The most important thing I have learned in this class is how to use tests as predictors of employee fit and performance. Tests are important in the recruitment process, as they help in narrowing down potential job applicants. I now understand the types of characteristics effective tests have. Tests must first be reliable, so that the score for an applicant is consistent. This ensures that the test will produce the same result for an applicant, which makes them easier to compare to other applicants. Before using a test for recruitment, it is important to put it through a trial. This trial could consist of having people within the company take the test multiple times to see if their results are consistent.
A good test must also be valid. It should be able to measure the things we want it to. There are three main types of validity: criterion validity, content validity, and face validity. Criterion validity measures how much a test predicts job performance. Content validity has to do with how well a test measures what it is specifically meant to measure. Face validity refers to hoe the test is perceived by the job applicant. If a test has high face validity, applicants see the test as being highly related to measuring the skills they will need for the job. Tests questions can make an impression on the job applicant. If the test taker does not think the questions relate to the job, they may start to view the company in a negative way. It is important to remember that before a test can be valid in any way, it must first be reliable.
Tests must also be cost efficient. The cost of a test must be compared to how well it predicts the certain things it is meant to measure. Part of deciding on the cost of a test, is considering the importance of the position you are hiring for.
Lastly, tests should be fair. They should put applicants on an level playing field. Some tests are more fair than others. Because many tests are made to predict performance, they cannot be entirely fair. People have different cognitive abilities and tests may be seen as easier or more difficult because of this. Test creators should make tests fair in as many ways as they can, so that they only measure the specific things they are meant to.