alireza khoshgooyanfard; Mohammadreza Falsafinejad; noorali farrokhi
Abstract
Nonresponse is an inevitable challenge to large-scale studies and can result in wasting money, time and human resource involved in data collection and can also prevent the studies from obtaining their objects especially scores distribution. Imputation methods have thus been invented to estimate item ...
Read More
Nonresponse is an inevitable challenge to large-scale studies and can result in wasting money, time and human resource involved in data collection and can also prevent the studies from obtaining their objects especially scores distribution. Imputation methods have thus been invented to estimate item nonresponses in order to make inference from a completed data set. Using a simulation study on a real data set in the form of a multivariate experimental design, this paper evaluates the accuracy of three models including cumulative logit model, graded response model and explanatory item response model. The results show that the imputed values of all three models are acceptable under random nonresponse mechanism although the imputed values of the explanatory item response model are always more accurate than those of the other models. If nonrandom nonresponses are occurred, explanatory item response model has acceptable imputed values only at 5% nonresponse rate and the other models are not accurate at all. The results also show that it is more accurate to impute individual item nonresponses and then compute the total score instead of directly imputing the total score.
Mohammadreza Falsafinejad; Noorali Farroukhi; laila bahrami
Abstract
Background: High school final exams are one of the most decisive tools for scientific assessment of students. Given the importance of this examinations, carried out systematic research on the quality and functionality of their questions is necessary to separate volunteers. Aim: The aim of this study ...
Read More
Background: High school final exams are one of the most decisive tools for scientific assessment of students. Given the importance of this examinations, carried out systematic research on the quality and functionality of their questions is necessary to separate volunteers. Aim: The aim of this study is to determine the psychometric properties and capabilities of the final exam questions in the field of experimental biology and Persian literature in the selection of candidates for admission to undergraduate courses. Methodology: The population, all of the questions of the third year high school final exams string in June 2011. For determining the psychometric properties of these questions, performance of 600 students, selected randomly, in aforementioned lessons in the school districts of the city of Khorramabad were used. Findings: Estimated reliability coefficient in biology and Persian literature was determined by Cronbach alpha, (0.97, 0.96). According to CTT, the average coefficient of difficulty and discrimination in Persian literature and biology test were (0.65, 0.57) and (0.50, 0.65) respectively. In analyzing the IRT; two-parameter model fitted to the data revealed more. Also, in IRT models, the average difficulty and discrimination of questions for Persian literature and biology courses were (-0.69, 1.03) and (-0.09, 0.96) respectively. The most informant rate for two tests of Persian literature and biology, belongs to levels of ability which are (-0.7, 0.1) respectively, the agreement between the CTT and IRT in terms of discrimination parameter the two courses, were (%98.36, %93.59) respectively. Conclusion: Due to the important decisions based on high stakes tests, implications of final examinations in the selection of candidates were discussed.
