1. Preferred Theory
When it comes to test fairness, I firmly believe that the IRT remains the preferred theory. One of the reasons for this preference revolves around the fact that the IRT utilizes information curves with the sole purpose of adapting tests more precisely or accurately to each of the individuals involved in the testing process. In this ways, the testers find the best possible opportunity to gather effective information from a given test without necessarily depending on a prescribed format. Undoubtedly, the various IRT information curves play a central role in the removal of CTT and related biases wherein a broad range of difficulty levels tend to fluctuate (Li et al., 2013). Additionally, differential functioning (DIF) increases fairness, given it reduces variables, which often result from linguistics, culture, and gender. Most importantly, IRT allows individual test developers and responsible stakeholders to tailor their tests around the very traits of the tested individuals. The soundness, as well as mutability of IRT makes it the most preferred theory and method in the whole processes of shaping and analyzing tests (Raykov & Marcoulides, 2011). Accordingly, the IRT remains the most appropriate method and theory when addressing fairness questions in testing.
2. Advantages and Disadvantages
a) Classical Test Theory (CTT)
One of the major advantages of CTT involves its broad application, given most researchers have attested to the fact that they are conversant with the theory’s basic approach, with program developers using the CTT’s perspective to create a variety of statistics-related software packages. Second, CTT has more than one model, including X = T + E, which revolve around different assumptions when it comes to determination of errors.
Delegate your assignment to our experts and they will do the rest.
Although the majority of IS and associated studies use CTT, the approach has certain theoretical shortcomings. One such weakness involves its assumption of the linear relationship that exists between the already observed scored and latent variables (Rusch, 2016). In this way, test developers lack the much-needed capacity to achieve the behavioral constructs’ empirical reality. Another shortcoming is the inability to estimate the true scores directly.
b) IRT
Apart from CTT, IRT has far-reaching positive effects on testing. The theory in question has the capacity to tailor a given test to needs. Additionally, it uses graphical illustrations, which remain helpful in the entire testing process. One of the major disadvantages of IRT is that it is quite difficult to use when compared to CTT (Raykov & Marcoulides, 2016). Apart from that, IRT models are complicated and multifaceted, making them difficult to understand.
4. Technology and Test Development
Most neuropsychological and associated assessments typically include measures, which are not only administered but also scored and interpreted by technologies, including computers. According to available studies, the continued adoption and integration of advanced or new technologies into these assessments play a fundamental role in ensuring the attainment of more comprehensive assessments and informed diagnoses. For instance, improved computer programming has enhanced the administration, scoring, and interpretation of measures without necessarily having direct interaction with physicians.
References
Li, F., et al. (2009). Model selection methods for mixture dichotomous IRT models. Applied Psychological Measurement, 33 (5): 353-373.
Raykov, T. & Marcoulides, G. (2016). On the relationship between classical test theory and item response theory: From one to the other and back. Educational and Psychological Measurement, 76 : 325-338.
Raykov, T. & Marcoulides, G. (2011). Introduction to Psychometric Theory . New York, NY: Routledge.
Rusch, T., et al. (2016). Breaking Free from the Limitations of Classical Test Theory: Developing and Measuring Information Systems Scales Using Item Response Theory. Information & Management : 1-15.
