Writing a Research Article: From Data Analysis to a Complete Report

1. The Overall Structure of a Research Article

A typical quantitative research article has six main sections. Each serves a distinct purpose and contains specific material, but all must connect closely to form a coherent account:

2. Introduction

The Introduction is the article's entry point. It convinces readers that your topic matters and is worth reading. According to Daniels & Minot (2023), a good Introduction follows four sequential steps.

"As scientific knowledge continues to grow exponentially, new technological developments emerge every day... questions are being raised about how it can be effectively used in various fields and what impact it will have." (an opening example from research on ChatGPT in education)

"Despite the potential benefits of ChatGPT as an educational tool, the full extent of its impact on education remains uncertain and requires further investigation... there is a scarcity of literature that delves into the potential educational use of ChatGPT, a novel tool in this domain." (an example of identifying a research gap)

3. Literature Review

The Literature Review synthesizes the most important research relevant to the research question. It should not list individual studies as an annotated bibliography would.

3.1. Organizing the Literature Review

Organize the review around themes or main arguments rather than giving each article its own paragraph. Weave different sources into the same paragraph according to their common features.

A common structure for a good Literature Review is:

1. A paragraph introducing foundational concepts or theories.
2. A paragraph presenting evidence supporting position A and the relevant studies.
3. A paragraph presenting contrary evidence or an opposing perspective.
4. A concluding paragraph identifying the gap your study will fill.

3.2. How many sources are enough?

There is no fixed number. Daniels & Minot emphasize that a clearer research question makes the scope of the literature easier to define. Prioritize peer-reviewed and widely cited sources. Recent articles with few citations should still be considered when directly relevant to your question.

3.3. When should you quote directly?

Use a direct quotation only when the original wording is so distinctive or precise that paraphrasing would lose its meaning. In most cases, paraphrase and cite the source.

4. Theory, Data & Methods

This is the article's most important section because it enables readers to assess the validity of the results. Provide enough detail for another researcher to reproduce the study.

4.1. Information to provide about the data

• When and where the data were collected.
• The organization or unit that collected the data.
• The sampling method and sample size, which reviewers examine closely.
• Data limitations or problems.
• Data adjustments and sampling weights, if any.

4.2. Information to provide about the methods

• Type of analysis: regression, descriptive statistics, t-test, ANOVA, SEM, and so forth.
• Expected results based on theory or hypotheses.
• Variable measurement: how each variable was measured and which scales were used.

5. Results

The Results section presents analyses relevant to the research question in a logical sequence. Interpretation and commentary belong in the Discussion.

5.1. Logical order in the Results section

Results usually follow the order of the questions or hypotheses stated in the Introduction. General results typically come first, followed by detailed analyses of each research question. For example, an article with two research questions should also have two Results subsections, each answering one question.

5.2. Using tables and figures

• Tables and figures are numbered separately and consecutively.
• Always capitalize Table or Figure when referring to them in the text.
• Each table or figure needs a complete title and notes so that it can stand independently of the main text.
• Avoid repeating table values in the text; describe trends or highlights instead.
• Use references such as "... (see Table 3)" or "The data in Figure 1 show..."

5.3. Reporting statistical results in APA style

How statistical test results are reported depends on the audience. Academic journals require full APA reporting. Basic APA rules include:

• Report descriptive statistics: M (mean) and SD (standard deviation).
• Include the test statistic, degrees of freedom (df), and p-value.
• APA generally reports inferential statistics such as t, F and χ² to two decimal places; report exact p values to two or three decimals as precision requires. Omit the leading zero for p and r.
• Italicize all statistical symbols: N, n, M, SD, p, t, F, r, and so forth.
• If p < .001, write p < .001; do not round it to p = .00.

The table below summarizes APA notation for each test:

5.4. Active voice and personal pronouns

Daniels & Minot recommend using active voice where possible. Compare the passive "It was found that students who listen to Beethoven earn higher scores" with the preferable active form: "Our results show that students who listen to Beethoven earn higher scores."

Regarding personal pronouns, academic writing often uses "we" rather than "I," even with a single author. This is called the collective we, inviting readers to observe the results alongside the author.

6. Discussion

The Discussion explains the meaning of the results and places them within the existing literature. It answers "What do these results mean?" rather than "What are the results?", which the Results section has already addressed.

6.1. What to include in the Discussion

• Comparison with previous literature: Do your findings agree or conflict with earlier studies? If they conflict, how might this be explained?
• Policy and practical implications: What do your findings suggest for managers, policymakers, or practitioners?
• Study limitations: Sample size, data collection scope, missing control variables, limits to generalization, and so forth.
• Future research directions: Unresolved questions and new perspectives to explore.

6.2. What to avoid in the Discussion

• Do not introduce new statistics or tables; this section is entirely prose.
• Do not repeat results already presented in the Results section.
• Do not make recommendations beyond what your data support.

7. Conclusions

The Conclusions section is shorter than the Discussion. It summarizes the most important findings and answers why they matter. According to Daniels & Minot, Conclusions should:

• Summarize the main results concisely, without repeating details from Results.
• Reaffirm the study's importance to the field.
• Optionally restate future research directions briefly.

8. Practical Writing Considerations

8.1. You do not have to write from beginning to end

Many researchers write Results first because the data are ready, then Methods, before returning to the Introduction and Literature Review. No writing order is mandatory.

8.2. Check each journal's requirements

Each journal has its own Author Guidelines for reference formatting, word counts, tables and figures, and sometimes article structure. Reading them before writing saves considerable revision time.

8.3. References

Most journals in the social sciences, management, and marketing use APA style. Reference managers such as Zotero, Mendeley, or EndNote can help prevent errors.

8.4. Writing for two audiences

Daniels & Minot emphasize that the same study can be written in two different styles:

• General audiences, such as journalism and policy: Focus on trends and meaningful findings, avoid technical notation, and use plain language.
• Academic audiences, such as research journals: Report complete test statistics, degrees of freedom, p-values, and confidence intervals in APA style.