![Sample our Education journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5932/TOC-Subject-Sample-2021-Education.jpg"})
Abstract
This paper presents a preliminary investigation of parameter estimation for planned incomplete data in educational research. An algorithm recommended by Beale and Little (1975) is used to estimate correlations for each of the two sets of six variables, based on correlational data obtained from previous research by the authors and from a paper in an educational publication. The algorithm, which gives an estimator which is maximum likelihood for data with a multivariate normal distribution, was found by Beale and Little to be superior to other existing estimation procedures for use with adventitiously missing data. The paper not only looks at the extension of the use of such algorithms to planned incomplete data, but also looks at the effect of several factors of planned incomplete designs on the estimation of correlations in order to suggest optimal designs for incomplete data. While the preliminary nature of the investigation limits the generalizability of the results, the findings suggest that the use of planned incomplete data collection in educational research has potential, and several design factors are nominated as worthy of further investigation.