Figure 1. Mean number of creeping juniper projections by location.The graph generated for the purposes of this post (Figure 1) displays the results for my hypothesis that the number of creeping juniper projections differs between side of the stair case (South-East versus North-West). As the predictor variable is categorical, a bar graph was used to display the results. Location is found along the x-axis and mean number of projections is found on the y-axis. Mean calculations and the Mann-Whitney U test (for non-parametric data) were performed using SPSS v24. The graph was generated using Excel.
Despite there being a visual difference in bar height, the data did not support the results I had expected.There was no statistical difference between conditions (p = 0.25). Given the small sample size, I suspect that if I were to increase the number of replicates I would be able to detect a significant difference. I was limited in the number of samples I could take based on the amount of creeping juniper present and ensuring that my data were randomized by varying the distance between samples taken. This method allowed me to make observations along nearly the entire available creeping juniper. Future work should seek to sample multiple similar locations (e.g., slope, geographical area, similar abiotic features present).
