your own research questions rather than starting from a blank page once data collection is already under way.