Best Practices to Follow When Writing the Results Section of a Quantitative Research Paper

Compared to the other parts of a quantitative research paper, the results section is relatively straightforward. While the previous sections need to convince the reader of something, the results do not.

Specifically:

• The introduction must convince readers that the research question matters.

• The methodology must convince readers that the present study is suitable for answering the research question.

• The results simply provide the evidence needed to answer the research question.

Still, if you fail to follow the best practices when writing your results, readers might have a hard time seeing how the data answer the research question.

In turn, journal reviewers may fail to see the value of your study and might reject your paper.

This article will guide you through the process of writing a results section for a quantitative research paper, which includes four main steps:

1. Remind the reader of the statistical analyses you conducted.

2. Present the results of the statistical analyses.

3. Tell readers what patterns in the data they should notice.

4. Repeat the previous steps as needed.

In addition, this article will help you avoid the most common mistakes authors make when writing this section:

• Don’t merely repeat the information in your figures and tables.

• Don’t provide more information than the reader needs. Specifically:

  • Only report results that are directly related to at least one research objective/question.

  • Refrain from interpreting or analyzing the results.

  • Write concise sentences.

Best Practices to Follow

1. Start by reminding the reader of the statistical analyses you ran.

In most papers, the results section begins with a quick description of the statistical tests used to analyze the data.

For example:

A one-way ANOVA was conducted to investigate whether there were any significant performance differences among the frameworks.

Even though this information will have already been provided in the methods section, it is standard practice to give readers a reminder.

2. Present the results of the statistical analyses.

Usually, authors begin describing a result by simply stating whether the data indicate a significant difference or effect, as in this example:

The results reveal a significant difference between students’ financial literature knowledge based on the type of high school they attended, F(5,513) = 9.78, p < .01.

Sometimes, authors will combine Steps 1 and 2 into one sentence, particularly when the statistical analysis is well-known and straightforward.

For example, here is the first sentence of the results section of a published paper:

A preliminary ANOVA revealed no effect of the two between-participant factors of stimulus set, F(1,17) = 0.06, p = .809, participant gender, F(1,17) = 0.79, p = .386, or their interaction, F(1,17) = 1.37, p = 0.258, on the looking preference for the smiling expression.

3. Highlight important patterns in the data.

When the data point to something important (e.g., a significant difference between groups), it is not enough to simply relay the data to the reader. You should also explicitly state how the data provide evidence that helped you answer the research question.

For example, if an independent variable significantly affected a dependent variable, was the effect positive or negative? If two groups had significantly different scores on a test, which group scored higher?

Going back to the study on how long infants gaze at faces, the authors present their next result as follows:

A repeated-measures ANOVA on the looking preference for the smiling expression further revealed a significant effect of face gender, F(1,20) = 16.68, p < .001.

Although this information addresses the authors’ objective, readers will want to know which face gender was associated with a preference for smiling.

While readers could figure this out for themselves by looking at any relevant tables or figures, it is the author’s responsibility to make sure their readers extract all important takeaways from the data.

Below is the full description of the result on smiling face preferences:

A repeated-measures ANOVA on the looking preference for the smiling expression further revealed a significant effect of face gender, F(1,20) = 16.68, p < .001. Infants looked longer at the smiling female face versus the neutral female face, t(23) = 2.16, p = .041, Cohen’s d = 0.44, but longer at the neutral male face versus the smiling male face, t(21) = -2.27, p = .034, Cohen’s d = -0.48.

4. Repeat.

The three steps above can then be repeated for each result. Or, if the same statistical test was used to analyze all the data, only Steps 2 and 3 will be repeated.

Common Pitfalls to Avoid

Don’t merely say what’s in your tables and figures.

To clarify, it is perfectly acceptable to repeat data from your tables in the main text. However, this shouldn’t be the only job the text is doing.

A poorly written results section might contain a paragraph like this:

As shown in Table 1, in the untreated blood samples, hemolysis was reported at 16.28%, and eryptosis was 10.05%. In the blood samples with urea, the hemolysis level was 8.74%, and eryptosis was 6.89% (both ps < 0.001). Meanwhile, in samples with sucrose added, hemolysis was reported at 7.87% (p < 0.001), while eryptosis was 10.01% (p = .789).

The problem with the above passage is that it only describes the data. But the text in the results section must tell readers what about the data is worth their attention.

Of course, the reader could determine which results are statistically significant and the directions of the effects based on the information the author has presented. However, notice how much easier it is to get that information from the revised passage below:

As shown in Table 1, compared to untreated blood samples, those with urea had significantly lower levels of both hemolysis (16.28% vs. 8.74%, p < .001) and eryptosis (10.05% vs. 8.74%, p < .001). The addition of sucrose also significantly decreased hemolysis levels to 7.87% (p < .001). However, the eryptosis level of 10.01% in the sucrose condition did not indicate a significant effect (p = .789).

A good way to test whether you’ve described your results effectively is to remove the numerical data and see if the outcomes are still understandable:

As shown in Table 1, compared to untreated blood samples, those with urea had significantly lower levels of both hemolysis and eryptosis. The addition of sucrose also significantly decreased hemolysis levels. However, the eryptosis level in the sucrose condition did not indicate a significant effect.

But if you try to do this with the original passage, you’ll end up with something meaningless, like:

As shown in Table 1, in the untreated blood samples, hemolysis and eryptosis were reported. In the blood samples with urea, there was a hemolysis level and an eryptosis level. Meanwhile, in samples with sucrose added, hemolysis was reported, as was eryptosis.

Don’t provide readers with more information than they need.

Authors commonly make three mistakes that bloat their results section:

1. They report results that don’t answer any research questions.

2. They interpret and analyze their results.

3. They write overly long sentences.

You can avoid these issues by following the tips below.

Only report results that answer your research questions.

When editing papers for my clients, I sometimes see sentences that answer questions that weren’t asked in the introduction. For example, an author might provide an in-depth description of participants’ demographic details, even though the study didn’t aim to evaluate demographic differences.

In such cases, there’s nothing wrong with providing the participants’ demographics in a table, and you might make a note of anything that could impact the generalizability of your findings.

But, in general, if the data aren’t crucial to answering any research questions, you shouldn’t describe these data in the text.

Don’t interpret the results.

Remember, the results section is supposed to (1) show the reader the data you collected and (2) identify important patterns in the data.

The results section is not where you interpret or analyze the data. This is the job of the discussion section.

To avoid this problem, check your results section for phrases like “it appears that,” “this seems to indicate,” or “this finding suggests that.” If you see any of these, it is worth taking another look and asking whether the sentence is in the right section.

Be concise.

This tip isn’t as vital as the others; no journal reviewer would reject a paper just because the occasional sentence contains an extra word or two.

However, a results section reads better when the author presents the facts in simple terms, as this helps the reader focus on what’s important.

For example, compare the two passages below:

1. An independent-samples t-test was conducted to examine whether there were any statistically significant differences between participants who completed the mindfulness intervention and participants who did not with regard to their self-reported levels of stress. The results of the analysis indicate that there was a statistically significant difference between the intervention group and the control group, t(98) = 2.84, p = .006. Specifically, compared to participants in the control group, participants in the mindfulness intervention group reported lower levels of stress.

2. The results of an independent-samples t-test revealed a statistically significant difference between the groups, with participants who completed the mindfulness intervention reporting lower levels of stress than the control group, t(98) = 2.84, p = .006.

As the example shows, removing repetitions and redundancies helps readers see what really matters.

Revisions for brevity, as illustrated in the above example, are included in Magnum Academic Editing’s Full Language Editing Package. If you’d like to work one-on-one with a language expert to improve your paper, click here for details.

Summary

Compared to the other sections of a research paper, the results section is straightforward. In most papers, the steps involved in writing this section are:

1. Reminding the reader of the statistical analyses you conducted.

2. Presenting the results of the statistical analyses.

3. Telling readers what patterns in the data they should notice.

4. Repeating the previous steps as needed.

Also, when writing this section, be conscious of these common mistakes:

• Don’t merely repeat information that you’ve presented in your figures and tables.

• Don’t provide more information than the reader needs. Specifically:

  • Only report results that are directly related to at least one research objective/question.

  • Refrain from interpreting or analyzing the results.

  • Write concise sentences.

Still unsure if your results section is strong enough?

I'd be happy to work with you one-on-one to ensure that every section of your paper clearly communicates the value of your work. Click here to learn more.