Interpreting results: trends, repeats, errors and the control group
Machine translation
Learn
When an experiment is finished, you have to interpret the results: decide what the data say and what they do not say. First you look at the trend: do the data go up, go down or stay almost the same? Then you check whether the data are reliable.
Living things vary: two identical plants do not grow the same. That is why several plants are used per group and the measurements are repeated. With them you work out the mean: you add them up and divide by how many there are. If three bean plants measure 12, 14 and 13 cm, the mean is (12 + 14 + 13) ÷ 3 = 13 cm.
plant 1 2 3 beans (cm) 12 14 13 Three plants: their mean is 13 cm If the means are almost the same, there may be no real difference, just chance. If one mean is much higher than the other, the difference is more reliable, and even more so if it appears again when you measure again. With more plants, the result is more certain.
To find out whether a change has an effect, you compare with a control group. It is a group that is the same in everything except what is being tested. If you test a fertiliser, the control group grows without fertiliser. Without a control you do not know whether the plants would have grown the same anyway.
The control group and the group that receives the change Every experiment has sources of error: an ill plant, a ruler put in the wrong place or one group watered more than another. Write them down. If a measurement comes out very odd, repeat it. Do not simply rub it out.
The conclusion answers the question with the data and says no more. With six beans you cannot talk about all the plants in the world. And it says whether the hypothesis was right, even when it was not.
From the data to the conclusion
Worked examples
Example 1
Mara tests a fertiliser on the plants in her shop, with an adult. She puts out 4 plants with fertiliser and 4 identical ones without fertiliser. After 10 days they measure 14, 16, 15 and 15 cm with fertiliser, and 11, 12, 13 and 12 cm without fertiliser. Which is the control group and what does she conclude?
| plant | 1 | 2 | 3 | 4 | mean |
|---|---|---|---|---|---|
| fertilised | 14 | 16 | 15 | 15 | 15 |
| control | 11 | 12 | 13 | 12 | 12 |
- The control group is the one that does not receive the change: the 4 plants without fertiliser.
- Mean with fertiliser: (14 + 16 + 15 + 15) ÷ 4 = 15 cm. Mean without fertiliser: (11 + 12 + 13 + 12) ÷ 4 = 12 cm.
- The plants in the fertiliser group grew 3 cm more, on average, than those in the control group.
The 4 plants with fertiliser grew 3 cm more, on average, than the 4 control plants. She cannot say that it is true for all plants.
Watch out! Many children compare a single plant from each group and not the means.
Example 2
Tony thinks that music helps seeds to sprout. He puts out 3 cups with 10 seeds and music, and 3 cups with 10 seeds without music. 7, 8 and 9 sprout with music and 8, 8 and 8 without music. What does he conclude?
| cup | 1 | 2 | 3 | mean |
|---|---|---|---|---|
| with music | 7 | 8 | 9 | 8 |
| without music | 8 | 8 | 8 | 8 |
- The control group is the cups without music.
- Mean with music: (7 + 8 + 9) ÷ 3 = 8. Mean without music: (8 + 8 + 8) ÷ 3 = 8.
- The means are the same: no difference can be seen.
In this experiment no difference can be seen: the music did not change how many seeds sprouted.
Watch out! Many children say that music helps, because in one cup 9 sprouted.
Example 3
Vera measures the same bean plant three times and writes down 10 cm, 10 cm and 16 cm. What does she do about the odd measurement?
| measurement | 1 | 2 | 3 |
|---|---|---|---|
| Vera | 10 | 10 | 16 |
- She wonders if she measured wrongly: maybe she misplaced the ruler.
- She repeats the measurement carefully, from the soil to the tip.
- She does not rub out the odd measurement or change it just like that: she writes down what happened.
She repeats the measurement and writes down what happened, without deleting the data.
Watch out! Many children rub out the odd measurement so that the data look nice.
A challenge to think about
Think of an experiment with seeds or plants that you could do at home with an adult. Write what the control group would be, how many plants you would put in each group and what sources of error could mislead you. Then write a conclusion that does not go beyond what you measured.
Practise
Rounds of 10 questions. Choose a level:
Your grade in this topic:No grade yetDo a round at the Basic level to get a grade.