Data Collection Methods: Types, Tools and How to Choose
Published 28 Jul 2025 · Updated 9 Oct 2026
Data collection methods are the ways you gather information to answer a research question. Primary methods, such as surveys, interviews and observation, produce new data. Secondary methods reuse data that others have already collected. This guide compares the main options, shows how to choose one and covers ethics.
Get feedback on your research plan
What data collection involves
Data collection is the planned gathering of evidence for a research question. Your methodology chapter must explain what you collected, from whom, how, and why that approach suits your question. Markers usually look for three things.
- A method that matches the research question
- A clear account of who took part and how you reached them
- Honest limits, such as a small sample or a low response
Types of data collection
Data comes in two main forms, and many projects use both.
Qualitative data
Qualitative data describes meanings, experiences and opinions in words. It answers 'how' and 'why' questions, for example why nurses find night shifts tiring. It usually comes from interviews, focus groups, open survey questions and observation. You analyse it by looking for themes.
Quantitative data
Quantitative data is numerical. It answers 'how many' and 'how much' questions, such as how satisfied students are with feedback on a scale of one to five. You analyse it with statistics. If the analysis worries you, see our statistics assignment help.
Mixed methods
Mixed methods combine both. A survey can show how common a problem is, and follow-up interviews can explain why. Say in your methodology why you need both, not just that you used both.
Methods of collecting data
Primary data is new data you collect yourself. Secondary data already exists, such as official statistics. The table compares the main choices.
| Method | Data | Best for | Main risk |
|---|---|---|---|
| Survey | Mostly numbers | Many people, quickly | Low or biased response |
| Interview | Words | Detailed personal views | Slow to run and transcribe |
| Focus group | Words | Shared opinions | Loud voices dominate |
| Observation | Either | What people actually do | Your presence changes behaviour |
| Experiment | Numbers | Cause and effect | Hard to control every factor |
| Secondary data | Either | Large samples at low cost | May not fit your question |
Surveys and questionnaires
Keep questions short, neutral and in a logical order. Use closed questions, such as rating scales, for numbers, and one or two open questions for detail. Test the survey on a few people before you send it, because a confusing question cannot be fixed afterwards.
Interviews and focus groups
Structured interviews ask everyone the same questions. Semi-structured interviews use a guide but let you follow up interesting points, so they suit most student projects. A focus group is a guided discussion with a small group. Record with permission and plan time to transcribe.
Observation and experiments
Observation records behaviour in a real setting, such as how shoppers move through a store. Experiments change one factor and measure the result, ideally against a comparison group. Both need a record sheet agreed before you start.
Secondary data
Sources include official statistics, company reports and archived survey data. The UK Data Service says most of its data are available for teaching, and most can be accessed free with a university or college login. Each data set has its own conditions. Check who collected the data, when and how before you rely on it.
Data collection techniques
A technique is how you carry out a method well. Three habits improve almost any project.
Sampling
Decide who you need and how you will reach them. A small, self-selected sample, such as friends who answer a message, cannot speak for a whole group. If your sample is small, say so and present your findings as examples, not proof.
Piloting
Try your survey or interview guide with two or three people first. Note the questions they misread, how long it takes and what you forgot to ask. Then change the tool before the real round.
Recording
Keep a log of dates, wording, changes and problems. Another researcher should be able to repeat what you did from your notes. The log also gives you material for your limitations section.
Tools of data collection
Choose the simplest tool that does the job and that your university allows.
Paper and offline tools
Printed questionnaires, observation checklists, field notebooks and audio recorders work well where internet access is poor or participants prefer paper. Type up or scan your records soon after collecting them.
Spreadsheets and statistics software
A spreadsheet is enough to store and clean survey answers for most projects. Larger data sets may need statistics software that your university provides. Keep a copy of the raw data and work on a duplicate.
Digital data collection methods
Digital methods reach people who cannot meet you in person, but they raise extra data protection questions.
Online surveys
Online survey platforms let you share a link, make questions required and export the answers. Use only a platform your university approves, and check where it stores responses before you collect any personal data.
Video interviews and transcription
Video calls let you interview people in other places and record with their consent. Automatic transcription saves time, but it makes mistakes with accents and technical words. Read the transcript against the recording and correct it.
Social media and web data
Public posts are not automatically free to use. Ask your supervisor and your ethics committee before collecting social media data, and do not name or quote people who could be identified.
How to choose data collection methods
Start with the research question, then match the method to it. This invented example shows the reasoning.
| Step | Example |
|---|---|
| Research question | How do first-year students use feedback on their assignments? |
| What it needs | Numbers on how many use feedback, and reasons why |
| Methods chosen | A short online survey, then four semi-structured interviews |
| Why | The survey shows how common each habit is; the interviews explain why |
| Main risk | Only keen students reply, so results may look too positive |
| Fix | Send two reminders, ask tutors to share the link, and state the limit |
- Which type of data does the question need?
- Who can answer it, and how will you reach them?
- How much time, money and skill do you have?
- What ethical approval does the method need?
- Has your supervisor agreed it?
Match your question wording to a method
The first words of your research question often point to the method. Use this as a quick check, not a rule.
| Question starts with | Likely method | Example question |
|---|---|---|
| How many or how much | Survey or secondary data | How many students work part-time? |
| Why or how do people | Interviews or focus groups | Why do nurses leave after one year? |
| Does X change Y | Experiment or comparison | Does a reminder text raise attendance? |
| What do people do | Observation | How do shoppers use self-checkouts? |
Ethics and data protection
If you collect data from people, your university will usually expect ethical review before you start, so ask your supervisor what yours requires. The ESRC's research ethics principles include minimising risk and harm and voluntary, informed participation. UK data protection law has provisions for research, which the Information Commissioner's Office (ICO) explains. Its guidance is under review, so check the current version. Our research ethics guide covers the basics.
Consent and withdrawal
Give participants an information sheet in plain English: what the project is about, what you will ask, how long it takes and how you will use the answers. Say they can stop at any time. Keep a record of consent.
Storing and sharing data
Collect only what you need. The ICO lists data minimisation and pseudonymisation, such as replacing names with codes, among research safeguards. Store files in university-approved storage and delete them when your university's policy allows.
Common mistakes
Avoid these errors when you plan and write up your methods.
- Choosing a method before writing the research question
- Asking leading questions that point to the answer you want
- Skipping the pilot
- Collecting more data than you can analyse
- Not recording how you collected the data
- Writing the methodology as a list, with no reasons
Writing up your methods
Your methodology chapter should name each method, give your reasons and describe your participants, tools, ethics steps and limits. Our dissertation structure guide shows where it sits, and our literature review guide shows how to link your findings back to existing research.
- Methods and the reasons for each
- Participants and how you chose them
- Tools and any pilot you ran
- Ethics steps and data protection
- Limits and what you would change
Need feedback on your research plan?
Our specialists give written feedback on your choice of method, survey questions or interview guide, and you write the final version yourself. See how it works, or ask about dissertation proposal help.
Related guides
- Dissertation structure
- Research ethics basics for students
- Dissertation proposal help
- What is a literature review
Sources
- UK Data Service: learning and teaching resources
- UK Data Service blog: Not the second choice: supporting students to do secondary analysis in their dissertations
- Information Commissioner's Office: the research provisions
- UKRI ESRC: framework for research ethics, our core principles
Frequently asked questions
What are the main data collection methods?
Surveys, interviews, focus groups, observation and experiments collect primary data. Existing sources, such as official statistics, give secondary data. Many dissertations combine two methods.
What is the difference between primary and secondary data?
Primary data is new data you collect for your own project. Secondary data was collected by someone else for another purpose, and you reuse it.
Which method is best for a dissertation?
The one that answers your research question with the time and access you have. Your supervisor can confirm whether your plan is realistic.
Do I need ethical approval to collect data?
If you collect data from people, usually yes. Rules differ, so ask your supervisor and read your university's process before you contact anyone.
How many interviews or survey responses do I need?
There is no fixed number. It depends on your question, method and course. Agree a realistic target with your supervisor and explain it in your methodology.
What is the difference between a survey and a questionnaire?
A survey is the whole process of asking a group of people questions. A questionnaire is the set of questions you use. Many students use the two words to mean the same thing.
Can I use secondary data only?
Often yes, if it answers your research question and your course allows it. Explain why you chose it and how you checked its quality.