Occupational burnout in a psychology master's thesis - how do you write the theory chapter?

You have a topic about occupational burnout, a handful of papers and a blank document. One paper describes three dimensions, another describes two, and your notes use “tiredness” for both. Start the chapter by deciding what burnout will mean in your thesis. Then compare the relevant approaches, explain which one you have chosen and show how that choice leads to your research question and what you plan to measure. This structure works for a literature-based thesis as well. By the end of the chapter, readers should know which concepts you will use to compare studies.

Start in a working document. Under the title, write one sentence naming whose experiences you will examine, what aspect of their work you will consider alongside those experiences and what you will call burnout. Then list the dimensions from your two most important sources. If one list has exhaustion and disengagement from work and the other also includes professional efficacy, flag that difference in your outline. In the worked example below, a student on a Psychology degree programme turns this short note into a paragraph, an instrument choice and a research question she can answer.

What belongs in the outline before you write the first section?

Decide whether you want to explain the structure of burnout, compare findings within an occupational group or explain why a particular working condition might be associated with burnout. A title such as “Occupational burnout among service workers” leaves all three options open. The outline for a question about time pressure, exhaustion and disengagement will differ from one about how two approaches define professional efficacy. The first thesis needs literature on time pressure and both subscales. The second needs more space for definitions of efficacy, how it is measured and the basis for comparison.

How do you narrow the topic to one research decision?

Write down the population, the work variable and the dimensions of burnout separately. An initial set might be customer service workers, self-reported time pressure, and exhaustion and disengagement from work. Decide whether participants will work by phone, in person, by chat or across several channels. The distinction is useful if it affects the kind of work or how you measure time pressure. Calls per hour, the share of cases resolved on time and a rating of how rushed someone feels provide different information. If you only have access to a survey, the last measure can tell you about perceived pressure. Use that term in the aim and research question as well.

Think about what you will need to justify the scope. If you plan to study several occupations, a label such as “working adults” calls for an explanation of why a shared question about working conditions has a similar meaning across those occupations. If you study a single organisation, its scheduling practices and local conditions matter when you interpret the responses. A sample restricted to one customer service channel may help you describe a shared kind of work. Check whether you can reach enough people in that group during the time available.

If your title contains two broad terms such as “stress and burnout”, decide what you need to examine on each side. “Stress” can refer to perceived strain, conditions experienced as demanding or a response to those conditions. Different approaches describe burnout through different dimensions. Once you have written down the group, the variable and the intended measure, you can judge whether a paper about general stress helps answer your question about deadline pressure.

In what order should you write the chapter?

For an empirical thesis, try these five steps. The order follows what the reader needs to know before reaching your question and description of the measures.

1. Concept and occupational context. Explain which part of life occupational burnout concerns and what scope you are adopting. Use an example to distinguish short-term tiredness from the dimension of exhaustion you plan to study.
2. Comparison of approaches. Show what they share, what people distance themselves from in each approach and how each treats professional efficacy. Finish by explaining which approach fits the research question.
3. The working condition in the project. If you ask about time pressure, explain what you mean by it and how it differs from other demands. Choose theoretical predictions that apply to the variables you will examine.
4. Previous research. Organise findings by question, population or dimension. For each important comparison, state the instrument and the feature of the study design that matters to the conclusion.
5. Summary and transition to the study. State the chosen definition, identify the dimensions represented in your measures and give the research question. Put the detailed scoring procedure in the methods chapter.

For a literature-based thesis, use the final step to explain how you will compare papers. You could say that you will discuss studies of exhaustion and disengagement separately from studies that also measure professional efficacy. Readers then know why you have divided the material before they reach the analysis. A list of sections named after authors tells them who wrote about burnout and leaves them to work out how the definitions compare.

Check the balance of the outline. If you have four subsections on the history of the term and a single paragraph on research involving your group, write down what each part contributes later in the thesis. History is worth expanding when it explains why the target of measurement changed, for example from attitudes towards service recipients to attitudes towards the work itself. A thesis about time pressure will often need more space for defining that demand, because the results depend on what the measure captures.

How do you explain burnout, exhaustion and stress?

Distinguish three levels of description. A deadline can be a job demand. “I felt tense towards the end of my shift” is a report of an experience. An exhaustion subscale score comes from a particular set of responses and a scoring procedure. Each kind of statement needs its own support. A claim about how work is organised needs information about how the organisation was studied. A statement about an experience should identify who reported it. An interpretation of a score depends on the scope of the measure.

What can you use the WHO definition for?

The World Health Organization (WHO, 2019) describes burnout in ICD-11 as an occupational phenomenon associated with chronic workplace stress that has been managed unsuccessfully. Its description covers exhaustion or a loss of energy, increased psychological distance from work or a negative or cynical attitude towards it, and reduced professional efficacy. WHO classifies burnout as an occupational phenomenon rather than a medical condition. In a theory chapter, this definition establishes the work context and the dimensions that the classification includes. To describe a research instrument, consult its own documentation.

You can bring the two levels together in two sentences. The first establishes the work context using ICD-11. The second says that your project will measure exhaustion and disengagement through the two-dimensional OLBI approach. When discussing the questionnaire results, use those two subscale names. When summarising WHO's description, retain all three dimensions. That wording helps you avoid claiming to have “measured burnout according to WHO” with a questionnaire that gives no separate measure of professional efficacy.

If your thesis involves students who work in Poland, explain which role your questions refer to. You can study the paid work experiences of people who also attend university. If you are asking about classes, assessments and how participants relate to their studies, look for literature and instruments on academic burnout. The participants' status as students is only one part of the description. The focus of the questions and the wording of the responses determine which concept fits.

Why is tiredness after work insufficient on its own?

Imagine two employees who want to rest after a demanding shift. One puts her tiredness down to an unusually large number of requests that day and is keen to get back to work the next morning. The other describes a recurring lack of energy and growing disinterest in her duties. This fictional example shows what the statement “both were tired” leaves out. It gives us no account of how those experiences developed, how they relate to work or how the employees would answer questions about other dimensions.

In some approaches, exhaustion covers physical, emotional and cognitive experiences. You can explain these as a lack of physical energy, feeling emotionally overloaded and finding it hard to summon the mental resources a task requires. These descriptions make the theory easier to read. In a study, use the instrument documentation to choose your indicators. A question you write yourself about sleepiness after work tells you about reported sleepiness over the specified period. If you include it, label the response accordingly.

For the same reason, state what a high score means in the specific scale. In an academic analysis, you can report that one group scored higher on exhaustion than a comparison group or that exhaustion scores were associated with responses about time pressure. An individual assessment would require a different basis. Survey data on two dimensions support interpretations of the two subscale scores in the sample.

Where do stress, job demands and resources belong in the chapter?

Define stress and burnout separately, then explain the proposed relationships between them. If your question is about time pressure, the Job Demands-Resources model, or JD-R, can help (Demerouti et al., 2001). Job demands are aspects of work that require effort. Resources can help people do the work or manage its demands. In the original study, demands were associated primarily with exhaustion, and limited resources primarily with disengagement. That gives you a basis for considering each dimension separately. Data from your group or closely relevant literature are needed to support the size and direction of an association in that setting.

You might illustrate time pressure with a queue of open customer requests and short deadlines. Access to advice from an experienced colleague when handling a difficult case is an example of a resource. The length of the queue alone gives no direct measure of how rushed someone feels. Likewise, saying that advice is available and having a record of advice that was actually given tell you different things about the same workplace. Choose terms that match the information you can obtain. If your question concerns time pressure alone, cover resources to the extent needed to explain the model. In your results, examine relationships between the variables you actually measured.

An early draft can easily mix possible causes, features and outcomes. Try rewriting “Haste, exhaustion and absenteeism are symptoms of burnout” by giving each term a role. Haste might represent a demand or how an employee appraises it. Exhaustion is a dimension in the chosen approach. Absenteeism is a separate indicator of functioning that would require its own data and rationale. The study design determines how each variable is used.

How do you compare Maslach's approach and OLBI without mixing up concepts?

Compare the approaches by asking what exhaustion covers, what people distance themselves from and how professional efficacy is treated. You can then explain which part of burnout a given study describes. When referring to the Maslach Burnout Inventory (MBI), name the version if the authors specify it. For the Oldenburg Burnout Inventory (OLBI), name the adaptation you plan to use.

What do the approaches share and what changes between versions?

Exhaustion is common to both, but the label alone does not tell you which experiences the items cover. In Maslach's classic approach developed in helping professions, emotional exhaustion is one of three dimensions, alongside depersonalisation and reduced personal accomplishment. The MBI-General Survey (MBI-GS) covers exhaustion, cynicism and professional efficacy. Read an efficacy score in the direction specified in the documentation for that version. Mańkowska (2018) draws attention to changes in the terminology. Chirkowska-Smolak and Kleka (2011) report a psychometric study of the Polish MBI-GS.

In the human services version of the MBI, “depersonalisation” concerns attitudes towards the people who receive the service. “Cynicism” in the MBI-GS has a broader target, the work itself. If the authors measured cynicism, keep that label when reporting their study. If they measured depersonalisation towards service recipients, retain that specific target. You can discuss both dimensions in relation to distance, as long as you state what the distance was from. Substituting the terms changes what the reader understands was measured.

The Polish version of the OLBI studied by Baka and Basińska (2016) has two theoretical dimensions, exhaustion and disengagement from work. Its exhaustion dimension includes more than emotional experiences, and disengagement concerns attitudes towards the work and its content. The instrument provides no separate professional efficacy subscale. Some aspects of disengagement may be related to how people view their own work, but the disengagement score does not give a separate efficacy measure. Conclusions about three distinct dimensions would require measures covering all three.

How do you compare studies that used different instruments?

Check what each study's result is based on. You can state whether both assessed exhaustion, how they measured it and which differences between the groups limit comparison. Mean scores from two instruments often come from different response ranges and scoring procedures. Converting them to percentages of their maxima does not show that both scores represent the same degree of exhaustion. In a descriptive comparison, you can report whether both studies found associations in the same direction for a particular demand. State how each study measured the demand and exhaustion.

Take particular care with studies that report percentages above a threshold. One may apply a threshold to exhaustion, while another uses a rule based on several subscales. Each percentage may be correct under its own definition. A claim that burnout was more common in one group would require comparable classification rules and comparable sampling. When that basis is missing, state the criterion behind each percentage and explain what it allows you to conclude. Leave the numbers out if putting them side by side would imply a comparison the studies cannot support.

Make a short note for each paper. Record the population, recruitment method, instrument language and version, subscale, type of result and where in the paper you can check it. Add a limitation relevant to your own question. If the main difference is the version of the instrument, explain that before presenting the results. If both papers used the same version, but one studied a single team and the other covered several workplaces, explain why that context matters. The notes let you see whether the difference is in the concept, measurement, sample or more than one of these.

How do you check whether an instrument fits the question?

Start by naming the result that will answer the question. If you ask about professional efficacy as a separate dimension, you need a measure of it. If you ask about time pressure and both OLBI dimensions, you need the two OLBI subscales and an appropriate time pressure measure. Check access to this set of measures and its suitability for your group before agreeing on the outline. Discovering a missing measure after collecting responses forces a change to the question and limits what you can conclude from the data.

What should you look for in an adaptation study?

Check the language, who took part and how the instrument was examined. Baka and Basińska (2016) describe the Polish OLBI using three samples of people in social service and other service occupations. They examined how responses grouped together, how consistent the subscale scores were and how stable they were across repeated measurements. They also assessed associations with perceived work stress and work engagement. Use several kinds of evidence to decide whether the instrument fits your project. Similarity in language, work setting and measurement purpose matters more than repeating the overall judgement in the abstract.

The paper also describes a difficulty relating to item wording. In the exploratory analysis, responses grouped partly according to whether the items were worded positively or negatively. This kind of analysis looks for patterns of responses that tend to occur together. The authors then analysed the theoretical subscales separately and described limits to testing the structure with more advanced methods. When justifying your choice, explain why you are analysing the two subscales separately and describe the limitations of the study of their structure. Positive and negative wording can affect responses, so an extra group of items in the analysis does not in itself establish another psychological dimension of burnout.

Reliability concerns aspects such as consistency or stability, depending on how you assess it. Validity concerns the evidence for interpreting a score in a particular way. If the authors report Cronbach's alpha, it describes one form of consistency among responses to items on a subscale. To justify interpreting a score as exhaustion in a given group, you also need information about item content, structure and research on what the score is associated with. In the theory chapter, explain how these types of evidence serve different purposes. Put numerical values where you describe the instrument or report your own results and identify the sample from which they came.

What should you check about thresholds, language and scoring?

Norms compare a score with those in a reference group. The Polish OLBI paper presents stanine norms for the subscales in a sample of social service workers. Stanines divide the reference scale into nine categories. A category based on that sample does not establish that a person has burnout or that the reference applies to another occupation. If you want to examine relationships between variables, ask what a set of categories would contribute. Analysing the subscale scores may answer your question more directly than dividing participants into groups using a cut-off with no clear basis.

Distinguish the language of your thesis from the language of the questionnaire. If you are writing in English or Ukrainian, you can give participants the documented Polish version of an instrument if Polish suits the group. If respondents need another language, look for the relevant version and evidence on how it has been used. To compare scores across languages, explain why the measure can be interpreted comparably across versions. Translating the items yourself changes the instrument and needs its own justification, even if you cite the Polish adaptation. Record the subscale names in both languages if the thesis and the documentation use different ones.

Before collecting responses, check the terms for using the chosen version, access to the scoring instructions and whether the survey matches those instructions. Record which items need reverse coding, how each subscale is scored and how you will handle missing answers. If you plan to use a combined score, check whether the instrument provides for one and whether it answers your question. The example in this guide needs two separate subscale scores. An average across all items could hide different patterns of association and make it harder to explain which result is related to time pressure.

What goes in theory and what goes in methods?

In the theory chapter, explain why the dimensions matter to your question and how they relate to the chosen definition. In the methods chapter, provide the full instrument name and version, the response range, the scoring rules and how you used the instrument with your participants. In the results, report what you found. If you assess reliability in your own data, specify the subscale and the set of responses on which the estimate is based. The reader can then follow the concept through to measurement and results without a change in meaning along the way.

Check the distinction by sorting the statements about your instrument. “Disengagement concerns attitudes towards work” explains a concept in theory. “We scored the subscale according to the instructions for the specified version” describes a method that needs more detail in the completed thesis. “Higher disengagement scores were associated with higher time pressure scores in the study sample” belongs in the results and can only be written after analysis. In your outline, leave room for such statements by using questions and planned actions.

A psychology example. How do you build a chapter from two publications?

Marta studies psychology in Poland and is considering a survey of adults who provide customer service by phone. She has two full-text papers, a working title about burnout and seminar feedback asking her to define the scope of the concept. This example shows how she develops her outline. The papers are real. Marta, her project and her notes are fictional, and no data were collected.

What does Marta record after reading the two sources?

First source note. Beata Mańkowska (2018). This paper critically examines how burnout is understood and measured. Marta uses the section on Maslach's approach and the passage comparing it with OLBI. She notes that her comparison must distinguish distance towards service recipients, cynicism about work and disengagement in OLBI. She also marks the separate role of professional efficacy as an important difference. This source helps her explain why she will compare the approaches using the same questions. She makes clear which judgements are Mańkowska’s.

Second source note. Łukasz Baka and Beata A. Basińska (2016). This study examines the properties of the Polish OLBI. Marta marks the passages explaining exhaustion and disengagement, the samples, the results and the limitations. She notes that her study needs two scores, one for each dimension. The paper supports her description of the specific adaptation. The third sample included 46 service workers and was used to assess stability over time. The sample description does not say whether they worked in telephone-based customer service. The Polish version and the inclusion of service workers give Marta a starting point. She still needs to check whether the instrument suits her group.

Marta adds “Professional efficacy” to both notes. In the first, she records its separate role in the three-dimensional approach. In the second, she records that OLBI gives no separate efficacy score. She also identifies what each source contributes. The first helps her explain where the approaches differ. The second shows which scores the chosen questionnaire will provide. For a more detailed description of the MBI-GS, Marta would also consult research on that version or its documentation.

How do those notes become a synthesis paragraph?

Marta's first draft reads “Mańkowska describes three dimensions of burnout. Baka and Basińska describe two. I will use the OLBI.” It leaves out which dimensions can be compared and why the difference matters to her study. She gives the paragraph a specific job. The reader should see what choosing two OLBI scores means for the question about time pressure. A revised passage might read as follows.

The approaches discussed by Mańkowska (2018) share a focus on exhaustion, although the scope of exhaustion varies by approach. They differ in what people distance themselves from and the role assigned to professional efficacy. The Polish OLBI measures exhaustion and disengagement from work as separate dimensions (Baka and Basińska, 2016). This project adopts that approach because it examines how perceived time pressure is associated with each dimension among telephone-based customer service workers. The OLBI will allow the two scores to be analysed separately. A separate assessment of professional efficacy falls outside the scope of this project.

The first sentence identifies common ground and the next identifies a difference. Marta cites the study of the specific adaptation when she describes the measure. The final sentences explain Marta's decision and its consequences, keeping her own research question distinct from the published work. Before this paragraph she could explain what “people distance themselves from” means in the two approaches. If her supervisor wants a separate section on models, she can explain the differences between versions there and keep this passage as the transition to her own project. For more on building this kind of paragraph, see the guide to writing a literature review.

How does the paragraph lead to a question, a measure and an analysis plan?

Marta's first question was “Does time pressure cause occupational burnout in employees?” A one-time survey can show whether scores are associated. She changes her question to “How is self-reported time pressure associated with exhaustion and disengagement from work among the adult telephone-based customer service workers in the study?” In her plan she sets out two analyses, one for time pressure and exhaustion and one for time pressure and disengagement. She explains the reasons for each and looks for studies that address time pressure directly. A conceptual review and an adaptation study are enough to plan a comparison of approaches. A directional hypothesis also needs evidence about the relationship itself and a similar group.

Marta identifies what still needs to be decided. She checks the OLBI documentation and looks for an instrument that measures perceived time pressure at work. If a measure covers workload more broadly, she checks whether it includes a separate time pressure score. If it does not, she can use a different measure or explicitly revise her question and outline. Her thesis will need to explain participant selection, the survey language and how she dealt with each subscale. She takes her question, two source notes and a description of the information she can obtain to a consultation. That gives her enough to discuss the scope before building the survey.

Marta also plans how to interpret the results. If the analysis eventually finds an association with exhaustion only, she will say so and present the disengagement result separately. If an estimate is imprecise, she will reflect that uncertainty in the interpretation. If both subscales are associated with reported time pressure, questions about the order of events and other working conditions will still matter. A one-time survey records responses in a single period and cannot establish by itself whether time pressure came before exhaustion. Marta records these limits before analysing any data.

How do you find papers that fill gaps in the chapter?

Search separately for sources on definitions, measurement and the relationship you plan to study. A conceptual review helps you identify debates about dimensions. An adaptation study provides evidence about a specific version of an instrument. A study of time pressure and exhaustion in a particular group supports a claim about an observed association. If you cite the same paper in all three places, check that it contains evidence or an argument for each claim.

How do you search when the results are too broad?

Combine a concept with the task at hand. For an instrument, try “wypalenie zawodowe zdystansowanie OLBI polska adaptacja”. For research on Marta's question, try “presja czasu wyczerpanie pracownicy obsługi klienta”. In a scholarly search engine, try an English query such as “time pressure exhaustion call centre workers” as well. If most studies cover other jobs, check whether the settings are similar in the particular working condition you are studying. Both jobs might involve a short response deadline, while differing considerably in autonomy or the consequences of missing a deadline. When you use a paper from an adjacent field, explain exactly what it contributes.

Once you find a relevant review, follow its references to the original study if you want to cite a result. Read the methods, results and limitations and note where the paper supports each statement. If you can only access an abstract, write down which details still need checking. Missing information about the instrument version or recruitment can make a paper suitable for background and unsuitable for a central comparison. Keep sources in the references if you use them in the chapter.

What if published findings differ?

Check whether the studies asked similar questions. One might examine perceived time pressure and another broad emotional demands. One might report exhaustion and another a composite of several subscales. The findings may only look inconsistent because they concern different relationships. You can present both studies and explain the variables and limits of comparison. When the questions are similar, check sampling, when measurements were taken, the measures and the precision of the estimates before attributing the difference to the occupations themselves.

In a review, distinguish a statement about the literature you found from a general claim. “The studies of telephone-based customer service I identified rarely reported disengagement separately” needs a description of the search and examples. “Nobody has studied disengagement” needs a much broader basis. If you find few studies that address your exact question, explain the limitation and why your question is worth pursuing. You can also leave the dimension with less supporting evidence in the research question without proposing a directional hypothesis about it.

What should you check before sending the chapter to your supervisor?

Read the introduction, the last sentence of each major section and then the research question. Together they should lead from the chosen meaning of burnout to the measures and aim. Where a transition is unclear, revisit the relevant paragraph and state the decision. If readers are introduced to three dimensions and then see two scores, explain why you selected OLBI and what it covers. If the theory concerns perceived time pressure, check that the research question still uses that term.

How do you catch conclusions that go too far?

Underline verbs such as “causes”, “leads to”, “proves” and “diagnoses”. For each, ask what evidence justifies the wording. In a one-time survey, “was associated with”, “co-occurred” or “participants reported” will usually fit. If you describe a theoretical mechanism, say that the model proposes it. If you cite an empirical study, state what the authors compared. Apply the same wording principles in the title, abstract and discussion.

Check nouns and group labels as well. “Higher exhaustion scores” states the comparison. “The more burned-out group” can suggest a comparison of the whole phenomenon or a classification of individuals. “Employees in the team studied” preserves the scope of the sample. “Customer service workers in Poland” names a larger population and requires an appropriate sampling basis and argument. Adding a few words about the measurement and where the data were collected can make the claim more precise.

What should you bring to a discussion about the next step?

Prepare a short outline, a working definition, your research question and two notes on the key papers. Add the decision you need help with. It may be choosing between two approaches, narrowing a broad question or deciding whether you can compare data collected in two languages. If your seminar specifies a chapter structure or citation style, bring the current guidance and see where your argument fits within it. If the thesis has an empirical component, discuss the instrument documentation and research with human participants before recruitment, following your unit's requirements.

On the last page of the chapter, try writing one sentence that names the chosen approach and one consequence. Marta might write “This project adopts the two-dimensional OLBI approach, which allows separate questions about how perceived time pressure is associated with exhaustion and with disengagement.” For a literature-based thesis, the equivalent could identify the dimensions you will use to organise research and when you will consider results comparable. That final sentence helps you check the chapter against the work you are actually planning.

Sources

• Baka, Ł., Basińska, B. A. (2016). Psychometryczne właściwości polskiej wersji Oldenburskiego Kwestionariusza Wypalenia Zawodowego (OLBI). Medycyna Pracy, 67(1), 29-41. DOI 10.13075/mp.5893.00353.
• Chirkowska-Smolak, T., Kleka, P. (2011). The Maslach Burnout Inventory-General Survey: validation across different occupational groups in Poland. Polish Psychological Bulletin, 42(2), 86-94. DOI 10.2478/v10059-011-0014-x.
• Demerouti, E., Bakker, A. B., Nachreiner, F., Schaufeli, W. B. (2001). The job demands-resources model of burnout. Journal of Applied Psychology, 86(3), 499-512. DOI 10.1037/0021-9010.86.3.499.
• Mańkowska, B. (2018). Wypalenie zawodowe. Dylematy wokół istoty zjawiska oraz jego pomiaru. Polskie Forum Psychologiczne, 23(2), 430-445. DOI 10.14656/PFP20180212.
• World Health Organization. (2019). Burn-out an "occupational phenomenon": International Classification of Diseases. Update of 28 May 2019.