User: | Open Learning Faculty Member:
I did not have too much difficulty organizing my table or graph. For the table I had sample #, pH as read off a meter, pH as found off pH strips, and the presence or absence of underbrush as the different columns. For the graph of my data I graphed the presence of underbrush as teh dependent variable against the soil pH as read off the pH meter. The graph felt a little funky for me since the presence of underbrush is a categorical variable. I assigned 1 to be presence of underbrush and 0 to be absence of underbrush so the data looked like a Bernoulli distribution to me. When a best fit line was placed in on the scatter plot there appeared to be a slight negative relationship, but when the means of each of the two sites were calculated they were so close together only differing by a value of 0.066 pH that without being able to evaluate the data using p-values, since I can’t assume normal distribution, I have to accept the null hypothesis. This outcome is unexpected as I went in thinking that soil pH was probably an influencing factor on the distribution of the underbrush, but there are definitely other factors that could be the cause of this pattern. If I were to explore this topic further I would look into those other factors like interspecies competition, shading, soil type, and soil moisture.
