Sample Results and Explanation
SAMPLE DATA CHART:
|
|
Nitrate
levels (ppm) |
Bacteria
Counts (#/cm2) |
Nitrate/Bacteria
Rate |
|||||||||
|
|
Site 3 |
Site 4 |
Site 3 |
Site 4 |
Site 3 |
Site 4 |
||||||
|
|
Dry |
Art. Wet |
Wet |
Dry |
Art. Wet |
Wet |
Dry |
Art. Wet |
Wet |
|||
|
Quadrat: |
|
|
|
|
|
|
||||||
|
|
|
|
|
|
|
|
|
|
|
|||
|
2 |
0 |
10 |
30 |
9.00E+06 |
3.60E+07 |
4.20E+07 |
0.00E+00 |
2.78E-07 |
7.14E-07 |
|||
|
3 |
0 |
7.5 |
10 |
1.10E+07 |
1.90E+07 |
1.00E+08 |
0.00E+00 |
3.95E-07 |
1.00E-07 |
|||
|
4 |
7.5 |
25 |
15 |
2.00E+07 |
3.20E+07 |
3.00E+07 |
3.75E-07 |
7.81E-07 |
5.00E-07 |
|||
The above chart is a sample from our complete data chart. We chose to organize our data in a spreadsheet for clarity and easy access.
The Nitrate almost always follows a common pattern. The dry samples have the least while the artificial wet and wet samples have higher nitrate levels respectively.
The Bacteria counts are harder to follow. In general, the wetter environments had more bacteria, however, the important data is really the rate at which nitrate is produced, or fixated by the bacteria.
The last category shows this. We expected the rates of Nitrate/ Bacteria/cm2 to be higher were there was more water. So, we expected the dry samples to have the slowest rate, while the naturally wet samples to have the most. In general this proved to be true.
The next step was to analyze this data to a further degree. We did three t-tests between the 3 means of fixation rates and a chi-test to verify whether or not the artificial wet fixation rate is close to what we expected it to be, which is the wet fixation rate.
STATISTICAL SIGNIFICANCE CHART:
|
|
T
value |
T
alpha (.05) |
Null
accepted? |
|
T-test
Dry: Wet |
2.52 |
2.228 |
No |
|
T-test
Dry: art. Wet |
1.977 |
2.228 |
Yes |
|
T-test
Art. Wet: Wet |
1.752 |
2.228 |
Yes |
|
Chi-test
O= Art. Wet, E=Wet |
2.1634E-06 |
18.307 |
Yes |
The purpose of the t-test between the Dry and Wet mean fixation rates is to positively confirm that there is a statistically significant difference between these two rates.
The t-test between the Dry and the Artificially wet rates is to see whether there is a dramatic difference between what the dry rates are and what the watered rates are. According to this t-test, is says that there is not a significant difference.
The result of the t-test between the Dry and the Artificially wet rates was surprising, so we chose to do a third t-test. This one is between the artificially wet samples and the naturally wet samples. The result came out that there was no significant difference between them. This made the data a little confusing because if there is no difference between the artificially wet and the naturally wet samples and there is a difference between the dry and the naturally wet samples, then there should be a difference between the dry and the artificially wet samples even though there was not.
The last test we did was the chi-test. Since the null hypothesis was accepted, it meant that the artificially wet fixation rate is indeed close to the naturally wet fixation rate that we expected it to be like.
According to most of our data, our original hypothesis was proven true. Two out of the three important data significance tests showed that our results confirmed that higher water levels in the soil increase the rate at which bacteria fixate nitrogen. It is unusual that the t-test between the dry and the artificially wet rates came out the way it did. If there were more samples from replication and repetition, the t-test may come differently.