Parametric and Non-Parametric Tests in Healthcare Study

Differences between Parametric and Non-parametric Tests

Parametric and non-parametric tests are used in research to analyze the statistical data received from the experiments. The main major difference between parametric and non-parametric tests is that the first one is characterized by specific assumptions made about certain parameters of the population, while the latter one is used when there are no assumptions whatsoever about these parameters. As a rule, most parametric tests concern themselves with finding means, while the main parameter within non-parametric tests is the median. Lastly, parametric tests concern themselves with variables only. Non-parametric tests are more flexible, allowing to deal not only with variables but with attributes as well.

Examples of Parametric and Non-Parametric Tests

An example of a parametric test would be a standard T-test. It is used to determine if there is a significant difference between the means of the two groups. In our research, the T-test will be used to compare the results derived from the test group with the results of the control group. An example of when a T-test is used could be when the researcher needs to find out how much money several people are spending on their groceries. It will allow discovering the mean (“T-test,” 2016).

An example of a non-parametric test would be a Mann-Whitney U-test, which is used in a similar way to a T-test and is very good at determining normal distributions. An example of a Mann-Whitney U-test could be observed in psychology, where it is used to compare correlations between behavior patterns and attitudes in patients test (“Mann-Whitney U-test,” 2016).

Assumptions in Parametric and Non-Parametric Tests

Every kind of test, whether parametric or non-parametric, has several assumptions that an investigator must take before going through with the test. In the case of the parametric tests, the assumptions are that the researcher knows the variable distribution. It also assumes that the interest variables are measured in intervals (“Difference between parametric and non-parametric test,” 2016).

For the non-parametric tests, on the other hand, it is assumed that the variables are calculated using nominal or ordinal methods. It is frequently used when the independent variables are not measured in any particular scale (“Difference between parametric and non-parametric test,” 2016).

Brief Data Analysis Plan Discussion

As it was mentioned previously, the main tool for analyzing the results of the experiment will be the T-test and the U-test, as we will be comparing the results of the test group to that of a control group. The U-test will be invaluable to us, as it will allow counting in certain parameters outside of the range of the T-test (“Mann-Whitney U-test,” 2016). The study hypothesis that we will be testing out states that the application of PEFM therapy will reduce neuropathic pain in diabetic patients.

The null hypothesis is that no change will occur. As for the demographical statistics, it is expected that the central measurements will have to deal mostly with the age of the patients involved in the experiment. Another variable would be gender, as it is known that men and women respond to pain differently, with females having a higher threshold. Other variables like social status, education, and family status will not be as valuable to use in the scope of study of the effects of PEFM therapy on patients with diabetic neuropathy.

References

Difference between parametric and non-parametric test. (2016). Web.

Mann-Whitney U-test. (2016). Web.

T-test (Independent samples). (2016). Web.