Measurement ScaleBefore planning a questionnaire or survey, test developers identify the scale of measurement of their data to make sure that they are in alignment with the objectives of the study. Th

Factor Analysis and Cross -Validation PSY3700 Multimedia Assessment and Psychometrics ©20 16 South University 2 Factor Analysis and Cross -Validation Exploring the Efficacy of the Test Factor Analysis Factor analysis is a statistical tool that can be used to reveal multiple interrelationships among variables (Kaplan & Saccuzzo, 2013). Test items can become the variables in the matrix that is generated from this data reduction technique. Factor analysis allows the test developer to distill data derived from items into user -friendly chunks of information (Kaplan & Saccuzz o, 2013). In test development, items can be correlated with each other to reveal the existence of several general underlying components that load onto the construct under investigation. If this occurs, it is possible to identify the essential components t hat comprise the construct (Gregory, 2013). If items are highly correlated to these components, they will also be good discriminators. Further, as an alternative to item analysis, factor analysis may be useful in the overall process of evaluating test item s. For example, it may be used to identify items that show acquiescence effects, that is, people tend to respond to them in predictable ways. Once identified, these items may be removed (Rust & Golombok, 2009). One of the drawbacks of the use of factor an alysis is the introduction of subjectivity in the interpretation of data. For this reason, researchers who use factor analysis should exercise caution to avoid compromising the overall objectivity of their study (Rust & Golombok, 2009). Cross -Validation The purpose of conducting item analysis after conducting the pilot study of an instrument is to identify poorly fitting test items that need revision or exclusion. It is common for test developers to discard up to one -half of the original items on the firs t draft of a test. Ideally, the revised test contains items that are better discriminators and have higher reliability and predictive validity. These items are used in the second tryout study of the questionnaire or survey. Accordingly, the demographic cha racteristics of this second group should approximate those of the pilot study group; otherwise, it can lower the concurrent validity index in the cross -validation process (Gregory, 2013). The next step in the process is to compare the results of the pilot ed study with those of the second tryout. A concurrent validity index score can be calculated to show the correlations between the test and a criterion when given during the same period of time (Kaplan & Saccuzzo, 2013). Accordingly, if items do not requir e too much tweaking, the finalized questionnaire or survey can proceed to the next stage in the process of cross -validation. Validity shrinkage may be observed from additional tryouts of the instrument, which is a common occurrence. Test developers can re -evaluate the demographic characteristics of the latter group of test takers to make sure that they are not negatively affecting the criterion validity. If this is happening, a different sample can be drawn. Likewise, sample size is checked to make sure th at it is sufficient (Gregory, 2013). PSY3700 Multimedia Assessment and Psychometrics ©20 16 South University 3 Factor Analysis and Cross -Validation Exploring the Efficacy of the Test References Gregory, R. (2013). Psychological testing: History, principles, and applications (7th ed.). Boston, MA: Pearson. Kaplan, R., & Saccuzzo, D. (2013). Psychological testing: Principles, applications, & issues (8th ed.). Belmont, CA: Wadsworth. Rust, J., & Golombok, S. (2009). Modern psychometrics: The science of psychological assessment (3rd ed.). New York, NY: Taylor & Francis. © 201 6 South University