Uncontrolled Variables in a Bounce Investigation
Mr. Chen's grade 8 science class investigates: 'Which surface makes a tennis ball bounce highest?' Six groups use a meter stick, a tennis ball, and three surfaces: tile, carpet, and grass. Group 3 drops the ball from 100 cm onto tile, 80 cm onto carpet, and 120 cm onto grass. They use one trial per surface and judge height by eye. They conclude grass produces the highest bounce. Other groups drop from 100 cm on all surfaces, use three trials, and measure bounce height with a slow-motion phone video. Their combined data show tile highest (mean 62 cm), carpet lowest (mean 28 cm), grass middle (mean 41 cm). The teacher has 20 minutes for discussion and wants to use the discrepancy to teach experimental design.
1. Critique Group 3's conclusion. Identify the uncontrolled variables and measurement problems, and explain how each affects validity.
Read the worked answer
Group 3 changed drop height across surfaces (100, 80, 120 cm), so surface is confounded with height. If grass was dropped from 120 cm, the ball had more energy and could bounce higher regardless of surface. One trial per surface cannot show reliability; a single bounce may be an outlier. Eyeball measurement is imprecise and subject to bias. Therefore the conclusion 'grass produces highest bounce' is not supported. The comparison is invalid because at least one alternative explanation-drop height-is not controlled. The class data also show the opposite pattern, which suggests Group 3's result came from procedure, not surface.
2. Write a revised procedure for Group 3 that controls variables and produces reliable data.
Read the worked answer
Drop from 100 cm for all three surfaces; use the same ball; measure bounce height with a meter stick or slow-motion video from a fixed angle; repeat three trials per surface; compute the mean and range; keep the same person dropping and measuring if possible. Record surface type as the independent variable. This controls drop height and ball type, reduces measurement error, and allows comparison of means. Repeating trials also lets the group see variability instead of treating one bounce as the result.
3. How should the class use Group 3's discrepant data?
Read the worked answer
Treat it as a design flaw to analyze, not discard. Compare Group 3's procedure with others; identify the confound; have Group 3 redo one surface at 100 cm and compare to class data. Discuss why scientists repeat trials and control variables. The discrepant data become evidence for why procedures matter. The class should not average Group 3's data with the others because the methods differ. This turns an error into a lesson about validity and reliability.