How to Collect Data for a University Assignment: A Complete Guide

When you start a university assignment that involves research, one of the first questions you may face is: Where am I going to get my data?

It sounds simple, but choosing and collecting the right information can take more thought than expected. You might need to conduct a survey, interview people, observe behaviour, carry out an experiment, or work with information that has already been collected by someone else.

The good news is that you don’t need to make the process unnecessarily complicated. I find it much easier to start with the assignment question and work backwards. Once you know exactly what you are trying to find out, choosing the right type of data becomes much more straightforward.

What Is Data Collection?

Data collection is simply the process of gathering information that you will use to answer a research question.

The type of data you need depends heavily on your subject and the kind of assignment you are writing. For one project, the evidence might be survey responses from students. For another, it could be statistics from the Office for National Statistics, interview transcripts, laboratory measurements or findings from previous academic studies.

There are two broad categories you will come across: primary and secondary data.

Primary data

Primary data is information that you collect yourself for a particular research purpose.

For example, imagine your assignment asks:

How does part-time employment affect the study habits of university students?

You could create a questionnaire and ask students about their working hours, study routines and academic workload. Those responses would become your primary data.

Other examples include:

How to Collect Data for a University Assignment

When you start a university assignment that involves research, one of the first questions you may face is: Where am I going to get my data?

It sounds simple, but choosing and collecting the right information can take more thought than expected. You might need to conduct

The main advantage is that you can design the collection process around your specific research question. The downside is that it can take time to recruit participants, collect responses and organise the results.

Secondary data

Secondary data has already been collected by another person or organisation.

You might use government statistics, academic datasets, company reports, published research or information from an established research organisation.

For example, the Office for National Statistics provides a large amount of UK statistical information that students can use for research.

Don’t assume that secondary data is automatically reliable just because it comes from a published source. You still need to check who collected it, when it was collected, what population it covers and how the information was produced.

Start With Your Research Question

Before you create a questionnaire or start downloading datasets, take a close look at your research question.

A question such as:

What do university students think about social media?

is probably too broad for a focused assignment.

You could narrow it down to something like:

How does daily social-media use relate to perceived academic concentration among undergraduate students?

Now you have a much clearer idea of what you need to investigate.

You may need information about:

Daily social-media use
Perceived concentration
Study habits
Year of study
Course or subject area

This is an important step because it stops you from collecting information simply because it looks interesting. Every piece of data should have a reason for being there.

Think about the type of answer you need

The nature of your question can also help you choose between qualitative and quantitative research.

If you want to measure something, compare groups or identify numerical patterns, quantitative data may be appropriate.

If you want to understand people’s experiences, opinions or reasons for behaving in a particular way, qualitative research may make more sense.

Some assignments benefit from using both. This is known as a mixed-methods approach.

For example, a survey might tell you how many students use generative AI regularly, while interviews could help you understand why they use it.

Decide Whether You Need Primary or Secondary Data

One mistake I often see students make is assuming that they must collect their own data.

That’s not always necessary.

Before spending several days designing a survey, search your university library, reputable databases and official statistical sources. You may discover that someone has already collected information that answers part of your question.

When considering a secondary dataset, ask yourself:

Who collected the information?
Why was it collected?
How old is the data?
Who does it represent?
How large is the dataset?
How were the variables measured?
Are there known limitations?
Can you access the original methodology?

For UK-focused research, the ONS data and statistics resources can be a useful starting point. Depending on your subject, you might also look at government departments, international organisations, research institutes and academic databases.

The important thing is to understand the dataset rather than simply copy figures from it.

Choose the Right Data Collection Method

Once you’ve established what information you need, you can decide how to collect it.

There isn’t one method that works for every assignment.

Surveys and questionnaires

Surveys are useful when you want to gather responses from a relatively large group of people and compare their answers.

For instance, if you were researching students’ attitudes towards online learning, you could create questions using multiple-choice answers or rating scales.

Keep the questionnaire focused. If it takes 20 minutes to complete, people may lose interest before reaching the end.

You should also avoid confusing or leading questions.

Instead of asking:

Don’t you agree that online lectures are more convenient?

a neutral question would be:

How convenient do you find online lectures?

The second version gives participants room to express their actual opinion.

Be particularly careful with questions that ask about two things at once.

For example:

Do you find online lectures convenient and academically effective?

A student could find them convenient but academically ineffective. Combining both ideas makes the response difficult to interpret.

Interviews

Interviews are useful when you want more detailed answers.

Suppose you’re investigating why some students struggle to balance employment and university study. A questionnaire might tell you how many hours students work, but an interview could reveal what actually makes balancing the two difficult.

You could use a structured interview with fixed questions, or a semi-structured interview that gives you a framework while allowing participants to expand on their answers.

For most small student research projects, semi-structured interviews can be a practical option.

If you record interviews, remember that participants should understand what is being recorded and how the recordings and transcripts will be used.

Observation

Sometimes the best way to collect information is to observe what people actually do.

For example, depending on your course, you might observe how people interact with a public exhibition, use a particular learning space or respond to an activity.

Before beginning, decide what you are looking for and how you will record it.

Otherwise, you can end up with pages of notes that contain plenty of detail but very little useful evidence.

Experiments

Assignments in subjects such as science, psychology and engineering may involve experiments or measurements.

Here, consistency matters enormously.

You should know in advance:

What you are measuring
Which equipment or instruments you will use
How measurements will be recorded
How many trials you need
Which conditions must remain consistent

Your methodology should explain these decisions clearly enough for your reader to understand how the data was produced.

Decide Who or What You Will Study

If you’re collecting information from people, you also need to think about your population and sample.

The population is the wider group you are interested in.

Your sample is the smaller group from which you actually collect information.

For example, your research might concern undergraduate students across a university, but your sample could consist of 100 students who complete your questionnaire.

How you select those 100 people matters.

You may come across sampling methods such as:

Random sampling
Stratified sampling
Convenience sampling
Purposive sampling
Snowball sampling

Each approach has advantages and limitations.

Convenience sampling, for example, can be easy for a student to organise because you can recruit people who are readily available. However, those participants may not accurately represent the wider population.

There’s nothing inherently wrong with using a practical sampling method for a small assignment. The important thing is to explain what you did and acknowledge the limitation instead of pretending that your sample represents everyone.

Create a Simple Data Collection Plan

A short plan before you begin can save you a surprising amount of trouble later.

Write down:

Area	What you need to decide
Research question	What exactly am I trying to find out?
Data required	Which information will answer the question?
Participants/source	Where will the data come from?
Method	Survey, interview, observation, experiment or existing dataset?
Sample	Who or what will be included?
Recruitment	How will participants be approached?
Timing	When will collection take place?
Storage	Where will the data be kept?
Analysis	How will I make sense of the results?
Ethics	Do I need consent or ethical approval?

Keep this document while you work.

When you eventually write your methodology section, you’ll already have a record of the decisions you made.

Test Your Questionnaire or Interview First

If you’re collecting primary data, don’t immediately send your questionnaire to 200 people.

Test it first.

Ask a few people to complete the questionnaire and tell you if anything is confusing. You may discover that a question has an unintended meaning or that one of your answer options doesn’t cover a common response.

The same applies to interviews.

Do a practice interview and listen to the answers you receive. If your questions consistently produce one-word responses, they may need changing.

This is where a small pilot study can be extremely useful.

For example, suppose you ask students how many hours they spend studying each week. One person might include lectures, while another might count only independent study.

A clearer definition before the main data collection begins can prevent that inconsistency from affecting your results.

Don’t Forget Research Ethics

Ethics should be considered before you collect data, particularly when human participants are involved.

Check your university’s research-ethics requirements before starting your project. Some types of research may require formal approval.

Participants should generally understand what they are being asked to do and how their information will be used. The UK Research and Innovation (UKRI) guidance on consent explains the importance of informed consent in research.

You also need to think about personal information.

The Information Commissioner’s Office (ICO) provides guidance on using personal information for research and explains issues surrounding data protection, anonymisation and pseudonymisation.

For a university project, that usually means being sensible about what you collect.

Don’t ask participants for their full name, phone number or other identifying information if you don’t actually need it.

You should also make sure that research files are stored securely and handled according to your university’s requirements.

Keep Your Data Organised

Data becomes much harder to work with when you don’t organise it from the beginning.

If you’re using a spreadsheet, decide how you’re going to record your variables and stick to the same format.

For example, if you’re recording responses about gender, don’t use “Male”, “M”, “man” and “1” interchangeably without a defined coding system.

For interviews, give your recordings and transcripts sensible file names. Keep track of which file belongs to which participant without unnecessarily storing identifying information.

If you’re working with a secondary dataset, keep the original documentation and codebook. Documentation helps you understand what the variables actually mean rather than making assumptions based on column headings.

It is also a good idea to keep the original dataset separate from the cleaned version. That way, you can always go back and check what changed.

Check the Data Before You Analyse It

Getting the data is only half the job.

Before analysing it, look for obvious problems.

With quantitative data, check for:

Missing responses
Duplicate entries
Impossible values
Inconsistent categories
Unexpected outliers
Data-entry mistakes

With qualitative data, check that your recordings and transcripts are complete and that your notes accurately reflect what participants said.

Secondary data needs checking too.

Look at the definitions used by the original researcher or organisation. Pay attention to the date, population, geographical coverage and collection method.

For example, a statistic describing UK adults in 2019 cannot automatically be presented as evidence about all UK university students in 2026.

The context behind a number matters just as much as the number itself.

A Worked Example

Let’s say your assignment asks:

What factors influence university students’ use of generative AI for academic work?

You could approach the question in several ways.

First, you might create a questionnaire asking students about:

How frequently they use AI tools
Why they use them
Which academic tasks they use them for
How useful they find them
How confident they are at checking AI-generated information
How familiar they are with university rules on academic integrity

You could then interview a smaller group of students.

The survey would give you comparable numerical information. The interviews could help explain the reasons behind those responses.

You could then compare your findings with published academic research and relevant university policies.

The point isn’t to use every possible research method. It’s to make sure each method has a clear purpose.

Common Data Collection Mistakes

Collecting more data than you need

A large dataset isn’t automatically a good dataset.

If you collect 1,000 responses but most of the questions have nothing to do with your research question, you’ve created extra work without necessarily improving your assignment.

Only asking your friends

Friends and classmates may be easy to reach, but they may not represent the population you’re studying.

If you use them as participants, explain the sampling limitation honestly.

Writing unclear questions

If participants can interpret the same question in several different ways, your results may become difficult to compare.

Read every question from the participant’s perspective rather than assuming they’ll interpret it exactly as you intended.

Ignoring existing information

Don’t spend a week collecting primary data before checking whether a reputable organisation has already produced a useful dataset.

Secondary research can sometimes provide better coverage than a small student survey.

Forgetting to document what you did

Write down important decisions while you’re collecting the data.

Trying to remember your entire research process several weeks later is much harder than keeping a simple research log as you go.

Trusting every online source

Finding information through Google doesn’t make the information academically reliable.

Check the organisation behind the source, publication date, methodology and references. University library resources can also help you distinguish between different types of evidence.

Where Can You Find Reliable Secondary Data?

The best source will depend on your subject.

For UK-related assignments, useful places to begin include the Office for National Statistics, government departments, university libraries and academic databases.

You can also look at organisations such as the OECD and World Bank when your research has an international focus.

Before using a dataset, find out:

Who produced it?
When was it collected?
What was the original purpose?
Who was included?
How was the information collected?
Are there methodological limitations?

Don’t rely solely on a chart or statistic taken from somebody else’s article when you can trace the information back to the original source.

How to Explain Your Data Collection in Your Assignment

When you eventually write the methodology section, your reader should be able to understand what you did and why.

A straightforward structure is:

Research approach: Explain whether your research is qualitative, quantitative or mixed methods.
Research design: Describe the overall approach.
Participants or data source: Explain where the information came from.
Sampling: Describe how participants or records were selected.
Data collection: Explain how the information was gathered.
Procedure: Describe what actually happened.
Ethics: Discuss consent, privacy and ethical approval where relevant.
Limitations: Explain weaknesses that could affect your findings.

Don’t try to make your methodology sound perfect.

Good academic writing usually acknowledges limitations. If you used a relatively small convenience sample, say so. If your dataset covers only a particular period, explain that.

Being transparent about limitations generally makes your research more credible, not less.

If you’re having difficulty turning your research into a properly structured university assignment, academic writing guidance such as assignment help oxford may also be useful for understanding structure, research and presentation. Any support should complement your own research rather than replace it.

Final Checklist

Before you start collecting data, take a few minutes to check the following:

Does my data directly relate to my research question?
Have I chosen a suitable collection method?
Do I know who my target population is?
Is my sample appropriate for the assignment?
Are my questions clear and neutral?
Have I tested my questionnaire or interview questions?
Do I need ethical approval?
Do participants need to give informed consent?
Am I collecting unnecessary personal information?
Do I know how the data will be stored?
Can I explain where secondary data came from?
Have I kept a record of my research process?
Do I know how I will analyse the information?

Final Thoughts

Learning how to collect data for a university assignment is really about making sensible research decisions.

You don’t need the biggest sample, the longest questionnaire or the most complicated methodology. You need evidence that actually helps answer your research question.

Start by defining the question. Then decide what information you need, where it can come from and which collection method makes sense. From there, pay attention to sampling, ethics, data quality and organisation.

If you take those steps before collecting your first response or downloading your first dataset, you’ll have a much clearer path through the rest of the assignment. More importantly, you’ll be able to explain not just what you found, but how you found it and why your evidence deserves to be taken seriously.