Friday Methods Sessions 2021

Each year StatUa organizes a series of talks/workshops/seminars on various methodological issues, in collaboration with FLAMES. Click on the titles below for more details and registration.

For more information, or if you want to contribute, contact Jarl Kampen.

February 26: Regression With Ordinal Variables (Online)

Prof. dr. Jarl Kampen (StatUa, UAntwerp)

The proper methods for analysis of ordinal variables, such as Likert-type items, have been subject of controversy for decades. Among other issues, the controversy concerns the problem of how one can prove that one measures what one is supposed to measure (Prytulac, 1975), whether or not measurement scales are relevant in statistical analysis (Gaito & Yokubinas, 1986), and how the scale of measurement is to be determined (Nunnally, 1967). Specific to the ordinal scale of measurement are the problems whether or not measurements on such a scale can be calibrated (Kampen & Swyngedouw, 2000), whether or not they can be interpreted as rough measures of continuous (underlying) variables (Yule, 1912; Kampen & Weeren 2017), and whether or not they may be modeled by parametric techniques (Boneau, 1961). 

In this Friday Methods Session, rather than choosing a side in the controversy, a systematic account of possible approaches to regression-type analysis involving independent and/or dependent ordinal variables is given. Topics that will be covered include ordinal dummy-coding, polychoric correlation, and (non-linear) ordinal response models. Special attention is given to the analysis of Likert scales. The approach will be pragmatic as well as scientifically justifiable.  Theory will be illustrated with practical examples. Students are assumed to be familiar with the basic principles of statistical analysis (hypothesis testing, correlation, simple and multiple regression, ANOVA). 

March 26: Text mining (online)

Dr. Koen Plevoets (UGent)

This session offers a comprehensive overview of the field of text mining. The focus is on tools, and topics to be discussed are stemming, tokenization, part-of-speech tagging and lemmatization, keyword extraction and collocations, (hidden) Markov models, text classifications and information retrieval, sentiment analysis and topic modelling. All tools will be exemplified in R, so a basic knowledge of the R language is necessary.

April 30: Statistical principles for clinical trials

Dr. Ella Roelant (StatUa, UAntwerp)

During this afternoon session the E9 Efficacy guideline ‘Statistical principles for Clinical Trials’ issued by ICH (International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use) will be discussed. Important statistical principles necessary to take into consideration when designing, conducting and analysing your clinical trial will be reviewed. Topics covered are primary and secondary variables, blinding, randomisation, type of comparison (superiority, equivalence, non-inferiority), interim analysis and early stopping, data analysis considerations (e. g. analysis sets, subgroups). Students are assumed to be familiar with basic principles of statistical testing.

You can register through this link

Due to the Covid situation, the event takes place online. A guest link for Blackboard collaborate will be sent to registered participants.

May 28 : Searching text patterns through regular expressions in R (online)

The analysis of large and complex datasets often starts with retrieving the records of interest, using a query that is based on a text pattern. Starting from a large database with student records, you may have to retrieve the students whose student-ID starts with s2016. Or you may need to retrieve files from which the filename starts with “results”, followed by a date, and ending with the extension “.txt”. And in these retrieved files, you may have to change the extension “.txt” into “.csv”.

This workshop illustrates functions in R that allow you to work with text data, including grep(), sub(), gsub() and strsplit(). Most of these functionalities are not unique to R – most of them are also found in other programming languages including perl, awk and Python.

An important notion in working with text is the concept of regular expressions. Regular expressions are a textual syntax for representing patterns for matching text – allowing to express patterns in character values. These patterns can then be used to extract parts of the dataset or modify these character values.

No knowledge in statistics is needed. Participants need to have the latest version of R and Rstudion installed on their laptop, and have some familiarity with R - if you have no idea what the following comands mean, the course is too advanced for you:

mydata <- read.table (“c:/temp/rawdata.txt”,header=T, dec=”,”)

Course format :
Due to the Covid situation, the event takes place online. A guest link for Blackboard collaborate will be sent to registered participants

Teacher :
prof. dr. Erik Fransen.​

June 25 : Research design in observational explanatory (causal) research - REGISTRATIONS OPEN

Prof. dr. Em. Joost Weyler

One of the (many) dogma’s in current (medical) scientific research is that the golden standard for causal research is the Randomised Controlled Trial (RCT). The theoretical design behind this type of study however is based on a functional relationship describing future occurrence (incidence, prevalence) of an outcome as a function of the intervention alternatives and therefore of a prognostic nature. As a consequence this design is in general not suited for appropriate assessment of the relevant exposure to the suspected cause nor for appropriate control for bias due to confounding.In this workshop we will approach observational explanatory research defining the theoretical design as a functional (causal) relationship between current occurrence (incidence or prevalence) of the outcome to be explained and the antecedents of that outcome: the antecedent suspected to be a cause, the antecedents that confound the crude relationship and the antecedents that determine the strength of the causal relationship.

You can register through this link