A NOTE ON DIFFERENTIAL INFORMATION LOSS IN MULTI-SCALE EDUCATIONAL ASSESSMENT

スポンサーリンク

概要

• 論文の詳細を見る

• Using the notion of differential information loss, this research note provides a theoretical explanation to the phenomenon of the exasperatingly slow convergence of the EM algorithm when it is applied to a multi-scale IRT model used in a large-scale national assessment in the United States. The result should be meaningful to researchers who work on international large-scale examinations with multiple-choice questions. 1. Introduction There is a growing interest, both internationally and within the USA, in applying hierarchical item response theory (IRT) for scaling multiple-choice items in large-scale assessments (e.g., Bryk and Raudenbush, 1992;Takane, 1996;Kamata, 2001;Rubin and Thomas, 2001;Scott, and Ip, 2002). This research note is motivated by a behaviormetric problem that arises during the analysis of large-scale assessment data for the National Assessment of Educational Progress (NAEP), which applies a hierarchical IRT model to scale student responses (Mislevy, Johnson, and Muraki, 1992). NAEP is a highly visible educational assessment program that aims to monitor the achievement of American students over time. Operating under a mandate from the U.S. Congress, NAEP informs the American public about student achievement on a broad range of subjects including reading, mathematics, writing, science, music, and performing arts. The reports from NAEP are of high interest to educational researchers, school district professionals, policy makers, parents, and the results contained in the reports were often cited by politicians in advancing their political agendas. The Educational Testing Service (ETS) is a major NAEP contractor to collect and analyze educational achievement data on a national sample of students from the United States. The basic statistical model employed by NAEP to scale student responses and report results on student proficiencies can be identified with a two-level hierarchical model (Mislevy, Johnson, and Muraki, 1992).

• 日本行動計量学会の論文

著者

• Ip Edward

  Department Of Public Health Sciences Wake Forest University School Of Medicine

• Ip Edward

  Department Of Biostatistical Sciences Public Health Sciences Wake Forest University School Of Medici

関連論文

• Measuring disability in older adults : The International Classification System of Functioning, Disability and Health (ICF) framework
• A NOTE ON DIFFERENTIAL INFORMATION LOSS IN MULTI-SCALE EDUCATIONAL ASSESSMENT

スポンサーリンク