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.     

In their very own words: authors and scholarly journal publishing

This paper presents the results of a major investigation by questionnaire of author attitudes towards scholarly publishing, which was funded by the Publisher's Association. Altogether, the views of nearly 4,000 authors were obtained and this paper concerns just those thousand or so authors who accepted the invitation to write freely about their own personal feelings and experiences. Using QSR N6 textual analysis software, their comments were processed and themes extracted. While the rest of the questionnaire, which was formed of closed‐question, tick‐box type questions, clearly must have had some impact on their thinking, the authors had considerable latitude in regard to what they could say. The paper provides a unique opportunity to hear their views directly. The pressing topics turned out to be (in rough order of concern): the peer‐review system, copyright, journal prices, alternative business models for journal publishing, big deals, Elsevier, electronic publishing and digital journals.     

Diversity in the Information Seeking Behaviour of the Virtual Scholar: Institutional Comparisons

The logs of four universities using the OhioLINK journal system were evaluated for a period of fifteen months using deep log analysis methods in order to compare and contrast the information seeking behaviour of their users. Large differences were found, especially between the research and teaching universities. Methodological problems associated with making the comparisons are discussed in some detail especially in terms of defining online sessions.