Most Canadian nursing theses should use the Copenhagen Burnout Inventory (CBI): it is free, validated, covers three domains, and may be reproduced in your appendix. Choose the Maslach Burnout Inventory (MBI) only when your research question requires direct comparability with the dominant literature and your budget covers a per-respondent licence.
Why does this one decision shape half your methodology chapter?
Your instrument choice is not a citation you drop into a sentence. It determines what you are allowed to print in your appendix, what your Research Ethics Board sees before it approves you, what your subscale scores mean, how you report reliability, and which studies you can legitimately compare your findings against in the discussion. Students routinely pick a scale because a supervisor mentioned it in passing, discover eight weeks later that they cannot afford or cannot reprint it, and lose a term rebuilding the survey.
The decision is also unusually consequential in nursing specifically. Burnout research in nursing is dominated by one instrument, which creates real pressure to conform — but that instrument is commercially licensed, which creates real pressure not to. Understanding the trade-off precisely is what lets you defend the choice at your defence instead of apologising for it.
What are the four instruments a Canadian nursing thesis actually chooses between?
| Instrument | Structure | Cost | Reproduce in appendix? | Best fit |
|---|---|---|---|---|
| Maslach Burnout Inventory (MBI-HSS / MBI-HSS-MP) | 22 items; emotional exhaustion, depersonalisation, personal accomplishment | Licensed per administration through Mind Garden, with minimum purchase quantities | No — licence forbids reproduction | Studies that must sit directly alongside the mainstream burnout literature |
| Copenhagen Burnout Inventory (CBI) | 19 items; personal, work-related and client-related burnout | Free for research use | Yes, with attribution | Most master’s theses; separates burnout attributable to patients from burnout attributable to the job |
| Oldenburg Burnout Inventory (OLBI) | 16 items; exhaustion and disengagement, positively and negatively worded | Free for research use | Yes, with attribution | Job-demands-resources framing; studies worried about acquiescence bias |
| Professional Quality of Life Scale (ProQOL) | 30 items; compassion satisfaction, burnout, secondary traumatic stress | Free, with registration and use conditions | Yes, under the stated terms | Compassion fatigue, moral distress and trauma exposure in oncology, ICU, emergency and palliative settings |
Four instruments, four different constructs. They are not interchangeable, and a thesis that treats them as interchangeable will be caught in the discussion chapter, where the numbers stop lining up with the literature you cited.
What does the MBI actually cost a student?
The MBI is distributed under commercial licence by Mind Garden. You purchase the right to administer it a fixed number of times, priced per administration, with a minimum order. There is no free student tier that lets you skip this, and there is no version of “I found the items in a published paper” that makes reproduction lawful. Confirm the current per-use rate and minimum quantity directly with the publisher before you build the budget line, because rates change and older forum posts are unreliable.
Two consequences follow that catch students out. First, the licence is normally tied to a projected sample size — if your response rate collapses and you need to reopen recruitment to a second site, you may need to buy more administrations. Second, the licence prohibits printing the instrument in your thesis. Since Canadian theses are deposited in an open institutional repository and typically in Library and Archives Canada’s national collection, “just put it in the appendix” means publishing a licensed instrument to the open web. Your library’s thesis office will not let that through, and if it slips past, you have a copyright problem attached to your name permanently.
What you may do is describe the instrument, cite it, reproduce one or two sample items if the licence permits, and state in your methods that the full instrument is available from the publisher. Confirm the exact wording against your own faculty’s rules for third-party material in a thesis; the reasoning is the same as for permission to reproduce published material in a thesis.
Is the Copenhagen Burnout Inventory really free?
Yes, for research and non-commercial use, and this is why it has become the pragmatic default for graduate nursing research. The CBI came out of the Danish PUMA study and was designed from the start to be publicly available. You cite the source, you may reproduce the items, and there is no per-respondent cost.
The structural argument for the CBI in nursing is stronger than the price argument, and it is the one to make in your methodology chapter. The CBI splits burnout into three sources: personal burnout (exhaustion regardless of cause), work-related burnout (exhaustion the respondent attributes to their job) and client-related burnout (exhaustion the respondent attributes to working with patients). For nursing that separation is analytically valuable. It lets you distinguish a nurse who is exhausted by staffing ratios and mandatory overtime from a nurse who is exhausted by patient contact itself — two findings with completely different implications for retention policy, and two findings the MBI’s structure blurs together.
The trade-off is comparability. There is simply more MBI literature, so a CBI study has a slightly smaller pool of directly comparable results. In practice this is manageable: report your subscale means, be explicit that scores are not convertible between instruments, and compare patterns rather than raw values.
When is the OLBI the better fit?
Choose the Oldenburg Burnout Inventory when your conceptual framework is the job demands–resources model, because the OLBI was built inside that framework and its two factors — exhaustion and disengagement — map onto it cleanly. It is also the right answer when you are worried about response bias: the OLBI deliberately mixes positively and negatively worded items, which discourages straight-lining down a column of agree responses. If your survey is long and going out to nurses at the end of a twelve-hour shift, that matters more than it sounds.
The cost of that design is scoring complexity. Reverse-coded items are the single most common place where a thesis dataset silently breaks, and a reversed item left uncoded will quietly destroy your reliability estimate and every analysis downstream. Recode before you do anything else, and verify by checking that reversed items correlate positively with their subscale.
When should you use the ProQOL instead of a burnout scale?
Use the ProQOL when your research question is really about compassion fatigue, moral distress or trauma exposure rather than burnout as such. It measures three things at once: compassion satisfaction (the positive return from caring work), burnout, and secondary traumatic stress. In oncology, intensive care, emergency and palliative nursing, the third factor is often exactly the phenomenon your participants describe, and a pure burnout instrument will miss it.
Do not administer the ProQOL alongside a full burnout inventory to be safe. Thirty items plus twenty-two items plus demographics plus your other variables produces a survey nurses abandon halfway through, and partial responses are worse than no responses. Choose the construct your research question names, and defend the choice in one paragraph.
What does your REB need to see about the instrument?
Every Canadian REB operating under TCPS 2 will expect the actual data collection instrument attached to your application, not a description of it. That means the full item set, the response scale, the demographic questions and the exact order the participant sees. If you are using a licensed instrument you cannot circulate, say so in the application and provide what the licence permits — REBs deal with this constantly and have a standard accommodation, but they need to be told rather than left to discover a gap.
Three instrument-specific points reliably attract reviewer comments in nursing applications. First, burnout items are mildly sensitive: some respondents will recognise their own distress while completing your survey, so your application should include a short debriefing statement pointing to the employee assistance programme and the provincial nursing association’s support line. Second, anonymity is fragile in small-unit sampling — if you collect unit, years of experience and shift pattern, a 14-bed unit becomes identifiable, so plan to collapse categories before anyone sees the data. Third, if you recruit through a workplace, the REB will scrutinise voluntariness closely, because a manager forwarding your survey link creates perceived pressure to participate. The full picture of what triggers review and what qualifies for delegated review is set out in the guide to REB approval under TCPS 2 in Canada.
Do you need a second approval to recruit nurses at a hospital?
Almost always, yes, and this is the timeline killer in Canadian nursing theses. Your university REB approves the research; the health authority or hospital where the nurses work runs its own review, and separately requires operational approval — sign-off from the unit or programme leadership confirming the study can run in their setting without disrupting care. Some provinces have harmonised processes that reduce duplication between institutions, but harmonised does not mean instantaneous, and operational approval is a distinct step that no harmonisation removes.
Budget for this properly. A realistic sequence is: supervisor approves the protocol, university REB review with one round of revisions, health authority review, operational sign-off, then recruitment. Students who assume one approval and plan a six-week data collection window routinely lose an entire term. Start the institutional paperwork while you are still finalising your literature review, not after.
How many nurses do you need to respond?
The honest answer is that it depends on the analysis you have committed to, not on a number someone quotes you in a hallway. A correlational design detecting a moderate association needs far fewer respondents than a multiple regression with six predictors and an interaction term, and a comparison across four unit types needs enough respondents in the smallest cell to be worth reporting at all. Work the calculation backwards from your planned analysis, and expect nursing survey response rates to be low enough that you should oversample your invitation list substantially. The mechanics of doing that calculation properly are in the guide to how many participants a thesis actually needs.
What do you report about the instrument in your results chapter?
Four things, in this order: the number of usable responses and how you handled missing items; the reliability coefficient for each subscale in your sample, not the coefficient from the original validation paper; the descriptive statistics for each subscale; and any deviation from standard scoring, with your reason.
The reliability point is where markers push back hardest. Reporting “the CBI has demonstrated good reliability (α = .87)” and citing the developers is not reporting your reliability — it is reporting theirs. Calculate alpha on your own data for each subscale separately and report it. If a subscale comes back weak, do not quietly drop it; report it, discuss what it means for the interpretation of that subscale, and consider whether an item performed badly in your context. The thresholds and the defensible responses to a low value are covered in the explainer on what counts as an acceptable Cronbach’s alpha.
How do you write the instrument paragraph?
Here is a worked example you can adapt. Notice that it names the instrument and version, gives the structure, states the response format and scoring direction, cites the validation evidence, reports sample-specific reliability, and justifies the choice against the alternative — in one paragraph.
Burnout was measured using the Copenhagen Burnout Inventory (Kristensen et al., 2005), a 19-item instrument comprising three subscales: personal burnout (6 items), work-related burnout (7 items) and client-related burnout (6 items). Items are rated on a five-point frequency or intensity scale and transformed to a 0–100 metric, with higher scores indicating greater burnout. The CBI was selected over the Maslach Burnout Inventory because its client-related subscale permits burnout attributable to patient contact to be distinguished from burnout attributable to the wider work environment — a distinction central to the present research question — and because the instrument may be reproduced in full (see Appendix B), supporting replication. In the present sample, internal consistency was acceptable for all three subscales (personal α = .87, work-related α = .85, client-related α = .82).
That paragraph belongs in your instruments section, which sits inside the wider structure described in the walkthrough of how to write the methodology chapter of a Canadian thesis.
Can Tesify help you write the instruments section?
Yes — drafting the instrument justification, the reliability reporting and the limitations paragraph that follows from your scale choice is exactly the kind of structured academic writing Tesify is built for. You supply the decision and the numbers; Tesify helps you turn them into defensible chapter prose with consistent terminology and correctly formatted citations, so you spend your supervisor meetings on the argument rather than on the wording.
Start your thesis chapter with Tesify
Frequently asked questions
Can I use the MBI for free if I found the items in a published article?
No. The MBI is licensed commercially, and items appearing in a published paper does not transfer any right to administer or reproduce them. Administering it without a licence is a copyright violation and a research integrity problem that a committee can raise at your defence.
Can I put the burnout scale in my thesis appendix?
You can reproduce the CBI and OLBI with proper attribution, and the ProQOL under its stated terms. You cannot reproduce the MBI. Because Canadian theses are deposited openly, reproducing a licensed instrument means publishing it to the open web, which your thesis office will refuse.
Can I change the wording of a validated scale to suit Canadian nursing?
Only minimally, and you must report every change. Substituting “clients” for “patients” is a routine, reportable adaptation. Rewriting items, dropping items or changing the response scale breaks the validation evidence, meaning you can no longer claim the instrument’s psychometric properties, and your reliability and comparability arguments both weaken.
Do I need a French version if I recruit in Quebec or in a bilingual hospital?
If your participant pool includes francophone nurses, yes, and you must use an existing validated French translation rather than translating it yourself. A do-it-yourself translation is a new instrument with no validation evidence. Both the CBI and ProQOL have published French versions; verify the specific version and cite it.
Should I combine two burnout instruments to be thorough?
No. It lengthens the survey, increases dropout, produces overlapping constructs you cannot cleanly interpret, and invites the committee question “why did you measure the same thing twice?” Choose the instrument that matches your research question and defend it in one paragraph.
Is there a cut-off score for “burned out”?
Treat cut-offs with caution. Burnout instruments were designed as continuous measures, and the widely circulated thresholds are conventions rather than diagnostic criteria — the ICD-11 classifies burnout as an occupational phenomenon, not a medical condition. Analyse continuous scores as your primary approach; if you also report a categorical breakdown, state the threshold, cite its source and label it descriptive.
How do I handle participants who skip items?
Decide the rule before you collect data and write it into your methods. A common convention is to compute a subscale score when a respondent has answered at least half its items, using their mean for the remainder, and to exclude the subscale otherwise. Report how many cases this affected.
Does my REB need the instrument before it will review my application?
Yes. The complete instrument as the participant will see it is a standard attachment, and applications submitted without it are returned unreviewed, costing you a full review cycle. If the instrument is licensed and cannot be circulated, state that explicitly and supply what the licence permits.
Can I recruit nurses through a Facebook group instead of a hospital?
Often yes, and it avoids the health authority approval chain — but your REB must approve the recruitment method and text, group administrators must consent, and you should expect questions about who your sample actually represents. Online convenience samples of nurses skew by age, specialty and province, and that belongs in your limitations.
How long does the whole approval process take for a nursing thesis?
Plan on months rather than weeks when a health authority is involved: university REB review with revisions, then institutional review, then operational approval at the unit level. Requirements differ by province, institution and faculty, so confirm the sequence with your own REB office and your supervisor before you commit to a data collection window.