There are two approaches to cross-validating and they are either empirical or statistical (Levy, 2013).To use the empirical approach you would take predictors and criteria from previous samples and compare them with other samples in an effort to see how they match with the original samples taken. This data is then entered into the predictor scores from the new samples to gain a cross validation score (Levy, 2013). By comparing the predicted score and the actual score shows how valid the sample is going to be. The other method is the statistical method and this involves adjusting the score for any measurement error (Levy, 2013). The idea is to gain a valid score either way one chooses to cross-validate. There are several formats for appraisal rating that can be used such as graphic rating scale, checklist and employee comparison procedures. An advantage to the graphic rating scale is that they are easy to develop and are among the oldest rating formats (Levy, 2013). A disadvantage is the scoring is left up to the opinion of the rater. The next format is the checklist and there are two categories for them and they are weighted or forced-choice. The advantage to these checklists is that the rater would simply check the box in which they think the employee matches. One of the major disadvantages is raters lose the control in the rating process (Levy, 2013). The final format is the employee comparison procedures. There are three types of this format and they are the rank-ordering, paired comparison and the forced distribution formats. While an advantages to this format is that it can show who excels among the employees this can also be a major drawback as the number of employees increases (Levy, 2013). In the measures of central tenden... ... middle of paper ... ...ple in the job and their behaviors used to complete tasks. Questionnaires are used to poll employees to help researchers focus on work behaviors to help with understanding and improving these behaviors. A disadvantage to this approach is that it doesn't look at the job and specific tasks needed to accomplish the job (Levy, 2013). In an unstructured interview the questions are random and not the same for all applicants. This is different in a structured interview in which the questions are job based and in a series that is asked by all applicants (Levy, 2013). The applicant is then scored based on the answers to each question by a point system from those questions. Using an unstructured interview is not very reliable and leaves to many unanswered questions about the applicants. The structured interview is the more reliable and uses a fair comparison on all applicants.