Quantitative research: what was measured and counted?

Before interpreting a number, ask what it represents. Is it a person, a visit, an answer, a duration, or a score defined by the authors? We can read more precisely by keeping the measurement and its unit beside the result, instead of letting a familiar-looking number explain itself.

Eight visits are not necessarily eight visitors

Consider an invented report about a small community games session. Its results table lists eight table visits during one hour. The notes explain that the same person could return more than once. A reader initially retells this as “eight people joined,” then corrects their note after reading the counting rule.

The number was not the problem: the reader had changed what was being counted. A second table gives the minutes each table was in use. That number answers a different question from the number of visits, and neither table tells us why someone chose a particular game.

A parent and teen work through the example together. They label one column “recorded visits” and another “minutes in use.” They leave “unique people” blank because the fictional report does not provide that count. They do not estimate it from the busiest-looking photograph.

Build a number's identity card

For one result, note:

• Value: which number are you reading?
• Unit: what does one counted or measured item mean?
• Group and period: whose observations, from when?
• Collection rule: how does the report say this was measured?
• Missing information: what is absent or excluded?
• Interpretation: what do the authors conclude, and what do they leave uncertain?

For a percentage, also locate the total it uses. For an average, find what values were combined. If those details are missing, keep the uncertainty in your retelling rather than substituting the interpretation that feels most natural.

A numerical answer still has a question

Use the report's stated purpose to guide your reading. Was it describing a set of observations, comparing groups, or testing a particular proposition? The presence of a chart does not answer that for you. Follow the methods and results rather than assuming the design from the graphic style.

The games-session example contains no research finding about ADHD. It does not show that one game, schedule, or environment improves anything. We are practicing the smaller skill of keeping a result connected to what was actually measured.

Several people can notice different gaps

A sibling might spot repeated visits. A grandparent might ask which hour was observed. An educator might ask how an unclear entry was handled. These are complementary reading questions, not tests of who is good at mathematics. Nobody has to disclose personal records or calculate a relative's performance to use this card.

Source

[1] Creswell JW, Klassen AC, Plano Clark VL, Smith KC. Best Practices for Mixed Methods Research in the Health Sciences. Commissioned by NIH OBSSR; exact publication date not stated in inspected front matter; accessed September 6, 2026. Printed pages 4–5, quantitative-research paragraph, supports [1]. No source effectiveness claims or datasets are reproduced.

Related guides

• Qualitative research
• Mixed methods
• Mean
• Sample size