#6 How to Write the Results Section of a Research Paper?

A strong results section is built by organizing findings in a clear, logical order—starting with how data was prepared, then moving through the main results using the right statistics, and ending with any reanalysis done to confirm conclusions. The core goal is straightforward: report what was found (and what wasn’t) without interpretation, so readers can verify the evidence and understand the patterns in the data.

Most results sections begin with data preprocessing. That can include describing steps taken before analysis, such as filling missing values (for example, using linear interpolation), transforming and normalizing data, and applying smoothing to reduce noise. When difficulties arose during data collection or processing, those should be documented as well. This transparency helps readers judge how the final dataset was produced and why the reported outcomes are trustworthy.

After preprocessing, the main findings should be presented in a logical sequence. The writing should answer three questions as results are reported: what was found, what was not found, and what was unexpected. In a psychology example about how different music types affect people, authors reported significant differences in memory recall between classical and pop music, found no differences in emotional response, and noted an unexpected similarity—both music types fatigued listeners at the same rate. The key is tone and structure: positive, negative, and neutral outcomes should appear in a balanced, unbiased way, with little to no interpretation.

Quantitative reporting matters. Results should use descriptive statistics and inferential statistics to communicate both the shape of the data and the strength of evidence. Mean and standard deviation are common for summarizing central tendency and variability, while confidence intervals and p values support inferential claims. In a social-sciences example linking social media use to mental health, anxiety differences between high and moderate users were described as statistically significant, with the p value specified. The groups’ social media time was reported as mean ± standard deviation, and sleep disturbance risk was expressed using an odds ratio with confidence intervals.

To make dense information readable, results should be paired with figures and tables whenever text would become cramped. A practical rule is simple: if the data can’t be presented clearly in one or two lines, move it into a figure or table. Choosing between them depends on purpose—figures are best for illustrating trends and patterns, while tables are better for showing exact values and large amounts of numerical detail. The typical approach is to describe the trend in the text, refer readers to the figure for visual support, and note that full numerical values are available in a table.

Finally, if the study required reanalysis to reconfirm findings, that should be reported. The overall payoff is credibility: using appropriate statistics and clearly formatted evidence helps summarize results meaningfully and makes the narrative easier for readers to trust and evaluate.