Anne Anastasi Dissertation Award for Carel Peeters

17 Feb

The dissertation of dr. Carel Peeters titled Bayesian exploratory and confirmatory factor analysis, Perspectives on constrained-model selection, has been awarded the prestigious 2015 Anne Anastasi Dissertation Award by Division 5 (Quantitative and Qualitative Methods) of the American Psychological Association. This award recognizes an outstanding dissertation that was completed in the previous three years and addressed a topic in the quantitative areas of assessment, evaluation, measurement, and/or statistics. The dissertation stood out among a strong field of candidates for its impact and importance.

In this dissertation factor analytic modeling, the Bayesian approach, and constrained-model selection are intrinsically interwoven. Both data-reductional and data-analytic factor analysis are reformulated in the Bayesian framework from a constrained-model perspective. The dissertation provides a Bayesian data-reductional factor model which may be used for the selection of the intrinsic dimension of the latent vector. Moreover, in the reformulation of data-analytic factor analysis (FA), structure is imposed on the model through inequality constraints on the parameters of interest, rather than through standard exclusion restrictions.

Dr. Carel Peeters worked on his dissertation in the VICI project “Learning more from empirical data using prior knowledge” van prof. Herbert Hoijtink and was, at the end of this project, also supervised by professor Peter van der Heijden. He is currently working at VU University Medical Center Amsterdam.

Advertisements

10th International Conference on Multilevel Analysis

7 Nov

Submission and registration is open!

Conference

The 10th International Multilevel Conference is held April 9-10, 2015. The conference will be about all aspects of statistical multilevel analysis: innovative applications, theory, software, and methodology. The conference will be in an informal style, with much room for discussion. It will be held at Utrecht University, which is easy to reach by public transport. Invited speakers are professors Paras Metha, Tom Snijders and Joop Hox.

Pre-conference workshop Multilevel SEM

The conference is preceded by a one-day workshop on April 8 on “Multilevel Structural Equation Modeling in R” that is taught by professor Paras Mehta  (University of Houston). As there are no computers available in the conference centre, all participants are requested to bring their own laptop computer to participate in the computer lab of the one-day workshop. For participants who are not acquainted with the software program R there will be a workshop “working with R” by Gerko Vink on April 7.

Abstract submission

Deadlines for submitting an abstract: If you submit before the 1st of December (2014), you will be notified on the 14th of December if your abstract is accepted. If you submit before 31st of January (2015), you will be notified on the 14th of February if your abstract is accepted.

Young Researcher Award

The abstracts of four PhD students will be selected and these PhD students will be invited to present their work in a special symposium. The best presentation will be awarded during the closing ceremony with the ‘Best PhD Multilevel Presentation Award’ and he/she will get the conference fee reimbursed.

A website with further information on abstract submission and registration can be found at http://multilevel.fss.uu.nl/

With kind regards,

Mirjam Moerbeek, Rens van de Schoot and Leoniek Wijngaards-de Meij

Utrecht University

upcoming post-doctoral courses

6 Nov

Please note the following Postgraduate Methodology & Statistics Courses we offer at Utrecht University:

Fall 2014

  • Mplus: How to get started; a compact 1-day course to get introduced to Mplus, a software package that allows for structural equation modeling, multilevel modeling, mixture modeling and other advanced techniques. December 4, 2014

Spring 2015

  • Multilevel analysis with Mplus; a 3-day course on linear and logistic regression, longitudinal data and SEM. January 19 – 21, 2015
  • Applied multiple imputation using SPSS and R; a 3-day course on how to appropriately handle missing data using a modern and flexible imputation technique. March 4-6, 2015

For more information, registration, and other courses we offer, see: www.uu.nl/postgraduatems.

If you have any questions, you can send an email to: MS.Graduate.Program@uu.nl.

new article: Get Over It! A Multilevel Threshold Autoregressive Model for State-Dependent Affect Regulation

31 Oct

By Silvia De Haan-Rietdijk, John M. Gottman, Cindy S. Bergeman,Ellen L. Hamaker

Published in Psychometrika, 10.1007/s11336-014-9417-x

Intensive longitudinal data provide rich information, which is best captured when specialized models are used in the analysis. One of these models is the multilevel autoregressive model, which psychologists have applied successfully to study affect regulation as well as alcohol use. A limitation of this model is that the autoregressive parameter is treated as a fixed, trait-like property of a person. We argue that the autoregressive parameter may be state-dependent, for example, if the strength of affect regulation depends on the intensity of affect experienced. To allow such intra-individual variation, we propose a multilevel threshold autoregressive model. Using simulations, we show that this model can be used to detect state-dependent regulation with adequate power and Type I error. The potential of the new modeling approach is illustrated with two empirical applications that extend the basic model to address additional substantive research questions.

vacancy: assistant professor

24 Oct

You will be employed at the Department of Methods & Statistics of the Faculty of Social and Behavioural Sciences at Utrecht University. The Department of Methods & Statistics provides teaching in methods and statistics within all bachelor’s and master’s degree programmes of the Faculty of Social and Behavioural Sciences and University College Utrecht. The department offers the only two-year master’s degree programme exclusively centred on methods and statistics in the social, biomedical, and behavioural sciences in the Netherlands (English-taught). The department also advises staff and students with respect to their research activities.

The research of the Department of Methods & Statistics comprises the development of new techniques in areas such as methodology, survey research, mixed-methods research, qualitative research methods, multilevel modelling, longitudinal analyses and applied Bayesian statistics, and the application of these techniques to the study of phenomena in the social and behavioural sciences.

The Department of Methods and Statistics has a job opening for a full-time assistant professor for a period of five years. The position consists of a combination of teaching and research activities, requiring you to contribute to our research programme as stated in our ‘Mission Statement M&S‘.

Tasks

The position involves 60% teaching activities and 40% research time. The core tasks will include:

  • independently providing and contributing to the bachelor’s and master’s degree programmes within the curriculum of Methods and Statistics;
  • carrying out and supervising research within the scope of the research programme;
  • contributing to the acquiring of substantial external research funding for the benefit of the research programme.

Requirements

We are looking for someone who:

  • holds a PhD degree in the field of Methods and Statistics;
  • has experience in conducting scientific research in the field of the research programme of Methods and Statistics;
  • boasts scientific creativity and productivity, as evidenced by being published in recent (international) peer-reviewed publications;
  • is able to contribute to the acquiring of external research funding for the benefit of the research programme;
  • has experience in co-developing and co-coordinating of academic courses;
  • is able to provide inspiring lectures, has an eye for the modernisation of teaching methods, and can apply own research expertise to their teaching, as evidenced by positive teaching assessments;
  • holds a Basic Teaching Qualification (BKO) or is willing to acquire this qualification in the near future;
  • is effective and efficient, able to work within a team, and to think conceptually;
  • is willing to learn Dutch. We expect a relevant level of competence within two years of taking up the appointment.

Conditions of employment

We offer a temporary full-time position for one year. Upon a positive performance the appointment will be extended by four more years. Extension of the contract is conditional to an evaluation based on pre-set criteria and with the restriction that the Collective Labour Agreement in force at the time of the appointment allows for a fixed-term extension. Your salary will range from a minimum of €3,259 to a maximum of €5,070 per month based on a full-time appointment (i.e. salary scale 11/12 according to the Collective Labour Agreement for Dutch Universities). In addition, we offer attractive and flexible secondary terms of employment such as an annual holiday pay of 8%, an end-of-year bonus of 8.3% per year and partially paid parental leave. We also provide an attractive working environment and a vibrant research climate.

Organisation

Utrecht University has great ambitions for its teaching quality and study success rates. This also applies to its clear research profiles which are centred on four themes: Sustainability, Life Sciences, Dynamics of Youth, and Institutions. Utrecht University plays a prominent role in our society and contributes to finding the answers to topical and future societal issues.

The Faculty of Social and Behavioural Sciences is one of the leading faculties in Europe providing research and academic teaching in interdisciplinary social science, cultural anthropology, educational sciences, pedagogical sciences, psychology, and sociology. Research and teaching activities are concentrated in five areas: Behaviour in Social Contexts; Child and Adolescent Studies; Cognitive and Clinical Psychology; Education and Learning; and Methodology and Statistics.

More than 5,600 students are enrolled in a broad range of undergraduate and graduate programmes. The Faculty of Social and Behavioural Sciences has some 850 faculty and staff members, all providing their individual contribution to the training and education of young talent and to the research into and finding solutions for scientific and societal issues.

Additional information

Additional information about this position can be obtained from K. van Kats: K.vanKats@uu.nl

new article: Confirmatory Methods, or Huge Samples, Are Required to Obtain Power for the Evaluation of Theories

21 Oct

By Irene Klugkist, Freek Haarhuis, Laura Post and Floryt van Wesel
Appeared in Open Journal of Statistics, 2014, 4, 710-725

Abstract:
Experimental studies are usually designed with specific expectations about the results in mind. However, most researchers apply some form of omnibus test to test for any differences, with follow up tests like pairwise comparisons or simple effects analyses for further investigation of the effects. The power to find full support for the theory with such an exploratory approach which is usually based on multiple testing is, however, rather disappointing. With the simulations in this paper we showed that many of the common choices in hypothesis testing led to a severely underpowered form of theory evaluation. Furthermore, some less commonly used approaches were presented and a comparison of results in terms of power to find support for the theory was made. We concluded that confirmatory methods are required in the context of theory evaluation and that the scientific literature would benefit from a clearer distinction between confirmatory and exploratory findings. Also, we emphasis the importance of reporting all tests, significant or not, including the appropriate sample statistics like means and standard deviations. Another recommendation is related to the fact that researchers, when they discuss the conclusions of their own study, seem to underestimate the role of sampling variability. The execution of more replication studies in combination with proper reporting of all results provides insight in between study variability and the amount of chance findings.

upcoming postdoctoral courses

28 Sep

Please note the following Postgraduate Methodology & Statistics Courses we offer at Utrecht University during the fall semester of 2014:

  • Introduction to Meta-Analysis in the Social Sciences; a 3-day course in which you learn to carry out a meta-analytic review from A to Z; November 17 – 19, 2014
  • Mplus: How to get started; a compact 1-day course to get introduced to Mplus, a software package that allows for structural equation modeling, multilevel modeling, mixture modeling and other advanced techniques; Thursday December 4, 2014
  • Introduction to Analysis of Variance using R; an online course provided through Elevate; starting October 6,  for 3 weeks, with a workload of 14 hours per week

 

For more information, registration, and other courses we offer, see: www.uu.nl/postgraduatems.

If you have any questions, you can send an email to Marianne Geelhoed: MS.Graduate.Program@uu.nl.

 

 

10th International Conference on “Multilevel Analysis”

1 Jul

The 10th International Multilevel Conference is held April9-10, 2015. The conference will be about all aspects of statistical multilevel analysis: innovative applications, theory, software, and methodology. The conference will be in an informal style, with much room for discussion. It will be held at Utrecht University, which is easy to reach by public transport. Invited speakers are professors Paras Metha, Tom Snijders and Joop Hox.

 

Pre-conference workshop Multilevel SEM

The conference is preceded by a one-day workshop on April 8 on “Multilevel Structural Equation Modeling in R” that is taught by professor Paras Mehta  (University of Houston). As there are no computers available in the conference centre, all participants are requested to bring their own laptop computer to participate in the computer lab of the one-day workshop. For participants who are not acquainted with the software program R there will be a short workshop “working with R” by Gerko Vink onApril 7.

 

Abstract submission

Deadlines for submitting an abstract: If you submit before the 1st of December (2014), you will be notified on the 14th of December if your abstract is accepted. If you submit before 31st of January (2015), you will be notified on the 14th of February if your abstract is accepted.

 

Young Researcher Award

The abstracts of four PhD students will be selected and these PhD students will be invited to present their work in a special symposium. The best presentation will be awarded during the closing ceremony with the ‘Best PhD Multilevel Presentation Award’ and he/she will get the conference fee reimbursed.

 

A website with further information on abstract submission and registration can be found at http://multilevel.fss.uu.nl/ 

 

With kind regards,

 

Mirjam Moerbeek, Rens van de Schoot and Leoniek Wijngaards-de Meij

Utrecht University

new article: Re-estimating sample size in cluster randomised trials with active recruitment within clusters

5 Jun

By Sander van Scjie and Mirjam Moerbeek.

appeared in Statistics in Medicine (link)

Often only a limited number of clusters can be obtained in cluster randomised trials, although many potential participants can be recruited within each cluster. Thus, active recruitment is feasible within the clusters. To obtain an efficient sample size in a cluster randomised trial, the cluster level and individual level variance should be known before the study starts, but this is often not the case. We suggest using an internal pilot study design to address this problem of unknown variances. A pilot can be useful to re-estimate the variances and re-calculate the sample size during the trial. Using simulated data, it is shown that an initially low or high power can be adjusted using an internal pilot with the type I error rate remaining within an acceptable range. The intracluster correlation coefficient can be re-estimated with more precision, which has a positive effect on the sample size. We conclude that an internal pilot study design may be used if active recruitment is feasible within a limited number of clusters.

 

6th MPLUS user meeting. Registration open

22 Apr

The sixth Mplus users meeting is scheduled for the 21st of July 2014, just before beginning of the VI European Congress of Methodology (eam2014.fss.uu.nl). The program offers a wide range of new developments and solutions for typical Mplus problems, see below. For more information and registration go to mplus.fss.uu.nl/
———————————————————-

Program

09:00 – 09:30 registration with coffee and tea.

09:30 – 10:00 Key-note presentation – Errors and biases in Structural Equation Modelling – Jelte Wicherts (Tilburg University, The Netherlands)

10.00-10.30 Detecting and correcting for ERS using nonlinear SEM in Mplus (Meike Morren, VU University Amsterdam)

10.30-11.00 Decomposing State and Trait Variance using Structural Equation Modeling – Eva Zijlmans (Utrecht University, The Netherlands

11:15 – 11:45 A theory of causal effects – Rolf Steyer (Friedrich-Schiller-Universitaet Jena, Germany)

11:45 – 12:15 Analysis of causal effects with Mplus – Axel Mayer (Friedrich-Schiller-Universitaet Jena, Germany)

13:30 – 14:00: Bayesian Mixture Modeling – Sarah Depaoli (University of California)

14:00 – 14:30: Sample size requirements to accurately compare development between unbalanced subgroups in latent growth models – Mariëlle
Zondervan-Zwijnenburg (Utrecht University, The Netherlands)

14:30 – 15:00: Priors when to worry? Rens van de Schoot (Utrecht University, The Netherlands & North-West University, South-Africa)

15:15 – 15:45: Using Alignment optimization in establishing measurement invariance. An illustration with the value scale across countries – Jan Cieciuch & Eldad Davidov (University of Zurich, Switzerland)

15:45 – 16:15 Testing for measurement bias across large numbers of groups – Suzanne Jak (Utrecht University, The
Netherlands)

16:15– 16:45: Direct and indirect influences in the use of travel time: insights from UK national travel survey 2002-2010 – Kaveh Jahanshahi (University of Cambridge, UK)