FACULTY OF SOCIAL AND BEHAVIOURAL SCIENCES
Tilburg University is a modern, specialized university. The teaching and research of the Faculty of Social and Behavioural Sciences are organized around the themes of Health, Organization, and Relations between State, Citizen, and Society. The faculty’s inspiring working environment challenges its workers to realize their ambitions; involvement and cooperation are essential to achieve this.

DEPARTMENT OF METHODOLOGY AND STATISTICS
The Department of Methodology and Statistics is responsible for teaching methodology and statistics to students of the Faculty of Social and Behavioural Sciences. Central in its research program are the study, development, and application of latent variable models. Among these models are latent class models, item response models, and linear structural equation models. Other research themes are the analysis of missing data and outliers in test and questionnaire data, and response bias in attitude measurement in survey research. The Department’s research program is part of the national graduate school Interuniversity Graduate School
for Psychometrics and Sociometrics (IOPS). At present, more than 10 PhD students work in the department’s research program.

We have a vacancy for a PhD–student (M/F), 0.8-1.0 fte, for a period of 1+3 years

Job number 400.06.087

TITLE PROJECT
Person-misfit in item response models, explained by means of nonparametric and multilevel logistic regression models

PHD PROJECT
Multi-item tests and questionnaires are used in the social and economic sciences to assess characteristics of individuals. An important tool for assessment of individual response behavior is the person response function (PRF). An individual’s PRF should decrease when items become more difficult. Estimated PRFs may show that for some individuals their response function is
not decreasing. In achievement tests (e.g., educational tests) these violations may be due to guessing, lack of concentration, and test anxiety, and in personality (psychology), attitude (sociology), and preference (marketing) testing to response inconsistencies, multidimensionality and lack of traitedness.

Multilevel analysis will be used to model individual differences in PRFs. In particular, multilevel analysis of longitudinal data has to be adapted to model nonlinear functions for describing the responses of individuals to a set of items in test. In the context of PRFs, in the multilevel model the items replace the measurements at different points in time. The multilevel model also has to contain explanatory variables that help to pinpoint such causes of deviant behavior. The project requires knowledge of statistical modeling (multilevel models, item response models) and programming skills but also aims at the analysis of several real data sets to investigate the practical usefulness of the proposed methods.

A full description of the PhD proposal that was financed by NWO can be obtained upon request (use one of the email addresses of the contact persons mentioned below).

REQUIREMENTS
Applications are invited for an appointment as a PhD student in methodology and statistics at the department of Methodology and Statistics. As a PhD student you will do a literature search, develop statistical and psychometric methods, and apply these methods to meaningful data sets. You are required to present papers about you research at international conferences and write papers for international journals to be included in your PhD dissertation.

We are looking for a strong candidate with an outspoken interest in methodology and (applied) statistics, psychometrics, econometrics, and in analysis of data from tests and multi-item questionnaires. You have a doctoral or master degree in the social or behavioral sciences, with a special interest in methodology and (applied) statistics, and/or psychometrics, or a degree in
econometrics or another comparable area. English has to be mastered at a level that is sufficient for reading and writing and also presenting papers at conferences. Knowledge of Dutch language is an advantage but not a prerequisite. Moreover, the candidate is ambitious, able to work independently but willing to cooperate with the colleagues in the group.

TERMS OF EMPLOYMENT
Tilburg University is among the top of the Dutch employers and has an excellent policy concerning terms of employment. The job is initially for one year, but the purpose is to extend the contract with three years, depending on the performance in the first year. The gross salary starts with € 1956,- (full-time) per month in the first year, up to € 2502,- in the last year.

INFORMATION AND APPLICATION
More information about Tilburg University can be found at http://www.tilburguniversity.nl/
You can also contact the initiators of this project, Prof. Dr. K. Sijtsma (tel. 013-4663222 or 013-4662544, secretary; email: k.sijtsma[ at ]uvt.nl), Dr. W. H. M. Emons (tel. 013-4662397; email: w.h.m.emons[ at ]uvt.nl) or Dr. M. A. L. M. van Assen (tel. 013-4662362; email: m.a.l.m.vanassen[ at ]uvt.nl).

The closing date for applications is July 20, 2007. Your application, including a curriculum vitae should be submitted to drs. M. van Ieperen, managing director of FSW, Tilburg University, PO Box 90153, 5000 LE, Tilburg, the Netherlands, or: pz-fsw[ at ]uvt.nl. (please print job number 4000.07.05 on the envelop).