The data analysis chapter, often titled ‘Results’, ‘Findings’ or ‘Data Analysis and Interpretation’, is where a quantitative thesis delivers its evidence. Examiners read it to check that you analysed the data appropriately, reported the results accurately, and answered each research question. This guide gives a practical structure for the chapter, explains what belongs in each section, and lists the mistakes that most often lead to revisions. The structure is a common one, but always check your university’s thesis guidelines and your supervisor’s preferences first.
Results vs discussion: keep them separate
In most quantitative theses, the results chapter reports what you found, objectively and without speculation, and the discussion chapter interprets it: what the findings mean, how they compare with previous research, and their implications and limitations (Creswell & Creswell, 2018). A short sentence explaining what a result shows in plain language is fine in the results chapter (‘teachers in private schools reported higher job satisfaction’), but explanations of why and comparisons with the literature belong in the discussion. Some programmes combine the two, so follow your guidelines.
A typical structure
- Introduction to the chapter
- Data preparation and screening
- Response rate and sample characteristics
- Reliability (and validity) of the instruments
- Descriptive statistics for the main variables
- Assumption checks
- Hypothesis testing, one research question at a time
- Summary of findings
1. Introduction
Begin with one or two short paragraphs that remind the reader of the purpose of the study and explain how the chapter is organised. Do not repeat the full methodology; a sentence such as ‘Data were analysed using IBM SPSS Statistics (Version 29)’ is enough, with details of the analysis plan already given in the methodology chapter.
2. Data preparation and screening
Explain what happened between data collection and analysis. Examiners value transparency here:
- How many responses were received and how many were excluded, and why (for example incomplete questionnaires, failed attention checks, duplicates).
- How missing data were handled (listwise deletion, pairwise deletion or imputation) and how much was missing.
- How outliers were identified (for example boxplots or standardised scores beyond ±3.29) and what you did with them (Tabachnick & Fidell, 2019).
- Which items were reverse-scored and how composite scores were calculated (sum or mean of items).
3. Response rate and sample characteristics
Report the response rate (completed questionnaires divided by those distributed) and describe the final sample with a demographic table: frequencies and percentages for categorical variables such as gender, age group, qualification or school type, and the mean and standard deviation for continuous variables such as age. Keep the text brief and highlight the main features; do not repeat every number from the table.
4. Reliability of the instruments
Report the internal consistency of each scale and subscale for your sample, usually with Cronbach’s alpha. A small table with the number of items and alpha for each construct works well. If you removed items, say which ones and why. See Cronbach’s alpha interpretation for benchmarks, and use the Cronbach’s alpha calculator for item-total statistics. If you conducted factor analysis to check construct validity, report it here too.
5. Descriptive statistics for the main variables
Present the mean, standard deviation, minimum and maximum (and often skewness and kurtosis) for each main variable. For Likert-type composites, readers find it helpful if you explain how to read the means, for example on a 1–5 scale. A correlation matrix of the main variables is also useful at this point, especially before regression analyses. The descriptive statistics calculator gives all of these values in one table.
6. Assumption checks
Before the inferential tests, summarise how you checked their assumptions: normality (histograms, Q–Q plots, skewness and kurtosis), homogeneity of variance (Levene’s test), linearity and multicollinearity for regression (scatter plots, VIF), and sphericity for repeated measures ANOVA. Report the conclusion and what you did if an assumption was violated, for example using Welch’s t test or a non-parametric alternative. Detailed plots can go in an appendix.
7. Hypothesis testing
This is the core of the chapter. Organise it by research question or hypothesis, in the same order as in your introduction, and use the same wording so the examiner can follow the link. For each one:
- Restate the research question or hypothesis.
- Name the test and briefly justify it if it is not obvious (for example ‘Because the data were not normally distributed, the Mann-Whitney U test was used’).
- Present the results in APA style: descriptive statistics, test statistic, degrees of freedom, exact p-value, effect size and confidence interval.
- Use a table when there are several variables or comparisons.
- State the decision clearly: ‘The null hypothesis was rejected’ or ‘The hypothesis was supported/not supported’.
- Add one sentence in plain language summarising what the result shows.
Example for one hypothesis:
H1: There is a significant difference in job satisfaction between public and private school teachers. An independent samples t test showed that private school teachers (M = 3.84, SD = 0.61) reported significantly higher job satisfaction than public school teachers (M = 3.52, SD = 0.70), t(118) = 2.67, p = .009, d = 0.49, 95% CI of the difference [0.08, 0.56]. H1 was therefore supported: private school teachers were moderately more satisfied with their jobs.
Our calculators produce an APA paragraph for every test, which you can adapt. For guidance on specific tests, see reporting a t test and reporting ANOVA. If you are unsure which test fits a hypothesis, see how to choose a statistical test.
8. Summary of findings
End the chapter with a short summary, often a table listing each hypothesis, the test used and whether it was supported. This gives the examiner a clear overview and provides a bridge to the discussion chapter.
| Hypothesis | Test | Result | Decision |
|---|---|---|---|
| H1: Difference in satisfaction by school type | Independent t test | t(118) = 2.67, p = .009 | Supported |
| H2: Relationship between workload and stress | Pearson correlation | r(118) = .41, p < .001 | Supported |
| H3: Difference in stress by experience group | One-way ANOVA | F(2, 117) = 1.36, p = .261 | Not supported |
Tables and figures
- Number tables and figures in the order they appear and refer to each one in the text (‘As shown in Table 4.3…’).
- Follow APA 7 format: a bold table number, an italic title on the next line, horizontal lines only, and notes below (APA, 2020).
- Do not paste raw SPSS output into the chapter. Rebuild the table with only the numbers the reader needs, and put the full output in an appendix if required.
- Use a figure only when it adds something, such as a scatter plot of a key relationship or a plot of means across time points. Avoid 3D and pie charts for statistical results.
- Do not present the same numbers in both a table and a figure.
Writing style
- Write in the past tense for your results (‘Scores increased…’).
- Be precise and neutral. Avoid words such as ‘proved’, ‘very significant’ or ‘almost significant’. A p-value of .07 is simply not significant at α = .05; report it and its effect size without spin.
- Report exact p-values (p = .031, p < .001) and effect sizes for all main results.
- Use consistent decimal places: two for most statistics and two or three for p-values.
- Italicise statistical symbols such as M, SD, t, F, r, p and N.
Common reasons for revisions
- Results not linked to specific research questions or hypotheses.
- Missing assumption checks, or tests that do not match the data (for example t tests for three groups).
- Missing effect sizes, or p = .000.
- Raw SPSS tables copied into the chapter.
- Interpreting results in the light of the literature inside the results chapter when the programme expects a separate discussion.
- Inconsistencies between numbers in the text and in the tables. Always double-check them.
Roman Urdu mein khulasa
Data analysis (results) chapter mein sirf nataij report kiye jate hain; un ki tafseel aur pichli research se muqabla discussion chapter mein hota hai. Aik aam tarteeb ye hai: introduction, data screening (missing values, outliers), response rate aur demographic table, har scale ka Cronbach’s alpha, descriptive statistics, assumptions ki checking, phir har hypothesis ko alag heading mein test karein (test ka naam, APA statistics, effect size, aur faisla: supported ya not supported). Aakhir mein summary table dein. SPSS ka raw output paste na karein; APA 7 format mein saaf table banayein.
Frequently asked questions
What should the first section of the results chapter contain?
A brief introduction to the chapter, followed by data screening: responses received, exclusions, missing data and outliers.
Should I interpret my results in the results chapter?
Give a one-sentence plain-language summary of each result, but keep explanations and comparisons with the literature for the discussion chapter, unless your programme combines them.
Can I paste SPSS output into my thesis?
Not in the main chapter. Create APA-style tables with the key statistics and put full output in an appendix if required.
How should I organise the hypothesis tests?
One section per research question or hypothesis, in the same order and wording as in your introduction, each ending with a clear decision.
What tense should I use?
Use the past tense to describe your results, for example ‘Scores increased significantly’.
Related calculators
- Descriptive Statistics Calculator
- Cronbach’s Alpha Calculator
- Independent Samples t Test Calculator
- Test chooser: which statistical test should I use?
References
- American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.). https://doi.org/10.1037/0000165-000
- Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE.
- Field, A. (2018). Discovering statistics using IBM SPSS Statistics (5th ed.). SAGE.
- Pallant, J. (2020). SPSS survival manual: A step by step guide to data analysis using IBM SPSS (7th ed.). Routledge.
- Tabachnick, B. G., & Fidell, L. S. (2019). Using multivariate statistics (7th ed.). Pearson.
Written and maintained by the ScholarCalc team. Last reviewed: November 2026.