This book is one of the five volumes in the National Assessments of Educational Achievement series, and provides step-by-step details on how to analyze data collected in a national assessment. Part I provides a general introduction to statistical analyses normally carried out in these large-scale assessments. These analyses cover measures of central tendency and dispersion, standard error, mean score comparisons, identifying low and high performers, correlation, introduction to regression, and charts and graphs. Part II includes item and test analysis using both classical test and item response theories. It covers analysis of pilot and final tests, treatment of missing data, differential item functioning, test dimensionality, linking national assessment results, rotated booklet designs, setting proficiency or performance levels, and scaling. An accompanying CD contains specially designed item and test analysis software (IATA), exercises and supporting data files for both parts of the volume. The five books in the series introduce key concepts in national assessments of student achievement levels, from policy issues to address when designing and carrying out assessments through test development, questionnaire design, sampling, organizing and carrying out data collection, data cleaning, statistical analysis, report writing, and using results to improve educational quality. This book will be of interest to assessment specialists of national, regional, and local educational institutions
researchers
and universities.