The results chapter reports what you found, in a fixed order, without interpreting it — and it fails for mechanical reasons far more often than analytical ones. Tables that duplicate the text, missing notes, decimals that change halfway down a column, a p-value with no effect size beside it. This guide walks the chapter step by step, with the APA 7 table anatomy set out in full and a worked reporting template for each of the tests an Indian thesis actually uses. Your department’s format manual has the final word wherever it differs from APA — confirm against your own.
Step 1: Fix the boundary between Chapter 4 and Chapter 5
Before you write a line, decide what this chapter does not do. Results report; discussion interprets. “Job satisfaction was significantly higher among permanent staff (p = .003)” is results. “This suggests that contractual insecurity undermines satisfaction, consistent with Sharma (2024)” is discussion, and putting it in Chapter 4 is the most common structural comment guides write in the margin.
Some departments prescribe a combined “Results and Discussion” chapter, and if yours does, follow it. Otherwise keep the wall clean: no literature citations in Chapter 4 except where you must reference a scoring procedure, and no explanations of why a result occurred. Expected output: a one-line rule at the top of your draft that you can check every paragraph against.
Step 2: Order the chapter the way the reader needs it
The sequence that works in almost every quantitative thesis:
- Response and data preparation — questionnaires distributed and returned, response rate, cases excluded and why, how missing data was handled.
- Sample characteristics — demographics, in one table.
- Instrument properties — reliability for each scale in your sample, and factor structure if you assessed it. What belongs here and how to phrase it is set out in our guide to an acceptable Cronbach’s alpha for a thesis.
- Descriptive statistics — means, standard deviations and correlations for your study variables.
- Assumption checks — normality, homogeneity of variance, multicollinearity, reported briefly and factually.
- Hypothesis tests — in the order your hypotheses were numbered in Chapter 1. Not in the order you happened to run them.
- Summary of findings — a short table mapping each hypothesis to supported or not supported.
That final table is disproportionately valuable. It is the page examiners flip to, and it forces you to confirm that every objective stated in Chapter 1 has an answer here.
Step 3: Learn the anatomy of an APA 7 table once
Every table has the same five parts, in the same order. Learn them once and you will format forty tables without thinking.
- Number. Arabic numerals, numbered sequentially in the order the tables are first mentioned in the text. Do not use suffix letters like Table 3a and 3b; combine the related tables instead. Tables placed in an appendix take a letter and a numeral, such as Table A1.
- Title. On the line below the number, in italicised title case, brief and clear. Where it helps, define an abbreviation parenthetically in the title itself.
- Headings. Every column needs one, written in sentence case and in the singular unless it names a group. Four heading types exist: the stub heading describes the leftmost column, column headings describe the entries beneath them, column spanners describe two or more columns at once, and a table spanner runs the full width. A heading should not be much wider than the widest entry beneath it.
- Body. Entries centred unless left-alignment reads better for long entries; word entries in sentence case. Leave a cell blank if the element does not apply; if a blank needs explaining, use a dash and a general note. Keep the number of decimal places consistent within a column and never change the unit of measurement partway down one.
- Notes. Three kinds, always below the table and always in this order: general notes (explaining abbreviations and the table as a whole), specific notes (superscript lowercase letters a, b, c, ordered left to right and top to bottom, with the first always a), and probability notes.
Two governing principles sit above the details. Each table must be intelligible without the surrounding text, which is why every abbreviation needs explaining even though you defined it three pages earlier. And a table is only justified if it earns its space: data requiring only two or fewer columns and rows belongs in a sentence, not a table.
Step 4: Make text and table do different jobs
Refer to every table and figure in the text — “Table 3 presents the correlations among study variables” — and then tell the reader what to look for in it. Focus on the important point and leave the details for the reader to examine themselves. What you must not do is walk through the table cell by cell in prose; that duplication is the second most common comment guides write, after the results/discussion leak.
Do not repeat the same data in two different tables either. And keep abbreviations, terminology and probability levels consistent across every table and figure in the thesis; a chapter where Table 2 says SD and Table 5 says S.D. reads as assembled rather than written.
Step 5: Use the reporting template for your test
Each test has a conventional form, and each includes an effect size. A p-value tells the reader whether an effect is distinguishable from zero; only the effect size tells them whether it matters. Reporting one without the other is the omission examiners most reliably flag.
| Test | Reporting template |
|---|---|
| Independent-samples t-test | Permanent staff (M = 4.21, SD = 0.63) reported higher satisfaction than contractual staff (M = 3.78, SD = 0.71), t(310) = 5.62, p < .001, d = 0.64. |
| Paired-samples t-test | Scores increased from pre-test (M = 3.10, SD = 0.55) to post-test (M = 3.64, SD = 0.58), t(89) = 7.41, p < .001, d = 0.78. |
| One-way ANOVA | There was a significant effect of department on satisfaction, F(3, 308) = 6.14, p < .001, η² = .06. |
| Chi-square test of independence | The association between gender and tool preference was significant, χ²(2, N = 312) = 11.27, p = .004, Cramér’s V = .19. |
| Pearson correlation | Satisfaction correlated positively with perceived support, r(310) = .43, p < .001. |
| Multiple regression | The model explained a significant proportion of variance, R² = .38, F(4, 307) = 46.91, p < .001; perceived support was the strongest predictor, β = .41, p < .001. |
| Mann-Whitney U | Scores differed significantly between groups, U = 6,412, z = −3.18, p = .001, r = .18. |
Four conventions carry across all of them. Italicise statistical symbols (M, SD, t, F, p, r) but not Greek letters. Put degrees of freedom in parentheses immediately after the statistic. Report exact p-values to two or three decimals, using p < .001 only below that floor. And omit the leading zero for any statistic that cannot exceed 1 — p, r and β are written .43, not 0.43, while M and SD keep theirs.
Step 6: Report the assumption checks factually
This section is short and it should be dull. State which assumption you checked, how, what you found, and what you did about it — in one or two sentences each, without apology or narrative. “Normality was assessed using the Shapiro-Wilk test; scores on the dependent variable departed significantly from normality (p = .008), so a Mann-Whitney U test was used.” That is a complete, defensible report.
Running those checks is a separate procedure that belongs to your analysis stage rather than your writing stage, and the package you run it in shapes how the output is labelled — the trade-offs between the options your library may already license are set out in our comparison of statistical tools for an Indian thesis. What matters in Chapter 4 is that the check appears, that its outcome is stated, and that any change of test is visibly a consequence of it rather than an unexplained substitution.
Step 7: Format figures to the same discipline
Figures follow the same logic as tables: numbered sequentially and separately from tables, with a number and an italicised title, a clear image, and a note beneath where explanation is needed. Axes must be labelled with units. Legends must be readable at print size, which is the check almost nobody performs — print one page and look at it before you decide your chart is finished.
If you reproduce or adapt a figure, table or dataset from another source, gather the full documentation you will need to attribute it properly at the point you save the file, not months later when the source has become an unlabelled screenshot in a folder. Where those datasets legitimately come from, by discipline, is catalogued in our guide to official data sources for an Indian thesis.
Step 8: Run the pre-submission checks
Six checks, in this order, catch nearly everything a guide would send back. Every table and figure is referred to in the text. Numbering runs in order of first mention with no gaps. Decimal places are consistent within every column. Every abbreviation is explained in a general note. Every hypothesis from Chapter 1 has a result here. And every significance test carries an effect size.
Two further ones specific to Indian submissions: your institute’s format manual governs margins, table caption placement and font, and it overrides APA wherever the two disagree; and the results chapter is where similarity checks sometimes throw up surprising matches on standard methodological phrasing, which the recovery sequence in our similarity report recovery plan addresses.
Building the chapter without losing the thread
Chapter 4 is mostly bookkeeping under time pressure: forty tables, each needing a number, a title, notes and a cross-reference, all of which must stay consistent while you are still adding analyses. The errors are not conceptual; they accumulate because the document grew faster than the housekeeping.
Tesify keeps your chapters structured and your references attached to real sources as the document grows, so renumbering a table or adding an analysis in week nine does not quietly break three cross-references elsewhere. The analysis and every claim in it remain yours.
Start writing your results chapter in Tesify
Frequently asked questions
What is the difference between the results and discussion chapters?
Results reports what you found; discussion explains what it means and how it relates to prior literature. Interpretation placed in Chapter 4 is the most common structural error in a quantitative thesis.
Should the results chapter include citations?
Rarely. Cite only where you must reference a scoring procedure or a formula. Comparisons with other studies belong in the discussion.
How do I number tables in APA 7?
Arabic numerals in order of first mention in the text, with no suffix letters. Combine related tables rather than splitting one into 3a and 3b, and give appendix tables a letter and numeral such as Table A1.
Where does the table title go?
On the line below the table number, in italicised title case. Keep it brief, and define an abbreviation parenthetically in the title where it helps the reader.
What are the three types of table notes?
General notes about the table as a whole, specific notes flagged with superscript lowercase letters starting at a, and probability notes. They appear below the table in that order.
Do I report exact p-values or asterisks?
Report exact values to two or three decimals in running text, using p < .001 below that. Asterisks in a table are acceptable when paired with a probability note that defines them.
Do I need to report effect sizes?
Yes, for every significance test. A p-value indicates whether an effect is distinguishable from zero; the effect size indicates whether it is large enough to matter, and examiners increasingly treat its absence as an omission.
Should I include SPSS output screenshots in my thesis?
Almost never in the body. Rebuild the numbers into a properly formatted table; if raw output is genuinely required, place it in an appendix.
How many tables is too many?
There is no limit, but any table that could be a sentence should be a sentence. Data requiring only two or fewer columns and rows belongs in the text.
Can I use IEEE style instead of APA for tables?
If your department requires IEEE, yes — many engineering programmes do. The structural discipline is identical; only the caption placement and citation form change, and your format manual specifies which applies.
What if a result contradicts my hypothesis?
Report it exactly as you would a supportive one, in the same neutral register. Non-significant and contrary findings are results, and concealing or softening them is a far more serious problem than the finding itself.
Where can I read the APA table rules myself?
The Purdue OWL pages on APA tables and figures set out the numbering, title, heading and note conventions in full, and they are freely available. Read them once alongside your institute’s format manual before you build your first table.