Question

1. You have been assigned to design an experiment to test the effect of 5 different kinds of fertilizer on the yield of strawberries grown hydroponically (using water rather than soil).

1. How many plants would you use in this study and why? (2 points)

2. What would you do for a control and why ? (2 points)

3. How would you assign plants between control and treatment groups and why would you do it that way ? (2 points)

4. What would be your null hypothesis (1 point)

5. What would be your alternative hypothesis (1 point)

6. What would you do in terms of carrying out the experiment itself (2 points)

7. What would you measure? (1 point)

8. What might be some confounding factors that you would like to try to minimize and how would you go about it

1. What statistical test would you use to test your hypothesis and why? (2 points)

10. If your p on that test was 0.021 what would you conclude and why? (2 points)

11. What would you do next if you wanted to be able to tell which fertilizer was better than which? (2 points)

50. Do you think your results could be used to assess which fertilizer would work best for strawberries grown in regular soil? Why or why not? (2 points)

Answer 1

a)

We should have at least 2 elements per groups. Hence we should have 6 * 2 = 12 plants in the study.

Here we used 6, because there are 5 types of fertilizers and one placebo(control) group.

Also, do note that we should have the numbers of elements to be as high as possible for the normality assumptions to be valid.

b)

For the control group, we will not use any fertilizer

c)

We would assign 2 to each 6 groups. Ie 2 to the placebo group and 10 to the five fertilizer groups (2 each)

d)

Null Hypothesis would be that all groups would have same mean

e)

Alternate Hypothesis would be that at least one group is different

f)

We will not give any fertilizer to the control group and one fertilizer each to the remaining five groups.

g)

We would measure the time taken by strawberries to grow from seeds

h)

The confounding variables would be the amount of water given and sunlight. To minimize the impact of these variables we would ensure that water and sunlight are given equally to the six groups.

i)

We would be using one way ANOVA to compare the six groups

j)

If our p value was 0.021, we would reject the null hypothesis as it is less than 0.05 (significance level)

k)

We would do a post hoc test

l)

No, results should not be used for the soil experiment as the fertility of soil will impact the results