Comparisons of Methodology on Estimating Different Reliability for Single Test

• Main Authors: TSAI, PEI-HUAN, 蔡佩圜
• Other Authors: WU, YUH-YIH
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/aw992r

博士 === 國立臺南大學 === 教育學系測驗統計碩博士班 === 105 === The purpose of this research is mainly to analyze the accuracy of different reliability index by employing ρ_glb 、λ_1 、λ_2 、λ_3 (α)、λ_4 、λ_5 、λ_6 、ω_h 、ω_t as the major arguments. Confirmatory Factor Analysis (CFA) is utilized for simulating data in this experiment, basically relying on independent variables (the number of test factors, the number of test items, the number of samples) and dependent variable (bias, absolute mean bias, root mean squared error). The statistical results and analyses are described as following: (1) λ_3 (α) , the most commonly and traditionally used, only suitable for the analysis of one-dimension test, reliability index value will be significantly underestimated if multi-factor test takes place. (2) The estimating errors of different reliability methods have the effect of interacting with the type of test, the number of factors, the number of items and the number of samples. However, ω_t 〖、λ〗_4 display best values of reliability estimation with extreme little error, it is recommended that these two can be used as much as possible. (3) The ratio of ω_h to ω_t is the ratio of the explanatory rate of g factor to the total explanatory rate of all common factors (including g and all f). It is recommended that it can be used as an indicator of whether the undergoing test is close to one dimension. (4) When the structure of factor of the test data is very clear, ω_t is the most suitable role to estimate the overall reliability. On the other hand, if it is not clear, then λ_4 is the appropriate candidate to do the work. (5) Unless it is for analyzing the parent group data, then ρ_glb shows a high estimated value of reliability which is not proper to name it as the greatest lower bound reliability. (6) The errors from these six major arguments (ρ_glb 、λ_2 、λ_3 、λ_4 、λ_6 、ω_t) for estimating reliability can be divided into two groups: (ρ_glb 、λ_4 、λ_6 、ω_t) and (λ_2 、λ_3). The former is less affected by the number of factors, the latter is only applicable to a single factor test, while (λ_4 、ω_t) and (λ_2 、λ_3) display the most similarity. In view of the theoretical simulation of this study, the results of this study can provide testing persons with more appropriate estimates of reliability indicators according to different test scenarios.