How to Report SPSS Results in a Thesis, Step by Step

Which numbers to take from SPSS output, how to format them in APA 7, and how to write each result in one clear paragraph, with before-and-after examples.

By Super Admin Published 6 min read Editorial policy

To report SPSS results in a thesis, do not paste the output tables. Pick the numbers that answer each research question, write them in one consistent format (for APA, an italic test symbol, degrees of freedom, the statistic, the exact p value and an effect size), and then say in plain words what the result means. One research question, one short paragraph, one clean table if needed.

Below is the order that works for most quantitative theses, with before-and-after examples you can copy the shape of. The numbers in the examples are invented for illustration only.

Step 1: Map each research question to one test

Before you open the output viewer, write your research questions or hypotheses in a list and put the test next to each one. For example:

  • RQ1: Do male and female students differ in study hours per week? Independent-samples t test.
  • RQ2: Does satisfaction differ across three faculties? One-way ANOVA.
  • RQ3: Is study time related to GPA? Pearson correlation.

This list becomes the skeleton of your results chapter. Each heading answers one question, in the same order you set them out in your introduction. Anything in the SPSS output that does not serve a question stays out of the chapter, or goes to an appendix.

Step 2: Report the sample and descriptive statistics first

Start with who took part and how many responses you analysed after cleaning. Then give means and standard deviations for the main variables, usually in one table. Use N for the total sample and n for a subgroup; the Purdue OWL summary of APA 7 numbers and statistics sets out this distinction.

If you removed cases (incomplete questionnaires, straight-line answers, outliers), say how many and why. Examiners notice when N in Chapter 3 does not match N in Chapter 4.

Step 3: Check assumptions and say what you found

Each test has conditions. You do not need a page on each one, but you need a sentence. The common example is Levene's test for equal variances in an independent-samples t test. According to Kent State University's SPSS guide, if Levene's test is not significant you read the "Equal variances assumed" row; if it is significant you read the "Equal variances not assumed" row. The same guide notes that from SPSS version 31 Levene's test is no longer shown automatically, so you have to tick the homogeneity of variance option yourself.

A typical sentence: "Levene's test indicated unequal variances, so the Welch-corrected t test is reported."

Step 4: Pull the right numbers from each table

Most students lose marks here, not in the analysis. SPSS gives you far more than you need. For the three tests above, these are the numbers usually reported:

  • Independent-samples t test: the two group means and SDs, t, its degrees of freedom, the two-sided p value, and an effect size such as Cohen's d.
  • One-way ANOVA: F with both degrees of freedom (between groups, within groups), p, an effect size, and post hoc results if the overall test is significant.
  • Pearson correlation: r, its degrees of freedom or N, and p.

Check with your supervisor or department handbook which effect size they expect. The worked write-up in the Kent State guide reports Cohen's d alongside the t test.

What to do when Sig. shows .000

SPSS rounds the Sig. column to three decimals, so a very small value can appear as .000. It is not zero. APA style says that when p is below .001 you write p < .001 (see the Purdue OWL page linked above). Never write p = .000.

Step 5: Format the numbers consistently

If your university follows APA 7, these rules from the Purdue OWL guide cover most of what you will write:

  1. Italicise statistical symbols such as M, SD, t, F, r, p and N. Greek letters stay in normal type.
  2. Report means and standard deviations to one decimal place, and correlations and inferential statistics to two. Exact p values take two or three decimals.
  3. No leading zero for values that cannot exceed 1, such as p = .042 or r = .31. Use a leading zero for values that can exceed 1.
  4. Put spaces around = and <, as in t(88) = 2.41.
  5. Degrees of freedom do not take a comma, even above 1,000.

If your department uses another style (Harvard, a journal template, or its own handbook), follow that instead. The point is consistency across the whole chapter.

Step 6: Write the sentence, then interpret it

A results sentence has three parts: what you tested, what you found with the statistics in brackets, and what it means for the research question. Keep deep discussion for your discussion chapter.

Before

The t test was conducted and the Sig. value was .000 which is less than 0.05 so the null hypothesis is rejected and there is a significant difference as shown in Table 4.3 below which shows the group statistics and the independent samples test.

After

Female students reported more weekly study hours (M = 14.2, SD = 3.9) than male students (M = 11.6, SD = 4.4), t(88) = 2.96, p = .004, d = 0.62. RQ1 is therefore supported: study time differed by gender in this sample.

The second version is shorter, gives the reader every number they need, and links the result back to the question. Notice it does not repeat the whole table in prose.

The same pattern for the other two tests:

  • "Satisfaction differed across faculties, F(2, 117) = 4.12, p = .019. Post hoc comparisons showed..."
  • "Weekly study hours were positively correlated with GPA, r(88) = .31, p = .003."

Step 7: Rebuild your tables instead of pasting SPSS output

Raw SPSS tables carry labels like "Sig. (2-tailed)", gridlines and rows nobody asked for. Rebuild the table in Word with only the columns you discuss, a numbered title, and a note for abbreviations. Refer to each table in the text ("Table 4.3 shows...") and do not describe every cell. If an examiner may want the full output, put it in an appendix.

Step 8: Report non-significant results too

A result that is not significant still answers your research question. Report it in the same format with the exact p value, and avoid wording like "almost significant". Write, for example: "There was no significant difference in satisfaction between first- and final-year students, t(118) = 1.12, p = .265." Leaving it out makes the chapter look selective.

A quick checklist before you submit

  • Every research question has its own heading and its own result.
  • N is the same in methodology and results, or the difference is explained.
  • No p = .000 anywhere in the chapter.
  • Symbols are italic and decimals are consistent.
  • Tables are rebuilt, numbered and mentioned in the text.
  • Interpretation stays brief; the discussion chapter does the heavy lifting.

If your methodology chapter also needs tightening, our guide on writing a methodology chapter without generic filler pairs well with this one.

FAQs

Should I paste SPSS output tables directly into my thesis?

Usually no. Rebuild the tables with only the values you discuss. Full output can go in an appendix if your department or examiner wants it.

How do I report p = .000 from SPSS?

Write p < .001. The .000 is a rounded display value, not an actual zero.

Which row do I use in the independent-samples t test table?

Check Levene's test first. If it is not significant, use "Equal variances assumed"; if it is significant, use "Equal variances not assumed".

Do I need to report effect sizes?

Often, yes. A p value on its own does not tell the reader how large a difference or relationship is. Check your department's requirements and report the effect size that matches your test.

Should results and discussion be separate chapters?

That depends on your university's thesis format. Some departments want a separate results chapter and some combine results and discussion, so follow your department handbook or supervisor.

Need help with your results chapter?

If you have your SPSS output but are not sure what to report or how to word it, our research paper support covers data analysis in SPSS or Excel, explanation of output tables, and support with the results and discussion, with you staying the author of the work. Support starts from 100 BDT and is arranged on WhatsApp; see pricing for details. If you only want your written chapter checked for grammar, clarity and tone, try the free 300-word trial first, no card needed.

More from the blog