Lesson 4: Interpreting heart rate data from the experiment
Independent groups and the unpaired t-test
SkillsAnalysing heart rate dataDesigning the heart rate experiment
Worked example
Another school randomly allocated 16 students to watch either the scary clip or the advertisement (8 in each). Changes in heart rate (bpm):
- Emotive: 9, 15, 6, 12, 18, 7, 11, 10 (mean 11.00, SD 4.00)
- Informative: 3, −2, 4, 1, 0, 5, 2, 3 (mean 2.00, SD 2.27)
A student analyses the data with a paired t-test and gets . Explain why this is wrong, give the correct test and its df, and state the conclusion given , .
Step 1: Why the paired test is wrong
Each student watched only one clip, so there are no pairs: the first emotive score and the first informative score belong to different people. Pairing them is arbitrary, and a different order would give a different t.
Step 2: The correct test
Independent groups → unpaired t-test, .
Step 3: Conclusion
, : reject H₀. The mean change in heart rate was significantly greater for the emotive clip (11.00 bpm) than the informative clip (2.00 bpm).
Step 4: A limitation of this design
Participant variables (resting heart rate, sensitivity to horror) differ between the groups; random allocation spreads them by chance, but with 8 per group a chance imbalance is possible.
Answer
There are no pairs (different students in each condition), so use the unpaired t-test with df = 14. , : reject H₀; the emotive clip raised heart rate significantly more.
Try a similar one
A fresh question of the same kind, checked as you answer, with hints and a full solution.
Method
Ask: did the same (or matched) people do both conditions? If not, the test is unpaired.
Unpaired df = n₁ + n₂ − 2.