In the results chapter you only report what your analysis found for each objective. The reasons behind it you keep for a later chapter.
Keeping the finding away from the explanation is the difficult part. After months of data collection, every scholar wants to explain a finding the moment it comes up in a table. Please hold that back. Most Indian theses keep results and discussion as two separate chapters, and in that format the results chapter only reports what happened. What it means, you write later in the discussion chapter. Keep the two apart and both chapters become easier to write, and easier for the examiner to check also.
Put the results in the order of your objectives
Before writing even one sentence, list all your objectives and hypotheses down the left side of a sheet. Against each one, note the tables and figures that answer it. This list itself becomes the outline for your results chapter.
Mostly the results chapter starts with a profile of the respondents or the sample. In a survey thesis this means age, gender, location and education. In an experimental thesis it is the specimen and test details. Don’t spend many pages on this profile, one or two tables with a few lines under them is usually enough. After that, go to the objectives and give one section for each objective.
Some scholars arrange the chapter in the same order they ran the tests in SPSS. Please avoid this. The examiner reads with your objectives in mind, and a chapter that follows them is quick to check.
What one result looks like, written out
Let us take one example, with made-up numbers. Suppose the study is on satisfaction with online classes among 300 engineering students. One of the objectives compares the satisfaction of students from urban homes with students from rural homes.
Too thin. “Table 4.6 gives the t-test results. The p-value is less than 0.05. So the null hypothesis is rejected.”
Too much. “Table 4.6 shows urban students are more satisfied than rural students. The reason is that urban students have better internet and more devices at home, as Sharma (2021) also found. This shows the digital divide in India is still wide.”
About right. “Students from urban homes reported higher satisfaction with online classes (M = 3.62, SD = 0.71) than students from rural homes (M = 3.38, SD = 0.80). The difference was statistically significant, t(298) = 2.64, p = .009, with a small effect size (d = 0.31). Hypothesis H3 is therefore supported (Table 4.6).”
In the thin version, nobody can tell which group scored higher. The long version goes into reasons and earlier studies, which is the discussion chapter’s work. The third version tells which group was higher, by how much, whether the difference was significant and how big the effect was. Only then is the result complete. Go through every result in your chapter and see that these four things are there.
Tables that stand on their own
Many times the examiner opens the tables first and reads the text later. So each table has to be understood without the paragraph around it. If a reader cannot able to follow your table without the text, the table needs more work.
We number tables chapter-wise. So in Chapter 4 they go Table 4.1, Table 4.2 and so on, and the title sits on top of the table. For column labels, write proper words. The examiner does not know your SPSS variable codes, so “INC_CAT” should become “Monthly family income (₹)”. Show the sample size where the reader can see it. Some tables have short forms or stars for significance. In that case add a small note under the table telling what each one means.
Many scholars are copying the SPSS output window as a screenshot straight into the chapter. Avoid that. Type the table fresh in Word with only the columns that matter, in the style you used for the other tables. SPSS gives columns like “Sig. (2-tailed)” and “Lower/Upper” which most readers don’t need in the main chapter. Some guides still ask to see the full output, and for them you can put it in appendix.
In the text, tell the reader what pattern the table shows, instead of reading out the whole table. Say you write “Table 4.3 shows that three of the five factors had mean scores above 3.5, with infrastructure scoring highest”. That one line already tells the reader what to look at. A paragraph that repeats all 25 numbers from the table does not help anybody.
Reporting the statistics consistently
Every test has a usual way of reporting. You give the statistic, its degrees of freedom if the test has them, the p-value and an effect size. The table above shows one way of writing these as per APA style. Many social science and management departments in India follow APA, but some departments have their own format and you should ask in the department once. Whatever you go with, use the same one for every result in the thesis.
A few habits help here. Write the exact p-value (p = .031) and not only “p < .05”. For very small values, writing p < .001 is fine. Give the standard deviation along with each mean. Report the effect size for every test also. With 400 respondents, almost any difference comes out significant, and it is the effect size that tells the reader whether the difference is big enough to matter.
How to write up each test one by one, we have covered in the guide onreporting SPSS results.
| Test | A sample reporting line for the text (numbers are made up) |
|---|---|
| Independent-samples t-test | t(298) = 2.64, p = .009, Cohen’s d = 0.31 |
| One-way ANOVA | F(3, 296) = 4.87, p = .003, η² = .05, followed by Tukey’s test |
| Chi-square test of independence | χ²(2, N = 300) = 9.12, p = .010, Cramér’s V = .17 |
| Pearson correlation | r(298) = .42, p < .001 |
| Multiple regression | R² = .36, F(4, 295) = 41.5, p < .001, and after that each β with its p-value |
| Mann–Whitney U | U = 8,412, z = −2.31, p = .021 |
When a figure earns its place
Use a figure when the shape of the data is important. Examples are a trend over the years, an interaction between two factors, a path model from AMOS or SmartPLS, or a stress–strain curve in an engineering thesis. Many scholar adds a bar chart of gender percentages next to a table that already gives the same percentages. That only adds one more page. If the table can show it, leave it in the table.
Check that your figures can be read in black and white also, because many university copies are printed that way. Label both the axes, with units. Keep the same font and size in all figures, which is easier when you make every figure in one software.
If your results are qualitative
For interview or focus group data, arrange the results chapter theme by theme instead of test by test. Start every theme with a few lines telling what the theme is about. After that put the quotations from your participants. Give each quote a code, for example “P7, woman, 42, Salem”, so the reader knows who said it. Mostly two or three quotes per theme is enough, if you put more it starts to look like the transcript itself.
Tell how common each theme was in your data. With twelve interviews, a line like “58% of participants” can give a wrong idea, so there you can simply write “most participants” or “a few of the younger participants”. Also report the views that did not match your themes. Examiners trust a qualitative results chapter more when some disagreement is shown in it.
Before the chapter goes to your guide
Keep the draft open next to your list of objectives and go through these points one by one.
- Each objective has its own section, in the same order as in Chapter 1.
- You mention every table and figure in the text before it comes on the page.
- Table number and title are on top of each table. Where short forms are used, a note under the table explains them.
- One statistic is written the same way every time, with the same decimal places.
- For H1, H2 and the rest, you have written clearly whether it is supported or not, and by which test.
- No sentence gives a reason for why a result came. Reasons go to the discussion chapter.
- Sample sizes in the tables match the sample given in the methodology chapter, or you explain the difference.
Sometimes the analysis is still half done, or Table 4.2 and Table 4.5 show a different total for the same sample. Sort that out first. Polishing the English can wait till the numbers are right. Our data analysis support can rerun or check the analysis together with you and explain each of the output, so you can defend it in the viva-voce. When the results chapter is final, move to the discussion chapter. That is where all the explaining you held back finally goes.
Questions scholars ask
Can I combine the results and discussion chapters?
Some universities and fields allow it, mostly in science and medicine, and a combined chapter can read well. Check your university format and ask your guide first. Even in a combined chapter, write the report of each finding and its explanation in separate paragraphs.
Do I report results that were not significant?
Yes, report them. When a hypothesis is not supported, that is also a finding of your study. Some scholars quietly drop that objective from the results chapter. The examiner compares with Chapter 1 and will ask in the viva-voce where it went, and by then it looks like you were hiding it.
How many tables are too many?
No fixed number is there. Take each table and ask which objective or hypothesis it answers. When you can’t name one, that table can go to the appendix or come out fully. Many results chapters become shorter and clearer after doing this.
Should I write “the hypothesis is accepted”?
Technically, you reject the null hypothesis or you fail to reject it. Many Indian theses write “the hypothesis is supported” when they talk about the research hypothesis. Follow whatever wording your guide likes, and keep it the same in the whole thesis.