Writing a PhD on Student Satisfaction: Constructs, the NSS and PTES Surveys, and How to Use the Published Data (2026)

The key point for a thesis on student satisfaction is that the major UK surveys measure different students on different scales: the National Student Survey collects final-year undergraduate opinions on the quality of their course, and the Postgraduate Taught Experience Survey covers taught master’s, certificate and diploma students. Neither measures research students, and neither is the same construct as engagement.

Why is student satisfaction a useful PhD construct?

Student satisfaction sits at the centre of higher education research because universities measure it, publish it and are judged on it. That gives a doctoral researcher three advantages: a large public data source, a body of instruments and a live policy debate about what satisfaction scores mean. It also gives three risks. The published surveys cover particular groups of students, their themes are fixed by the survey designers and not by your research question, and satisfaction can be confused with neighbouring constructs. A strong thesis defines the construct, chooses its data deliberately and is explicit about what the data cannot show.

What exactly does the NSS cover, and what does it leave out?

The Office for Students describes the NSS as gathering final-year undergraduate students’ opinions on the quality of their course. Its page lists the themes in the table above, and publishes the 2026 results with downloadable Excel and CSV spreadsheets. Note two details on that page: the single overall satisfaction question was asked only to students at providers in Northern Ireland, Scotland and Wales, on a five-point agreement scale that the page says is not directly comparable with the other questions, and the freedom of expression question was asked only at providers in England. That structure has three further consequences for design. First, the survey is of final-year undergraduates, so a thesis about postgraduate or part-way experiences cannot use it as a direct measure. Second, the published data are summaries by theme and by provider or subject, not individual-level responses, so analyses of NSS data are analyses of groups, and conclusions about individual students would commit an ecological fallacy. Third, the theme list is a fixed set of questions; if your construct is broader or narrower than the survey’s, you will need your own measures. This guide deliberately quotes no percentages from the survey: the figures change each year, and your thesis should take the current release directly from the source, with its date, rather than from any summary.

What does PTES add for postgraduate students?

Advance HE describes PTES as gathering insights from taught postgraduate students on master’s, postgraduate certificate or diploma courses. It examines learning and teaching experience, motivations for enrolling and, where relevant, experience of dissertations or major projects, and it covers aspects that the page says other surveys do not, such as student enhancement and engagement. It lets institutions benchmark their postgraduate provision against peers. If your thesis is about the taught postgraduate experience, PTES is a natural source of questions and context, but note that institutional access to the detailed data is typically governed by participation, so check what you may obtain and under what terms before designing a study around it.

How is satisfaction different from engagement and experience?

These are the neighbouring constructs an examiner will ask you to separate.

• Satisfaction is an evaluation: students’ assessment of how well their course or institution meets their expectations or needs. Oliver (1980), “A Cognitive Model of the Antecedents and Consequences of Satisfaction Decisions” in the Journal of Marketing Research, is a standard marketing-literature reference for how satisfaction is modelled, and Elliott and Shin (2002), “Student Satisfaction: An alternative approach to assessing this important concept” in the Journal of Higher Education Policy and Management, applies satisfaction thinking to students specifically. Read both before borrowing their logic, and cite the passages you use.
• Engagement is about what students do and how they invest effort. Kuh (2009), “The national survey of student engagement: Conceptual and empirical foundations” in New Directions for Institutional Research, summarises the history of the engagement concept and the development of the National Survey of Student Engagement.
• Experience is the broader description of what a student goes through, of which satisfaction is one evaluation. PTES, for example, covers experience, enhancement and engagement and not only satisfaction.

A thesis that uses “satisfaction” and “engagement” interchangeably will be queried. Choose one as your outcome, define it, and treat the other as an explanatory variable or a different outcome. The engagement construct on the employment side, with its own instrument, is set out in employee engagement as a PhD construct, which is a useful comparison of how a construct, definition and scale are fitted together.

How should you choose measures for your own study?

If you collect your own data, you have three options. You can adopt a published scale for the construct, adapt items from the national surveys, or write new items. Each carries an obligation. A published scale needs its validation evidence described and its reliability reported in your sample. Adapted items need the permission or terms of use from whoever owns them, and you should say precisely what you changed. New items need the full development path of expert review, piloting and testing. Because the NSS and PTES are designed to compare providers, their items are written for that purpose, which may not match an individual-level research question. Whichever route you take, state it in the methods chapter and keep the reasoning visible, so that an examiner can see why the instrument fits your construct.

Which designs fit a student satisfaction thesis?

1. Secondary analysis of published survey data. Using NSS or PTES releases at provider or subject level to test group-level hypotheses. It is quick and transparent, but limited to the published categories and to group-level inference.
2. Primary survey of your own sample. A questionnaire built from the constructs you define, possibly alongside borrowed items. It allows individual-level analysis, and it needs the instrument evidence set out in what examiners expect on validity and reliability and a platform choice described in survey platforms compared for PhD data collection.
3. Mixed methods. A survey followed by interviews that explain why particular groups score differently, which is a common way to add depth to a satisfaction thesis.
4. Qualitative case study. A single department or programme, which suits questions about how students interpret quality and feedback.

Whatever the design, think early about sampling and access, for which researching a specific population in an education PhD offers a worked example of access and ethical design. If your framework needs a theory, the choices used in education doctorates are compared in which theoretical framework an education PhD should use.

What ethics and access questions come up?

Studies of students in your own institution raise their own issues. If you survey students you teach or assess, consent may not feel free, so recruit through a third party where you can. Satisfaction data can reveal criticism of named staff or courses, so agree early how results will be reported and who sees them first. If you use institutional survey data rather than publicly released data, you need written permission and a clear statement of what you may publish. Your university’s research ethics committee will want to know all of this, and the answers belong in the methods chapter.

What hypotheses can you test?

These are illustrative statements that name a defined construct and a level of analysis. They are examples, not findings.

• H1 (provider level): Subjects with higher published scores on the assessment and feedback theme also have higher published scores on the academic support theme, using the most recent release.
• H2 (individual level): Perceived feedback usefulness is positively associated with course satisfaction among final-year students at one institution.
• H3: The association in H2 is stronger for students who report more frequent contact with academic staff.
• H4 (taught postgraduates): Satisfaction with dissertation supervision is positively associated with overall course satisfaction.
• H5 (mixed methods): Students who describe feedback as actionable in interviews score higher on a feedback scale than those who do not.

Ten researchable titles

These are illustrative titles shaped by design and population, not published studies.

1. Assessment and feedback and overall satisfaction across subject areas in Scotland, Wales and Northern Ireland: a secondary analysis of published NSS data.
2. What counts as helpful feedback? A mixed-methods study of final-year students in one faculty.
3. Satisfaction, engagement and attendance: separating three constructs in a first-year cohort.
4. Dissertation supervision and course satisfaction among taught postgraduate students.
5. Do international and home students interpret “academic support” differently? A measurement invariance study.
6. Student voice and the sense of being listened to: a qualitative case study of course representation.
7. Satisfaction with learning resources in blended courses: a survey of two institutions.
8. How students explain a low satisfaction score: interviews following a survey.
9. Expectation and disconfirmation in postgraduate satisfaction: a longitudinal study across one master’s year.
10. Comparing a published satisfaction scale with an institutional survey: convergent validity in one department.

What goes wrong in satisfaction theses?

• Treating published group-level data as individual-level evidence.
• Applying NSS findings to students it does not survey, such as postgraduate research students.
• Quoting a figure without its release year and source. Survey results change each year.
• Using satisfaction and engagement as synonyms.
• Assuming a high score means high quality. Satisfaction is an evaluation, and the thesis should say what it does and does not indicate about teaching quality.
• Ignoring response patterns. If your own survey has a low or uneven response, say so and discuss what it limits.

Frequently asked questions

Who does the National Student Survey cover?

The Office for Students describes it as gathering final-year undergraduate students’ opinions on the quality of their course. It does not survey postgraduate research students.

Which survey covers taught postgraduates?

The Postgraduate Taught Experience Survey, run by Advance HE, covers taught postgraduate students on master’s, postgraduate certificate or diploma courses.

Can I download NSS data for my thesis?

The Office for Students page offers the 2026 NSS data for download as Excel and CSV spreadsheets. Cite the release and date you used.

Is the NSS data individual-level?

The published data are summaries by theme and by group, so analyses of them are group-level analyses. Do not draw conclusions about individual students from them.

What is the difference between satisfaction and engagement?

Satisfaction is an evaluation of how a course or institution meets expectations or needs; engagement concerns what students do and the effort they invest. Kuh (2009) summarises the engagement concept and the development of the National Survey of Student Engagement.

Which theory frames satisfaction research?

Marketing and higher education literatures both model satisfaction, for example Oliver (1980) and Elliott and Shin (2002). Choose the one whose assumptions suit your design and cite the original.

Can I use NSS items in my own questionnaire?

Check the terms of use for the survey questions with the Office for Students before you reuse them, and say in your methods chapter what you adapted.

Can I study satisfaction among PhD students?

Yes, but use a source that covers them. PRES is the research student survey, discussed in the linked guide, and the NSS and PTES do not cover that group.

How many respondents do I need?

It depends on your design and analysis. Justify the number with a power or precision argument tied to your hypotheses, not a rule of thumb.

Is a cross-sectional survey enough?

For associations, yes, provided the claims stay at that level. For change over a course, use a longitudinal design.

From construct to chapters

A satisfaction thesis depends on careful definitions and a clear account of what its data can show. Tesify is used by 9,000+ students across 15,000+ chapters to organise a thesis chapter by chapter, and the thesis stays 100% written by you. The construct, the data and the interpretation are yours; a chapter plan helps you see whether each is stated where an examiner will look for it.