Best Online Survey Tools for a South African Dissertation (2026): Free, Institutional and Offline Compared

Tool Cost to a South African postgraduate Offline capture Best suited to Main limitation
Microsoft Forms Included with the Microsoft 365 account most SA universities issue No A straightforward online questionnaire with a modest respondent pool Limited question logic and export options compared with research-grade tools
Google Forms Free with any Google account No Fast, low-friction surveys where respondents have data and a browser No research features — no piping standards, no audit trail, no institutional governance
KoboToolbox (Community Plan) Free — the nonprofit category explicitly includes educational institutions. 5 000 submissions/month, 1 GB file storage, unlimited projects and forms Yes, via its mobile collection app Fieldwork, township and rural data collection, anything where connectivity is unreliable Built for humanitarian and field research, so the interface assumes that context
REDCap Free to the researcher if your institution is a consortium member and hosts an instance Yes, via its mobile app Health, clinical and any study where an audit trail and institutional hosting matter You cannot use it unless your institution runs it; there is no personal sign-up
Qualtrics Free to you if your university holds a licence; otherwise enterprise-priced Yes, via its mobile app Complex designs — randomisation, branching, experimental blocks Entirely dependent on the institutional licence; access typically ends when you graduate
LimeSurvey Cloud Basic 34€/month billed monthly (1 admin user, 1 000 responses/month, 250 MB storage); Expert 29€/month billed annually at 348€ (3 admin users, 10 000 responses/year, 1 GB) Yes, in the offline app for supported plans Multilingual surveys and complex question types when no institutional tool fits You are paying in euro out of a rand budget
LimeSurvey Community Edition Free, open source since 2006; current stable release 7.0.10, installed on your own server Yes A department that already has hosting, or a student with the technical skills You are the system administrator, and a broken server mid-fieldwork is your problem

Prices and limits above were read from each vendor’s own pricing page on 18 August 2026. Two we could not verify and therefore do not state: SurveyMonkey’s current free-tier limits (its pricing page renders almost no readable text to our tools) and Microsoft Forms’ published response ceilings (the Microsoft support page for form limits returned a 404). Check both yourself before relying on either.

Before anything else: what does your university already give you?

The same rule that governs statistical software governs this. Nearly every South African university issues students a Microsoft 365 account, which includes Forms. Many health and science faculties host a REDCap instance through the international consortium, and a licence you already hold is worth more than a marginally better tool you have to buy.

Ask your faculty research office three questions before you evaluate anything on this page: does the institution host REDCap, does it hold a Qualtrics licence, and is there a preferred platform your ethics committee expects to see named in the application. That third question saves the most time, because a committee that has already assessed a platform’s data handling will move faster than one meeting your choice for the first time.

A secured data container inside a bounded territory outline, representing where survey responses are stored and under whose rules
Your ethics committee will ask where the responses live. Have the answer before you apply.

The ranked shortlist

1. Your institution’s own platform — REDCap or Qualtrics, if you have it

Who it suits. Everyone who has access. REDCap was created at Vanderbilt University in 2004 for researchers who needed secure, compliant data collection, and the consortium that formed in 2006 now reports 8 384 institutions across 166 countries, 2,8 million projects and over 56 000 citations. Those are the platform’s own live figures, and the citation count is the useful one: it means your methods section can name a tool an examiner has read about many times.

Where it falls short. There is no route in except through an institution. If your university does not host REDCap you cannot simply sign up, and Qualtrics access almost always ends when your student status does — so export your raw data before you graduate, not after.

2. KoboToolbox — the best free option for South African fieldwork

Who it suits. Anyone collecting data away from reliable connectivity, and anyone who needs a properly featured research tool at no cost. KoboToolbox’s Community Plan is free, and its nonprofit category explicitly covers educational institutions alongside nonprofits, government agencies and UN organisations. The free plan includes 5 000 survey submissions per month, 1 GB of file storage and an unlimited number of projects and forms — comfortably more than a master’s dissertation will use.

Where it falls short. It was built for humanitarian data collection, and the vocabulary and interface reflect that. If your study is a straightforward online questionnaire to a connected professional sample, Microsoft Forms will take you fifteen minutes and Kobo will take you an afternoon.

A device collecting data with a broken link to a distant network node, representing offline survey capture that syncs later
Offline capture is not a luxury feature in South African fieldwork. It is the feature.

3. Microsoft Forms — the default you already own

Who it suits. A conventional online questionnaire distributed by link or email to respondents who have data and a browser. It costs nothing extra, your institution already governs it, and your ethics committee has almost certainly seen it before.

Where it falls short. Question logic is basic, and there is no offline mode. If your design needs randomisation, complex branching or experimental blocks, you will hit the ceiling and have to migrate mid-project, which is worse than choosing correctly at the start.

4. Google Forms — fastest to build, thinnest on research features

Who it suits. Pilots, expert-panel rating sheets and short instruments. It is free with any Google account and exports cleanly to a spreadsheet, which makes it genuinely useful for the content-validity panel described in our guide to proving your questionnaire is valid.

Where it falls short. It is a form builder, not a research platform. There is no institutional governance, no audit trail and no configurable retention. Some ethics committees accept it without comment and some ask questions; find out which yours is before you build.

5. LimeSurvey — the paid option worth the rand

Who it suits. Multilingual surveys and complex question types when no institutional platform fits. LimeSurvey states it is used by more than 250 universities and research institutions, and its cloud plans start at 34€ per month billed monthly for 1 000 responses a month and 250 MB of storage, with the annually billed Expert plan at 348€ a year for 10 000 responses a year and 1 GB.

Where it falls short. You are paying in euro from a rand budget, and the exchange rate is not on your side. The free route is LimeSurvey Community Edition — open source since 2006, current stable release 7.0.10 — but that means running and maintaining your own server for the length of your fieldwork.

The three things that actually decide it in South Africa

Can your respondents afford to answer?

A twelve-page questionnaire with embedded images costs your respondent data to load. That is a real barrier in a country where connectivity is uneven and prepaid data is expensive — the scale of it is set out in our figures on student internet and data access in South Africa. Keep the instrument short, avoid images that carry no measurement value, and make it work on a phone, because that is what most of your respondents will use.

Will you be collecting where there is no signal?

If any part of your fieldwork happens away from reliable connectivity, offline capture stops being a nice-to-have. Only tools with a dedicated mobile app that stores locally and syncs later will work — KoboToolbox, REDCap and Qualtrics among the options here. A browser-based form is unusable the moment the signal drops, and a half-completed response is a lost respondent you probably cannot re-recruit.

Where will the responses be stored, and who else can see them?

Your ethics committee will ask, and the answer belongs in the application rather than in an email afterwards. Know which country the servers are in, who at your institution can access the account, how long responses are retained and how they will be destroyed. The full set of obligations, including gatekeeper letters and retention, is covered in our guide to ethics clearance at a South African university — this is one paragraph of that, and it is the paragraph that decides your platform.

The recommendation

Use your institution’s platform if it has one. If it does not, use KoboToolbox.

That holds for almost every South African dissertation. An institutional REDCap or Qualtrics account is free to you, already assessed by your ethics committee and hosted under rules your university controls. Where neither exists, KoboToolbox’s Community Plan gives an educational user a research-grade tool with offline capture at no cost, with a submission allowance no dissertation will exhaust.

Pay for LimeSurvey only when you have a specific requirement — a language combination or a question type — that the free options genuinely cannot meet, and you have confirmed that with a test form rather than assumed it.

Two decisions to make before you build the form

  1. How many responses do you need? Build the instrument after you know the number, not before — the routes to a defensible target are in our answer to how many participants you need for a dissertation. A tool’s response cap only matters once you know what you are asking of it.
  2. How will the export reach your analysis? Every tool here exports to CSV or Excel; what differs is how the variable names, value labels and missing data arrive. Export ten test responses and import them into your analysis package before you go live, following the codebook discipline in our guide to analysing questionnaire data in SPSS. Discovering a messy export after 200 responses is a week you do not have.

From responses to a chapter

The gap where dissertations lose time is between a clean export and a written methodology — the codebook that lived in your head, the field-visit dates in a WhatsApp thread, the reason three responses were excluded. Tesify keeps the methodology chapter and its sources together while the fieldwork is happening, so what you write in October matches what you actually did in June. The data and the decisions stay yours.

Write your methodology chapter in Tesify

Frequently asked questions

What is the best free survey tool for a South African dissertation?

KoboToolbox’s Community Plan, which is free for educational institutions and includes 5 000 submissions a month, 1 GB of storage, unlimited forms and offline mobile capture. If your university hosts REDCap or licenses Qualtrics, that is better still because it is already governed by your institution.

Is Google Forms acceptable for academic research?

Many supervisors and committees accept it for straightforward, low-risk questionnaires, and it exports cleanly. It offers no institutional governance, audit trail or configurable retention, so confirm with your ethics committee rather than assuming.

Can I use Microsoft Forms for my dissertation?

Yes, and it is free with the Microsoft 365 account most South African universities issue. Its constraints are limited question logic and no offline capture. We could not verify its current published response limits, because the Microsoft support page for form limits returned a 404 — check it before committing to a large sample.

Does KoboToolbox really cost nothing for students?

KoboToolbox’s own pricing page places educational institutions in its nonprofit category, for which the Community Plan is free with 5 000 submissions a month and 1 GB of file storage. Confirm your eligibility when you register.

How do I collect survey data offline?

Use a tool with a dedicated mobile app that stores responses on the device and syncs when connectivity returns — KoboToolbox, REDCap and Qualtrics all offer this. A browser-based form cannot do it.

Can I get REDCap as an individual student?

No. REDCap is distributed to institutions through its consortium, so access comes through a university that hosts an instance. Ask your faculty research office whether yours does.

What does LimeSurvey cost?

LimeSurvey Cloud’s Basic plan is 34€ a month billed monthly for 1 admin user, 1 000 responses a month and 250 MB of storage; the Expert plan is 348€ billed annually for 3 admin users, 10 000 responses a year and 1 GB. LimeSurvey Community Edition is free open-source software you install on your own server.

Does my ethics committee care which tool I use?

It cares where the data is stored, who can access it, how long it is kept and how it is destroyed. The tool matters because it determines those answers. Name the platform in the application and answer those four questions explicitly.

Should I worry about respondents’ data costs?

Yes. A long questionnaire with heavy images costs a respondent money to complete, which depresses your response rate in exactly the populations that are hardest to reach. Keep it short and make it work on a phone.

Can I switch survey tools halfway through data collection?

You can, but you will have two datasets with different structures and a change in the instrument’s presentation that you must report. Run a ten-response test export before you go live so that you do not have to.

Do I need a paid tool for a master’s dissertation?

Almost never. Between an institutional platform and KoboToolbox’s free plan, the great majority of South African master’s projects are fully served without spending anything.

How many test responses should I collect before going live?

At least ten, from people who are not your respondents, and take them all the way through export and import into your analysis package. The test is not whether the form works; it is whether the export does.