Exploratory Research: Turn Early Evidence into Open Questions

Exploratory research helps you make sense of a topic when the key questions or ideas are still unclear. It can show what people mean by a term, suggest reasons worth looking into, or help you decide what a later study should ask. The result is an account of what you have learned and what still needs checking.

Start by asking what is unknown, not by choosing the answer you want to find. You can explore through reading, interviews, observation, or numerical data. What matters is how those sources help you refine the inquiry and how honestly you describe their limits.

What exploratory research can answer

You might know that people rarely use a service without knowing how they decide whether to use it. Exploration can help identify experiences, terms, or practical issues worth studying. The Open University's methods guidance distinguishes learning about a little-known issue from describing it in detail or testing explanations about it.

These goals can overlap. An early study might describe a process while raising new questions about it. KPU's research methods text treats exploration, description, and explanation as goals that can support one another. A label does not decide what a particular claim can support.

In clinical research, the MRCT Center glossary describes early work that informs future studies. Its illustration below links initial information about people, samples, and observation to a later study. The health setting is an example of that relationship, not a required method for every exploratory project.

The MRCT Center's illustration shows early inquiry informing a later study. It is reproduced unchanged under CC BY 4.0.

Choose methods for the unknown

Choose sources for the kind of uncertainty you need to resolve. Interviews can help you understand how people describe an experience. Records may show what was logged. Neither is a complete view by itself. Your question should make clear which account you need and what would remain unseen.

Read before you collect

Read past studies, policies that affect the project, and records you can access. A topic may be new to you without being new to the field. Use a bounded literature review process to find existing concepts and conflicting findings before declaring a gap.

Chris Drew's teaching guide emphasizes refining concepts and identifying future research priorities. Treat those as possible outputs. Do not assume that every project needs to end with a testable hypothesis; a clearer description of a process may be the useful next step.

Match collection to the question

Interviews let people explain what happened to them in depth. A focus group can show how people discuss and challenge shared ideas, though some may hold back views in front of others.

Watching a setting can show what happens while leaving motives unclear. A survey can describe what people report, if its questions and the way you select respondents fit the purpose.

Tegan George's methods guide includes both direct collection and existing sources. The distinction matters for access and ethics: reading a published paper differs from asking people for personal accounts. Plan permissions, consent, recruitment, and analysis with the relevant research supervisor or ethics process.

Exploration can use numbers as well as words. The Open University includes surveys among early ways to gather information.

The data format alone does not make a study exploratory, explanatory, or confirmatory. Explain what you are trying to learn and how each source can help.

Worked example of an open question

Imagine a fictional campus team asking why evening students seldom use its shuttle-information page. The team has a draft project note and hopes to hear from students. It also plans to read past transport studies. None of this work has yet shown the reason for low use.

The first question is: “Why does confusing wording stop students using the page?” It assumes that wording is confusing and causes the problem. A more open question is: “How do evening students look for shuttle information, and what do they encounter along the way?”

Use the following fictional plan to link each question to what prompted it. The note keeps other possible reasons and gaps visible, so the team can check what it needs to learn before choosing a method.

Table 1: These rows describe a fictional planning exercise, not observed transport findings or completed interviews.

Keep more than one explanation

Low visits might mean students get clear answers elsewhere. They might not know about the page, struggle to open it on a phone, or have no need for the service. There could also be a flaw in how visits were counted. Keep these as ideas to check. Let students describe routes that work as well as ones that fail.

This is the difference between a concern and a conclusion. A project note can raise a useful concern without showing its cause.

When you synthesize research papers, keep each study's setting and source contribution visible rather than blending them into a single explanation for your project.

Record why the question changes

Suppose early reading shows that students often use a separate app. You may shift from page wording to how they choose where to look. Record the source that prompted the change and when you made it. Check whether it changes whom you need to hear from or the consent you need to collect data.

The Center for Open Science distinguishes planned work from ideas and analyses that arise later. You can plan exploratory work and still change questions as you learn. Make those changes visible rather than presenting a later choice as a test specified from the start.

Trace sources to questions in Atlas

Use Atlas to compare the background papers, project context, and draft questions you are allowed to upload. Select the actual sources you need; keep personal interview material outside the project unless your research permissions allow that use. The goal here is a checked rationale note for your own study planning.

1. Open the project with the relevant sources and wait for them to finish processing.

2. In Ask a question, type @ and select the background papers, context document, and draft questions.

3. Ask: “Compare these sources with my draft exploratory question. Separate established findings, possible explanations, and missing evidence. For each proposed question, cite the passage that motivates it and state its limit.”

4. Open each citation and read the surrounding passage. Check the study setting, people, evidence source, and whether it supports the proposed interpretation.

5. Revise questions that assume an answer or extend beyond the source. Ask the response to separate the contribution from each paper when claims are hard to trace.

6. Select New → Note and save the corrected rationale with source links and open questions. Wait for Saved before closing it.

Repair a source leap

In the fictional project, an answer might say a past paper proves confusing wording caused low page use at this campus. Check where the paper was conducted and what it actually measured. If it concerns another setting, correct the note to say it motivates asking about wording; it does not establish the local cause.

Keep the open question in the note, with its source and other ideas still worth checking. Mark staff accounts as staff accounts and planned interviews as planned. Do not let a cited sentence turn material you hope to gather into material you already have.

Keep early findings in their scope

An interview account can be useful without showing how common an experience is. A convenience sample may miss people who do not use the service or cannot join the study.

If you need a population rate, review a design that can estimate it. If you need an effect, review how the study can address cause and alternative explanations.

KPU's methods guidance links the purpose of inquiry to methods and ethical choices.

Flexibility should help you follow relevant evidence. It does not remove the need to document sampling, changes, missing material, and limits on what the findings mean.

Finding a pattern and testing it with the same data differ from a test planned in advance. COS guidance explains why reports should distinguish these stages. A later test needs a sound design and analysis plan. More of the same data does not, by itself, confirm the claim.

Decide what the next study needs

Finish with a question you can explain, a record of what motivates it, and the unknowns it leaves open. Turn the inquiry into bounded research objectives only after checking that the scope, access, and proposed sources fit what you need to learn.

Take the note to your supervisor or research team. Decide whether the next step is more source reading, ethical approval for new collection, a clearer measure, or a separate test of a proposed explanation. Keep that decision distinct from what the exploratory work has already established.