Fundamental Rules for Reliable Research

A Research Question Is A
Constraint

The first rule of reliable research is simple: the question must
constrain the work.

Without a question, research becomes accumulation. The researcher
collects articles, interviews, numbers, examples, quotations, and
opinions until the project feels substantial. It may still be shapeless.
A pile of material is not an answer.

A good question tells the researcher what belongs in the study and
what does not. It clarifies the object of attention. It also creates
discipline. If the question asks how small firms manage cash pressure,
the project cannot wander into every interesting idea about
entrepreneurship. If the question asks how students experience online
feedback, the project cannot become a general essay on technology.

The question is not a heading. It is a boundary.

Rule One: Define The Claim

Every research project should be able to state the claim it is
testing, developing, or challenging.

Sometimes the claim is explicit: a policy improves access, a method
predicts performance, a programme changes behaviour. Sometimes it is
exploratory: a group of participants experience a problem in ways that
are not yet well understood. In both cases, the researcher must know
what kind of knowledge is being produced.

This prevents the common error of writing around a topic instead of
arguing through it. A topic says, “research skills.” A claim says,
“research skills improve decision quality because they force evidence,
assumptions, and uncertainty into view.”

The second sentence is easier to challenge. That is why it is
better.

Rule Two:
Separate Evidence From Interpretation

Evidence is what the researcher has. Interpretation is what the
researcher makes of it.

The distinction matters. A survey result, interview transcript,
document, observation, or dataset does not speak for itself. The
researcher selects it, organises it, codes it, compares it, and explains
it. That work must be visible enough for a reader to follow.

Weak research hides interpretation inside confident language. Strong
research shows the path. It says what was found, how it was read, what
alternatives were considered, and why the interpretation is
reasonable.

This does not make research neutral. It makes it accountable.

Rule Three: Let Method
Match The Question

Method should not be chosen because it sounds impressive or because
the researcher is comfortable with it. Method must fit the question.

If the question requires meaning, experience, language, or identity,
qualitative methods may be appropriate. If the question requires scale,
frequency, association, or comparison, quantitative methods may help. If
the question asks how something changes over time, the design must make
time visible.

There is no honour in using a complex method to answer a simple
question badly. There is also no honour in using a simple method to
avoid the difficulty of the real question.

Reliable research is not method worship. It is method fit.

Rule Four: Write Down
Assumptions

Every research project rests on assumptions. Some are theoretical.
Some are practical. Some are hidden in the language of the field.

Writing them down improves the work. It allows the researcher to ask
what would change if an assumption were false. It also gives the reader
a fairer view of the argument.

For example, a study of workplace learning may assume that managers
can observe capability accurately. A study of entrepreneurship may
assume that survival is the best measure of success. A study of
education technology may assume that access to a platform is the same as
access to learning.

Each assumption may be reasonable. Each may also be incomplete.
Research becomes stronger when those limits are named rather than
buried.

Rule Five: Keep
Uncertainty In The Answer

The purpose of research is not to remove all uncertainty. It is to
reduce confusion honestly.

A useful conclusion should say what the evidence supports, what it
does not support, and what remains unresolved. This is not weakness. It
is the difference between knowledge and performance.

Decision-makers often want certainty. Researchers should offer
clarity. Clarity may include a strong recommendation, but it should not
pretend that the world has become simpler than it is.

Reliable research leaves the reader better informed and less easily
seduced by false confidence.

The Discipline

Research is a discipline of restraint. It asks the researcher to
resist the attractive answer, the convenient source, the dramatic
conclusion, and the paragraph that sounds clever but carries no
evidence.

The five rules are practical: constrain the question, define the
claim, separate evidence from interpretation, match method to question,
write down assumptions, and keep uncertainty in the answer.

That is how research earns trust. Not by sounding academic, but by
making its path visible enough that another serious person can inspect
it.