AP Statistics Project Proposal
(It has to be an experiment, no observational study will be accepted)
- Topic (What is the research question)
Comparison of the Effects of Gibberellin and Brassinolide on Soybean Seed Germination
- Sampling (Be specific about how subjects might be selected)
Randomly select 120 soybeans that are purchased from the same source.
Type of plant hormone being used: categorical, no plant hormone (control), Gibberellin, Brassinolide, or both.
Response Variable: the height of seedlings after 2 days; the height of seedlings after 7 days.
We have 4 kinds of treatments.
For the selected seeds, randomly assign them to different treatments, which is soaking the soybean seeds in specified solution for 12 hours. Each treatment has 30 seeds:
|No plant hormone||10-6 mol/L Gibberellin|
|10-6 mol/L Brassinolide||5×10-7 mol/L Gibberellin + 5×10-7 mol/L Brassinolide|
After soaking, transplant the seeds to soil. Water the seeds every 12 hours, take an account of the proportion of germinated seeds. Also measure the height of seedlings after 2 days and 7 days.
- What extraneous variables might influence the response?
Environmental factors, like temperature, light intensity, water, and even soil conditions all could have impact on seed germination.
- How does the design protect against its potential influence on the response through blocking, direct control, or randomization?
We would use preliminary experiment to ensure the optimal conditions for plant growth. Despite that, we would also apply:
Direct control: All the seeds are grown in the same room and are expose to the same environmental factors so that the extraneous factors’ effects are not confounded with those of the experimental variables.
Randomization: Randomly assign the seeds to different treatment groups to ensure that the experiment does not systematically favor one treatment over another.
Replication: There is considerable amount of individuals for each treatment to achieve an adequate number of observations for each experimental condition.
- Statistical method (e.g. we plan to use…method to explore … We hope to have a …result)
Normally, we would apply what we learned from descriptive statistics: Calculating important statistics, like the mean heights of seedlings, and representing them in the form of a bar chart for comparison.
For the part of inference, we plan to use two-sample t-test to explore whether there is a significant difference in the effects of different plant hormones in the heights of seedlings. We hope that there is significant difference so that we could have evidence to support that one plant hormone is greater in promoting germination than the other.
We also plan to use regression analysis to explore whether there is a linear relationship between the heights of seedlings after 2 days and those after 7 days for a specific treatment group, thus implicitly see if each hormone has prolonged effect.