- Home
- ICDS
methodological and statistical support to help make a difference
ICDS provides complementary support in methodology and statistics to our research community, for both individual researchers and research groups, in order to get the best out of them. ICDS aims to further enhance the quality of both the research and how it is communicated.
At Your Service
for our VUB-UZ Brussel research community
Coordinator
Prof. Dr. Kurt Barbé
Statistical Consultants
Dr. Wilfried Cools (Jette)
Dr. Lara Stas (Etterbeek)
Supporting Researchers (BISI)
Prof. Dr. Kurt Barbé
Prof. Dr. Ronald Buyl
Dr. Sven Van Laere
Drs. Susanne Blotwijk
Dr. Iris Steenhout
Dr. Oscar Olarte
Drs. Hanif Shaikh
Drs. Camille Raets
Whom We Serve
Master Student, authorized by promoter
to get a push in the right direction
General guidance on methodology, statistics and its communication. For Humanities, the focus is on writing scholarship applications and articles. For Life Sciences & Medicine, the focus is on the master thesis.
Researcher
to get your study and analysis up and running
A look over your shoulders to ensure appropriate methodology and statistics and help its communication, possibly taking over parts of the data analysis.
Research Groups
to get fully dedicated support for your project
Full involvement in a project, for study design and/or data analysis, in optimizing the data collection and processing for recurrent studies and in bringing in the latest statistical techniques.
Committees
to help safeguard research quality and ethics
Reviewing and refereeing research protocols and proposals to promote efficient / effective studies, reducing costs for patient, lab animal, researcher.
How We Serve
Consultancy
asking the right questions the right way
Statistical and methodological advice on data analysis and data collection (eg., sample size), on communicating statistics and results, on writing protocols and research proposals.
Data Analysis
getting the relevant information out of it
A full collaboration implies taking our responsability for implementing (parts) of the analysis, communicating the results and responding to remarks made by reviewers.
Tutoring
learning to do things yourself
Workshops and seminars are provided to enhance the existing proficiency in statistical and methodological reasoning where needed.
Automation
getting the computer to do it for you
Parts of the data collection, data analysis and reporting can be automated to avoid repetitive work, errors and time loss, and to promote uniformity.
Financially supported by
Research board, University Medical Campus
Vice-rectorat Research Policy
Project based support by
Research groups Frailty in Aging, Pain in Motion & Rehabilitation: Ivan Bautmans, Jo Nijs & David Beckée with project Exercise therapy for osteoarthritis pain: how does it work?
Laboratory of Molecular and Cellular Therapy: Karine Breckpot (PI) with project SPARC - Smart design nanoParticles to Activate immune Responses against Cancer.
Department of Radiotherapy (UZ-B): Mark De Ridder (PI) with project Societal Benefit of Markerless Stereotactic Body Radiotherapy: A statistical Support based on Quantitative Imaging.
Resources
Workshops
= Sample Size Calculation with GPower =
Workshop to introduce the necessary concepts, covering power, effect size and type I/II errors, with small illustrative exercises to touch on relevant statistical reasoning. While focusing mainly on the comparison of 2 independent groups, several extensions are highlighted. [slides] [pdf].
Presentations
= Methodological & Statistical issues in research proposals =
Info session for experienced researchers to highlight the research aim and design as part of a research proposal, touching on types of research aims (confirmatory, exploratory, preparatory, techn(olog)ical) and different aspects of research design (quantity & quality of observation and generalization). [webpage] [pdf].
= Data Representation =
Short info session to help structure data in preparation of statistical analyses; highlighting errors, inconveniences, common problems and solutions. [webpage] [pdf].
Technical Notes
= Corona hospitalisation SIR model =
Covid-19 offers challenges for statisticians as well. A technical note addresses how to approximate the popular SIR-equations analytically, which after reparametrization allow for straightforward parameter estimation and prediction of hospitalisation sustainability. Extensions that add flexibility are highlighted which allow stable identification of the model with a limited time series. [pdf]
Please reference: Barbé, K., Blotwijk, S., & Cools, W. (2020) Data-driven epidemiological model to monitor the sustainability of hospital care, VUB Covid19 Technical Note No. ICDS043020.
Shiny Apps
all Shiny Apps are deployed on our own shiny server and are available online.
alternatively, the Apps can be run locally using the R package `shiny` and the code on GitHub
sample size simulation
interim analysis based on alpha spending for t-test and one-way ANOVA
run locally with runGitHub('simAlphaSpending','ICDS-VubUZ')
effect size calculation
eta squared effect size specification for two-way ANOVA and repeated measures ANOVA
run locally with runGitHub('effectSizes','ICDS-VubUZ')
didactical tool
sample sizes for one-sided t-test
run locally with runGitHub('shinyT','ICDS-VubUZ')