ACADEMIC WRITING GUIDE

How to Calculate and Report a Survey Response Rate in a Dissertation (With Worked Examples)

By Acadelyra Editorial TeamPublished

Suppose you invited 300 students to complete your dissertation survey and received 174 questionnaires. Can you report a 58% response rate? Maybe—but only if those 300 invitations represent the appropriate eligible sample and the 174 responses meet the completion criteria you defined. If some invitations reached ineligible people, some addresses were undeliverable, or some respondents submitted only partial answers, the calculation needs more explanation.

A survey response rate tells readers something about participation. It does not, on its own, establish that respondents represent the people you intended to study. This guide shows how to keep track of survey outcomes, calculate defensible rates, distinguish response rate from completion rate, assess possible nonresponse bias, and write a transparent dissertation methods or limitations paragraph. All counts and study situations below are hypothetical teaching examples, not evidence from real participants.

What is a survey response rate?

A survey response rate is the share of the relevant sampled eligible units that provided a response, calculated under an explicitly defined method. A unit might be an individual student, a household, a school, or another entity in the sampling design. Professional survey standards distinguish complete interviews, partial interviews, refusals, noncontacts, ineligible cases, and cases whose eligibility is unknown. Those categories can affect the denominator.

For a simple, fully known eligible invitation list with no partial interviews, the basic formula is:

Response rate (%) = (number of eligible completed responses ÷ number of eligible sampled people) × 100.

That formula is appropriate only under its stated assumptions. More complex surveys should use an identified standard definition, such as an appropriate American Association for Public Opinion Research (AAPOR) response-rate formula, and report how unknown eligibility and partial interviews were treated.

A high response rate can reduce some concerns, but it does not certify representativeness. A low response rate signals a need to examine potential nonresponse bias; it is not proof that bias occurred. AAPOR's response-rate guidance explains why rate calculation and survey quality assessment should be reported together.

Response rate, completion rate, and item response: do not confuse them

Survey response, questionnaire completion and item response compared
MeasureWhat it asksIllustrative calculation
Survey response rateWhat proportion of the eligible sampled cases responded under the stated definition?174 complete responses ÷ 300 eligible invitations = 58%
Questionnaire completion rateOf participants who started a questionnaire, what proportion completed it?174 completed ÷ 210 started = 82.9%
Item response rateOf people asked a particular question, how many gave a usable answer?162 usable answers ÷ 174 asked = 93.1%

These percentages describe different stages and use different denominators. A web-survey platform may display a “completion rate” for visitors who started the survey. Do not relabel that as the response rate for everyone invited. When survey links are open to the public and the number of eligible people exposed to the invitation is unknown, a conventional invitation-based response rate usually cannot be calculated reliably. You can still report valid completions and the observable recruitment or participation metrics, with that limitation stated.

Step 1: Define exactly who was invited

Begin with your sampling and recruitment records. Identify the unit of analysis, target population, sampling frame, invitations sent, and eligibility criteria. For example, “first-year students enrolled in a particular programme” is more precise than “university students.”

A sampling frame may contain outdated addresses or people who do not meet your criteria. A count of 300 emails sent is not necessarily a count of 300 eligible students contacted. Do not silently exclude people who refused or failed to reply simply to produce a higher rate. Where people never received an invitation, or eligibility is unknown, record this and use a rate definition that handles their disposition appropriately.

Your sampling-method and sample-size decisions determine which population your participation rate describes. A convenience sample does not become a probability sample because a large proportion replied.

Step 2: Create a transparent survey disposition record

Before calculating percentages, record the final status of every sampled case where possible. In a simple student survey, your tracking table might distinguish complete questionnaires, eligible partial questionnaires, explicit refusals, people who never replied, undeliverable invitations, and people later confirmed ineligible.

Hypothetical survey disposition: all 300 sampled records accounted for
Final statusHypothetical countReporting decision
Complete eligible responses174Count as complete responses
Eligible partial responses16Describe separately; inclusion depends on the specified rate formula
Eligible refusals35Remain in the eligible denominator
Eligible noncontacts/nonreturns65Remain in the eligible denominator
Confirmed ineligible10Exclude under the declared rule
Total sampled records300Account for every record

In this teaching example, the known eligible cases total 290 (300 minus 10). Do not discard the 16 partial cases or quietly merge them with complete questionnaires; first decide and disclose your definition of a usable response. In an actual study, there may also be unknown-eligibility records, which can require a more advanced formula. The example deliberately assumes eligibility is established for all records.

Keep an auditable log without exposing participants' identities unnecessarily. Recruitment and consent procedures should be consistent with the study's ethics approval. Our data-collection procedure guide explains how to describe invitation and collection steps.

Step 3: Calculate a defensible rate and show your working

Worked example 1: All invitees are known to be eligible

A researcher has a verified list of 250 eligible postgraduate students. After one defined recruitment period, 155 students submit complete usable questionnaires; no partial submissions are counted as responses under the project's declared complete-response definition.

Complete-response rate = (155 ÷ 250) × 100 = 62%.

The report should identify the invitation list, the eligibility verification, the field period, and the definition of “complete.” It should also disclose any other outcomes (refusals, noncontacts and partials) if available. A 62% rate does not justify claiming that the respondents mirror all postgraduate students.

Worked example 2: Confirmed ineligible cases and partial interviews

Use the 300-record disposition table above. Ten cases are confirmed ineligible, leaving 290 known eligible cases. There are 174 complete and 16 eligible partial questionnaires.

Under a complete-only definition:

Complete-response rate = (174 ÷ 290) × 100 = 60.0%.

Under an explicitly defined complete-plus-qualifying-partial calculation, if those 16 partial cases meet the prespecified criteria for inclusion:

Complete-or-qualifying-partial rate = ((174 + 16) ÷ 290) × 100 = 65.5% (rounded).

These are illustrative calculations, not an assertion that either equals a particular AAPOR standardized rate. To make a formal AAPOR claim, classify dispositions using its current definitions and apply the corresponding formula exactly. Explain whether partially completed questionnaires satisfy the definition of a partial interview; any partial questionnaire does not automatically qualify.

A transparent results sentence would be:

Of 300 sampled records, 10 were confirmed ineligible, leaving 290 eligible cases. We received 174 complete questionnaires (60.0% of eligible cases) and 16 additional partial questionnaires. Including the latter under our stated qualifying-partial definition yields a 65.5% participation measure. These figures use known-eligibility denominators and are not presented as standardized AAPOR response rates.

Worked example 3: An open online link with an unknown denominator

A dissertation survey link is distributed through student social-media groups. The researcher records 210 survey starts and 147 complete submissions. The number of eligible students who saw the invitation is unknown, and the link may have been reshared.

Completion rate among recorded starts = (147 ÷ 210) × 100 = 70%.

The researcher cannot defensibly describe 70% as an invitation-based survey response rate. The number of people who were eligible and had the opportunity to respond is not known. A suitable statement is:

The open-link survey produced 147 complete submissions from 210 recorded starts, a 70.0% completion rate among starts. Because the invitation circulated through student groups and the number of eligible recipients was unknown, an invitation-based response rate could not be calculated. The self-selected recruitment method limits generalisation beyond participants.

This distinction is particularly important for convenience and snowball sampling.

Step 4: Describe missing answers separately

Unit nonresponse occurs when an eligible sampled person does not provide a usable survey response. Item nonresponse occurs when a survey participant skips or cannot answer a particular question. These require separate counts and potentially separate decisions.

Suppose 174 participants completed enough of the questionnaire to qualify under your defined rule, but only 162 answered the study-time question. That question has an item response rate of 162 ÷ 174 × 100 = 93.1% among those asked. Its analytic sample size may be 162, not 174.

Do not imply that the dataset is complete merely because the questionnaire was submitted. Explain your missing-data approach, variable-specific denominators and any relevant analysis exclusions. Our guide on cleaning research data before analysis helps students distinguish missing values from invalid and inconsistent entries.

Step 5: Evaluate possible nonresponse bias—not just the percentage

Nonresponse bias can arise when those who respond differ from those who do not in ways related to the variables being studied. For example, if students struggling most with academic workload are less able to complete a time-management survey, results may overstate how successfully students manage their time.

The response rate alone does not quantify that bias. Pew Research Center's survey-methodology discussion explains why response patterns, population benchmarks and research design matter alongside the rate itself.

Where feasible, consider the following checks:

  1. Compare known respondent characteristics with the sampling frame. If enrolment records lawfully provide programme, year or other non-sensitive aggregate characteristics for everyone invited, compare those distributions with the respondents. A difference suggests possible representation problems, though similarity on measured characteristics cannot rule out bias on unmeasured ones.
  2. Compare early and later respondents cautiously. Later respondents may differ from early ones, but late response is not a validated stand-in for true nonresponse in every study. Treat differences as clues, not proof.
  3. Examine whether important groups were under-reached. Accessibility barriers, language restrictions and contact methods may systematically exclude groups. Document recruitment limitations rather than declaring the sample representative.
  4. Consider weighting only with adequate justification. Weighting may help align known demographics with credible benchmarks in suitable designs, but it cannot automatically repair self-selection, coverage errors or unmeasured differences. Explain the method and its assumptions if used.
  5. Report what cannot be tested. If no information exists about nonrespondents, say so. Do not invent comparisons or assume the observed sample reflects the intended population.

In a small dissertation, a careful limitations analysis may be more defensible than a complex adjustment unsupported by the sampling design. Nonresponse bias is one part of broader research quality, along with measurement error, sampling coverage and analysis decisions.

Step 6: Write the methods and limitations paragraphs

A strong dissertation report gives the actual counts, method, eligibility decisions, completion rules, recruitment context, and a proportionate interpretation.

Complete worked methods paragraph

This hypothetical cross-sectional questionnaire study sampled 250 students from a verified list of eligible postgraduate enrolments at one institution. Invitations were distributed through an approved institutional email channel during a defined recruitment period. A complete questionnaire was defined in advance as one satisfying the study's prespecified core-item requirements; 155 eligible students submitted responses meeting that definition. The complete-response rate was therefore 62.0% (155/250). Partial submissions, refusals and nonreturns were recorded separately. The survey responses were analysed according to the prespecified missing-data rules; questions with missing answers used their applicable analytic denominators.

Complete worked limitations paragraph

Although 62.0% of the eligible sample submitted complete questionnaires, the achieved response rate does not establish that respondents were representative of every eligible student. Students experiencing heavier workload or lower engagement may have been less inclined to participate, which could affect estimates of the academic behaviours under study. Because comparable outcome information for nonrespondents was unavailable, the direction and magnitude of potential nonresponse bias could not be established. The findings are therefore interpreted as evidence from respondents in the study setting rather than as unbiased estimates for all postgraduate students or other universities.

These passages work because they report actual design features within the hypothetical scenario and distinguish measurable facts from plausible, unverified limitations. In a real dissertation, replace every detail with the procedures and counts actually observed. If you are still writing a proposal, use future tense to describe intended reporting rather than claiming to have collected responses.

Where should response-rate information appear in a dissertation?

Where to report survey participation and response-rate information
Dissertation locationWhat belongs there
Methodology: sampling/recruitmentWho was approached, eligibility, invitation method, field period and planned response rules
Results: sample or participation flowAchieved counts, exclusions, complete/partial breakdown and computed rate
Results: variable analysesItem-specific missingness and analytic sample sizes as relevant
Discussion or limitationsPotential nonresponse bias, coverage limits and effects on interpretation
Appendix (if useful and permitted)An anonymised disposition table or recruitment flow, without exposing identifiable records

Some institutions combine these sections differently. Your department's dissertation handbook and supervisor-approved structure take precedence. For wider reporting advice, our guide to reporting statistical results in a dissertation explains how to present denominators and uncertainty clearly.

Common survey response-rate mistakes

Dividing by website views. Page views can include repeated visits, ineligible viewers and automated traffic. They are rarely an appropriate eligible-sample denominator.

Treating all partial questionnaires as complete. Set a defensible rule before analysing participation and explain what “usable” means. Do not shift the rule after seeing the outcome simply to increase the reported rate.

Removing nonrespondents from the denominator. A refusal or nonreturn is not the same as proven ineligibility. Excluding nonrespondents artificially inflates the reported participation measure.

Claiming that a good rate proves representativeness. Coverage, sampling design and differences between respondents and nonrespondents also matter. AAPOR specifically cautions against judging survey quality from response rate alone.

Using a universal response-rate cutoff. There is no single percentage that automatically makes every dissertation survey valid or invalid. The consequences depend on the research question, design, recruitment and available quality evidence.

Calling a convenience sample random. Random selection is a property of the sampling design; participation percentage does not change it.

Hiding an unknown denominator. If the survey was openly shared without a traceable eligible invitation count, report starts, submissions and limitations rather than manufacturing an invitation-based response rate.

Forgetting missing items. Different analyses may have different usable sample sizes. Show the relevant denominator.

Frequently asked questions

Is a 50% or 60% response rate acceptable for a dissertation?

No universal cutoff decides quality. Report the calculation accurately, assess the sampling method and possible nonresponse bias, and explain what conclusions the evidence supports. Your institution may have practical expectations, but these do not replace methodological justification.

Do I calculate a response rate for qualitative interviews?

You can report the number invited, eligible, declined, and interviewed, with an appropriate participation measure where the denominator is known. However, many qualitative studies use purposive recruitment and seek information-rich accounts rather than statistical representativeness. Explain recruitment adequacy under that design rather than applying a survey threshold mechanically.

Are incomplete questionnaires respondents?

That depends on your declared response definition and any standardized formula used. Distinguish complete questionnaires from partial ones, specify the criteria for qualifying partial interviews, and report both counts transparently.

Can I estimate a response rate from a Google Forms link shared publicly?

Usually not an invitation-based response rate when the number of eligible people who received or saw the invitation is unknown. You may report complete submissions and, if reliably measured, completion among people who started the form.

What if I achieved the intended sample size but had many refusals?

Report both the achieved sample and the recruitment outcomes. Reaching a numeric target does not eliminate the possibility of selection or nonresponse bias. Avoid generalisations that the design cannot support.

Final dissertation response-rate checklist

  • [ ] I identify the eligible target population and sampling frame.
  • [ ] I account for sampled cases, including ineligible, partial, refusal and noncontact outcomes where known.
  • [ ] I distinguish response rate from completion and item response rates.
  • [ ] I show the formula and the correct numerator and denominator.
  • [ ] I disclose how partial responses and unknown eligibility were handled.
  • [ ] I report recruitment method, timing and relevant limitations.
  • [ ] I consider possible nonresponse bias without claiming it is proven or absent.
  • [ ] I avoid treating the percentage as evidence of random or representative sampling.
  • [ ] I accurately identify the analytic sample size for each relevant result.
  • [ ] I follow institutional and disciplinary reporting requirements.

Final thoughts

The strongest response-rate section is not necessarily the one with the largest percentage. It is the one readers can understand and evaluate. Keep an accurate account of who was approached, who was eligible, who responded, which answers were usable, and what participation patterns may mean for interpretation. Transparent reporting makes a dissertation more credible—even when participation is lower than hoped.

If you would like feedback on your research-methods explanations, dissertation structure or presentation of quantitative findings, explore Acadelyra's academic-support services. Support should strengthen your own understanding, decisions and honest reporting—not substitute invented data or methods you did not use.

Authoritative further reading