PROPOSING AN ACHIEVEMENT SIMULATION METHODOLOGY TO ALLOW THE ESTIMATION OF INDIVIDUAL IN CLINICAL TESTING CONTEXT

Enio Galinkin Jelihovschi,Cristiano Mauro Assis Gomes

doi:10.28951/rbb.v37i4.423

Abstract

Information from the population should be directly transposed to the individual level only under strict conditions of stationarity and homogeneity. In general, psychological phenomena are neither stationary nor homogeneous. Furthermore, the individual parameters must be estimated. The usual techniques of estimating the individual are rarely attainable. This occurs because in order to get valid estimates of individual parameters, many occasions of the same test must be performed. In this article, we propose an approach (simerg algorithm) to estimate the individual, so that a proper clinical testing for the individual in the context of testing setting that require the respondents' performance becomes viable. We present the rationale of simerg and apply it to estimate the verbal comprehension ability of three individuals, which performed, in 90 different occasions, the 60 items of this ability. In this application, we show how simerg uses the initial empirical performance of these persons and simulates their whole performance. We evaluate the validity of simerg using different measures to predict the empirical performance. We conclude that simerg produces valid estimates. Further studies in a variety of populations and tests that require respondents' performance are needed to enable a stronger conclusion about the validity of simerg.

Highlights

A psychological testing of individuals is understood as the practice of making inferences about a particular person by using psychological tests
As we have been pointing out in this article, we propose a new method in order to make the estimation of individual parameters viable in the context of clinical testing
We stated that the usual methods for estimating the individual demand a huge number of testing occasions, which led us to claim that the clinics of the individual is, nowadays, an almost impossible task

Summary

Introduction

A psychological testing of individuals is understood as the practice of making inferences about a particular person by using psychological tests. The function simerg tries to simulate in the same way as people learn when submitted to many testing occasions This means that the number of correct answers and the ability parameter θ increase as the testing goes on until they arrive to a plateau. It is up to the user to define which type of correct prediction is more important in his or her study; in ours, the empirical and simulated testing of this paper, the type 1-1 correct answer regarding sensitivity, is more important than the type 0-0, concerning specificity, because it reflects the natural movement of the learning process within the empirical which is going from 0 (wrong answer) to 1 (right answer) This movement produces the variation needed to estimate the parameters.

Participants and empirical data

Findings

Conclusion