Selecting a Statistical Test
The third hypothesis proposes that there exists no difference in perceptions of employees on whether they are well trained to do the job after or before the training program. This information will be collected from drivers on their perceptions about the necessity of the training to do the job. The data used will be either existing or generated data. The data for the drivers’ perceptions before the study is existing data while the data on drivers’ perceptions after training will be generated. The dataset will be based on the drivers’ responses to a five point Likert scale on their perception about training before and after training. This will be drivers’ response to Survey09_2 and Survey10_2.
The third hypothesis will be addressed by use of Likert scales to find the levels of opinions by the drivers who have been assessed. In the analysis process, the researchers will use descriptive statistics such as mean and median to determine the statistical test procedure. Since the mean can be found from the available information, it is possible to find the standard deviation. The mean and standard deviation will make the t-test the most appropriate statistical procedure for the researchers.
The t-test is most appropriate because the researchers are trying to investigate an independent variable i.e. training whose two levels are after or before training. The dependent variable is the perception of the drivers. The t-test is selected because the research is about establishing the difference between tow variables, one that has two levels and another variable. The t-test is found by the use of descriptive statistics like mean, which is found from the Likert scale responses.
References
Greene, J., & D'Oliveira, M. (2005). Learning To Use Statistical Tests In Psychology (3, revised, reprint ed.). New York: McGraw-Hill International.
Peers, I. (1996). Statistical Analysis for Education and Psychology Researchers (illustrated, reprint ed.). New York: Routledge.