Early Literacy Measures for Improving Student Reading Achievement: Translating Research Into Practice

The importance of early literacy instruction and its role in later reading proficiency is well established; however, measures and procedures to screen and monitor proficiency in the area of early literacy are less well researched. The purpose of this study was to (a) examine the technical adequacy and validity of early curriculum-based literacy measures, Letter–Sound Correspondence, Onset Phoneme Identification, and Phoneme Segmentation, developed for use within the problem-solving model in the Minneapolis Public Schools and (b) describe the district-wide implementation of these measures. In general, these measures were found to have adequate reliability and validity, have moderate to moderately high correlations with criterion measures (oral reading, report cards), and be sensitive to growth across the school year. A case study of how these measures are used for screening and progress monitoring to improve reading achievement within 1 school is included. Limitations and future directions are also presented.     

Methodological quandaries in studying process and outcomes in peer assessment

Peer assessment is very various in its implementation. Six studies of peer assessment are reviewed, four of them in higher education. A literature review is followed by five empirical studies. Strengths and weaknesses of each study are considered and issues are raised. Variables in peer assessment needing further exploration are extricated – in sample definition, specification of type of peer assessment, organisational nature of the intervention, and the measures which are used to evaluate change. It is suggested that further experimental and quasi-experimental studies are necessary to contrast one or two of these variables at a time.     

RIM: A Random Item Mixture Model to Detect Differential Item Functioning

In this paper we present a new methodology for detecting differential item functioning (DIF). We introduce a DIF model, called the random item mixture (RIM), that is based on a Rasch model with random item difficulties (besides the common random person abilities). In addition, a mixture model is assumed for the item difficulties such that the items may belong to one of two classes: a DIF or a non-DIF class. The crucial difference between the DIF class and the non-DIF class is that the item difficulties in the DIF class may differ according to the observed person groups while they are equal across the person groups for the items from the non-DIF class. Statistical inference for the RIM is carried out in a Bayesian framework. The performance of the RIM is evaluated using a simulation study in which it is compared with traditional procedures, like the likelihood ratio test, the Mantel-Haenszel procedure and the standardized p -DIF procedure. In this comparison, the RIM performs better than the other methods. Finally, the usefulness of the model is also demonstrated on a real life data set.