In one line: Data only means something once you can read it, summarise it, and judge how far it can be trusted.
Data analysis and interpretation is studied by all psychology students (assessed HL-only on Paper 3). It covers reading graphs, summarising data with statistics, judging significance, and analysing qualitative data through themes.
This links directly to the concept of measurement: analysis turns raw numbers and words into meaning. On Paper 3, you interpret sources and data, so these skills are essential.
Memory hook: Read it, summarise it, judge it. Data analysis makes evidence meaningful.
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
Key idea: Quantitative data is read through graphs and descriptive stats; qualitative data through themes.
The analysis toolkit
Graphs
Bar charts compare groups; box-and-whisker shows spread and median; scatter plots show a relationship between two variables.
Descriptive stats
Central tendency (mean, median, mode) and spread (range, standard deviation) summarise a data set.
Significance
A significant result is unlikely to be due to chance alone — but 'significant' isn't the same as large or important.
Thematic analysis
For qualitative data (interviews), identify recurring themes in what people say to capture meaning.
Graphs · Descriptives · Significance · Themes
Take a study comparing two teaching methods. A bar chart compares average scores; the mean and standard deviation summarise each group; a significance test asks if the difference is beyond chance; and if interviews were done, thematic analysis captures why. Each tool turns data into measurement-based meaning.
Histogram
Touching bars for continuous scores (e.g. reaction times grouped 200-249 ms, 250-299 ms). Shows the shape of the distribution.
Frequency table
Counts how often each score or category occurs; the first step before drawing a bar chart or histogram.
Line graph
Shows change across time or across levels of a continuous variable, e.g. recall after 1, 5 and 10 minutes.
Normal distribution
A symmetrical bell curve: mean, median and mode are the same, and about 68% of scores lie within one standard deviation of the mean.
Skewed distribution
A long tail on one side. Positive skew = tail to the right (a few very high scores pull the mean above the median). Report the median.
Outliers and the semi-interquartile range
Outliers are extreme scores that drag the mean and stretch the range. The median and the semi-interquartile range (half the distance between the first and third quartiles, the spread of the middle 50%) resist them.
Go further — higher-level insight: Significant ≠ big ≠ important. A statistically significant result just means it's unlikely to be chance; it may still be a tiny effect with little real-world value. Distinguishing significance from effect size is a high-level Paper-3 interpretation point.
Stop wasting time on topics you know
Our AI identifies your weak areas and focuses your study time where it matters. No more overstudying easy topics.
Key idea: Good interpretation asks what the data can — and cannot — show.
Reading data well
| Reading data well | Do | Watch out for |
|---|---|---|
| Graphs | Read axes, units and scale carefully | Misleading scales that exaggerate differences |
| Averages | Report central tendency AND spread | A mean alone hides variation (use SD) |
| Claims | Match the claim to the design | Correlation presented as causation |
So interpreting data means reading graphs carefully (watching for misleading scales), reporting spread as well as averages, and matching conclusions to the research design — a correlation can't support a causal claim. On Paper 3 you also weigh credibility, bias and transferability of the sources.
Watch out: A mean without spread can mislead. Two groups can share a mean yet differ hugely in variation — always look at the standard deviation or the box-and-whisker.
Key idea: Descriptive statistics describe the sample; an inferential test asks whether a difference or relationship is likely to be real or just chance.
Every test starts from the null hypothesis. The test gives a p-value: the probability of getting a result this strong if the null hypothesis were true. Psychologists usually accept p < 0.05 (less than a 5% chance) as statistically significant, so the null hypothesis is rejected.
Which test?
| Which test? | Design | Data | Test |
|---|---|---|---|
| Difference between two groups | Independent samples | Interval or ratio, roughly normal | Unrelated (independent) t-test |
| Difference between two groups | Independent samples | Ordinal, or skewed | Mann-Whitney U test |
| Difference between two conditions | Repeated measures or matched pairs | Interval or ratio, roughly normal | Related (paired) t-test |
| Difference between two conditions | Repeated measures or matched pairs | Ordinal, or skewed | Wilcoxon signed-rank test |
| Difference or association in categories | Independent | Nominal (counts in categories) | Chi-square test |
| Relationship between two variables | Correlational | Two scores per person | Correlation coefficient (Pearson's r; Spearman's rho for ordinal data) |
Reading a result, as Paper 3 shows it
t = 2.41, p = 0.02
p is below 0.05, so the difference is significant: reject the null hypothesis.
U = 45, p = 0.21
p is above 0.05, so the difference is not significant: the result could be chance.
Effect size (Cohen's d)
How big the difference is: about 0.2 small, 0.5 medium, 0.8 large. A significant result can still be a small effect.
Test · p-value · Decision · Size
A stricter level (p < 0.01) makes a Type I error less likely but a Type II error more likely. Small samples also make Type II errors more likely.
Go further — higher-level insight: On Paper 3 you interpret, you never calculate. Say what the test compared, whether p is below 0.05, what that means for the hypothesis in the words of the study, and one limit: a significant result is not proof, may be a small effect, and says nothing about whether the sample can be generalised. This is the concept of measurement.