Management of Type 2 Diabetes in Adults
Parametric and Non-parametric Tests
Parametric test is a form of a statistic test that best on already established parameters and values. In these tests, the researcher has already developed a probability across the population, and only seeks to provide data that seem to support the assumption. Geisser and Johnson (2006) assert that the tests are fixed in regards to how the training data can influence the results of the research. The tests are mainly based on the assumption that the probability of the variables are similar to the data collected, and the expected result is likely to be proved by the training data.
On the other hand, non-parametric tests are based on the value of the training data. In essence, the test can increase or decrease based on the training data. For instance, the variables being tested in the research can change if new information is accessed. Murphy (2012) points out that the tests are versatile in regards to applying emerging data different from variables being tested. The main difference between the two tests if that parametric tests are based on specific values or parameters which the researcher seeks to approve, while the non-parametric tests can change based on the training data (Wasserman, 2007).
Data Analysis Plan
My data analysis plan will be based on already established variables, which assume that existing intervention and management methods are not enough in addressing management of Type 2 diabetes in adults. For this reason, the research will use parametric tests as it is expected that the data from the population will depict the same hypothesis as that stated in the research variables. In my research the central tendency that will be tested is how much adults are limited in accessing information on how to manage their type 2 Diabetes condition.
1. Geisser, S. & Johnson, M. (2006). Modes of Parametric Statistical Inference. New York: John Wiley & Sons.
2. Murphy, K. (2012). Machine Learning: A Probabilistic Perspective. Boston: MIT Press.
3. Wasserman, L. (2007). All of Nonparametric Statistics. Boston: Springer.