The Case of Misbehaving Data or the Proverbial Plan B - Essay Example

Plan B In using parametric tests, one must adhere to certain assumptions such as normality and homogeneity of variance. However, there are many situations wherein experimental situations do not conform to the requirements of parametric tests. As such, employing parametric tests when the assumptions are violated may lead to an erroneous interpretation of data (Cohen, 1988). Thus, the researcher must now turn to nonparametric tests in analyzing such cases. Nonparametric tests are sometimes called distribution-free tests because they make very few assumptions about the population distribution (Bluman, 2004).

Furthermore, while parametric tests most likely require numerical scores, responses in a nonparametric test are usually categorized. It should then be noted that putting responses under these classifications entail that these data involve measurement on nominal or ordinal scales and thus, cannot produce numerical values that can be used to calculate means and variances. This would mean that data for a number of nonparametric tests are simply frequencies. An example of a nonparametric test is the Mann-Whitney test which is used for testing differences between means when there are two conditions and different subjects have been used in each condition.

For example, an experiment may be carried out to investigate the depressant effects of certain recreational drugs (Leech, Barett, & Morgan, 2005). Twenty clubbers are tested with 10 clubbers given an ecstasy tablet to take on a Saturday night while 10 are allowed only to drink alcohol. Then, their levels of depression are measured. In using SPSS to analyze the given data, it should be noted that the data must be inputed using a coding variable. Thus, the data editor will have three columns of data where the first column is the coding variable with two codes (for example, 1 = ecstasy and 2 = alcohol).

The second column will contain values of the dependent variable while the third column will contain values of the independent variable. An exploratory analysis in SPSS will reveal that the data is not normally distributed, indicating that a nonparametric test should be used. To run the analysis, the main dialog box should be accessed by using Analyze, followed by Nonparametric Tests, followed by 2 Independent Samples. The dependent and independent variables should then be placed in their corresponding boxes and the coding variable should be placed in the box labeled “Grouping variable.

” the corresponding numeric codes should then be inputed and after clicking OK, SPSS should now be able to run the analysis (Leech, Barett, & Morgan, 2005).ReferencesBluman, A. (2004). Elementary statistics: a step by step approach, 5th ed. McGraw-Hill.Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Routledge.Leech, N., Barett, K., & Morgan, G. (2005). SPSS for intermediate statistics: Use and interpretation (2nd ed.). New York, NY: Routledge.