Reliability of Results for the Clerical Test and Work Samples
Before interpreting the reliability results for the clerical test and work sample it is a good idea to first define what reliability of measurement is. A measurement is reliable to the extent that it provides a consistent set of scores to represent an attribute. In the majority of the case perfect reliability is never achieved because of the errors that the distinct types of measurement have. If we test the same time more than once, we are going to have greater reliability.
As noted in the first table of the clerical test, the mean score in the results for the clerical test are very similar in the two tests made by the company, in Time one the mean was 31.61 and in Time 2 the mean was 31.22. On the results of work sample of tactfulness and concern of customer the mean scores are also pretty similar. The results for the standard deviations in Time 1 for the clerical test and work samples are also very close to the results of the standard deviations on Time 2. This is one indicator that the tests made are reliable. Besides this indicator, there other results that tells us that these tests are reliable. The test-retest value is .92 which is a very high value, and it is another indicator of reliability of the results. The coefficient alpha also has high values in both Times (.85 and .86), and the values of the mean and standard deviation and test retest combined with these high values of coefficient alpha gives us a good indicator of the reliability of the tests.
Given the numbers of the tables in the book, and the interpretation of the results I just made, I will consider these test results reliable and favorable enough for Phonemin to consider them in selecting new job applicants. If the company will like more reliability in the test results, one thing they can do is to conduct another test in Time 3, which will not be much of a difference given the results of the numbers. Validity...