Describing the variation in test scores between and within school districts is critical for: (1) for policy-related and descriptive work that investigates the sorting of students among districts and the differential effectiveness of those districts; and (2) for methodological work planning future experiments or interventions. Intraclass…

This chapter will provide potential models for analyzing learning data through a discussion of screening data and then analyzing that data using appropriate statistical techniques.

The growth in the availability of longitudinal data--data collected over time on the same individuals--as part of program evaluations has opened up exciting possibilities for evaluators to ask more nuanced questions about how individuals' outcomes change over time. However, in order to leverage longitudinal data to glean these important insights,…

We propose a new limited-information goodness of fit test statistic C[subscript 2] for ordinal IRT models. The construction of the new statistic lies formally between the M[subscript 2] statistic of Maydeu-Olivares and Joe (2006), which utilizes first and second order marginal probabilities, and the M*[subscript 2] statistic of Cai and Hansen…

Value-added modeling is an alternative approach to test-based accountability systems based on the proportions of students scoring at or above pre-determined proficiency levels. Value-added modeling techniques provide opportunities to estimate an individual teacher's effect on student learning, while allowing for the possibility to control for the…

This paper considers the model of school effectiveness (SE) currently dominant in research, policy and practice in England (although the concerns it raises are international). It shows, principally through consideration of initial and propagated error, that SE results cannot be relied upon. By considering the residual difference between the…

A multidimensional item response theory model that accounts for response style factors is presented. The model, a multidimensional extension of Bock's nominal response model, is shown to allow for the study and control of response style effects in ordered rating scale data so as to reduce bias in measurement of the intended trait. In the current…

In discussion about search keys, concern has been expressed as to how the number of items retrieved by a single value relates to collection size. This paper creates a statistical model that attempts to give some insight into this behavior. (6 references) (Author)

The (univariate) isotonic psychometric (ISOP) model (Scheiblechner, 1995) is a nonparametric IRT model for dichotomous and polytomous (rating scale) psychological test data. A weak subject independence axiom W1 postulates that the subjects are ordered in the same way except for ties (i.e., similarly or isotonically) by all items of a psychological…

THIS NOTE OFFERS A MEANS FOR ANALYZING THE DIFFERENCES IN THE LEVELS OF TWO RATIOS OVER A PERIOD OF TIME WHEN THOSE RATIOS COME FROM EMPIRICAL DATA. THE PARTICULAR EXAMPLE USED PERTAINS TO WHITE-NONWHITE PERSONAL INCOME DIFFERENTIALS OVER A TIME INTERVAL, BUT IT IS NOTED THAT THIS TYPE ANALYSIS CAN BE USED IN OTHER CIRCUMSTANCES. (HW)

The purpose of the research was to determine which community characteristics, among the 29 studied, were statistically most useful as predictors of per-pupil Federal aid to the 169 school districts of Connecticut. Three regression models were developed using community traits as predictors of Federal aid allocations. Community characteristics…

Discusses alternative forms of the desirability distribution for library materials and demonstrates the differences in short- and long-term patterns of frequency distribution in the circulation of library materials. Four figures, an appendix on statistical models for library circulation, and a reference list are included. (Author/JL)

A methodology is proposed to optimize the training classification performance of any suboptimal model. The method, referred to as univariate optimal discriminant analysis (UniODA), is illustrated through application to a two-group logistic regression analysis with 12 empirical examples. Maximizing percentage accuracy in classification is…

In confirmatory factor analysis (CFA), models exist that are admissible and demonstrably identified where the factor loading matrix from latent to manifest variables is of less than full column rank, even though other models are not identified. Implications for CFA are discussed. (SLD)