Statistics for Research, with Applications in Marketing, Management and Innovation


April 16-20, 2018

​5 ECTS credits

Goal:
The goal of this course is to provide the course participants with knowledge of certain data analytic and statistical methodologies that are useful in several research areas. Both univariate, bivariate and multivariate methods are presented in the course, including aspects such as how data and statistical results should be reported in research papers.

Content:
The course “Statistics for Research, with Applications in Marketing, Management and Innovation" aims to introduce a wide variety of statistical methods that are necessary to be familiar with for anyone confronting empirical research. It covers both univariate, bivariate and selected multivariate methods, with the emphasis on how to select the appropriate method, depending on the data and the research objective, how to interpret the method’s results, and how to present the results in a research paper, especially in graphical and/or tabular form.

Examples of data will be drawn from the fields of marketing, management and innovation, so the course would be useful to a wide range of researchers. Real data will be analysed, results will be given and interpreted, and it will be shown how to report the results.

Generally, the statistical programming language R will be used, but other packages such as SPSS (commercial), XLSTAT (commercial add-on to Excel) and JASP (freeware: “A Fresh Way to do Statistics”) will be demonstrated as well and compared to R. Not knowing R will be no hindrance to following the course, but participants should have some knowledge of at least one statistical package.

Learning outcomes:
Based on this course the student should have
 
Knowledge

  • of a set of statistical methods that are useful to their research;
  • of the differences in approaches to a data set, and which analytical approach is the most appropriate: univariate, bivariate or multivariate treatment of a particular data set, or a combination of treatments;
  • of various statistical methods such as analysis of variance, regression analysis, generalized linear modelling, principal component analysis, correspondence analysis, structural equation modelling and network analysis, specifically how they are defined and implemented in practice;
  • of the difference between parametric and non-parametric statistical methods;
  • of modern resampling methods for hypothesis testing, such as permutation tests and bootstrapping;
  • of different statistical packages

Skills

  • to be able to recognise their needs when analysing empirical data and select the most appropriate statistical method for the research objective;
  • to be able to interpret the statistical results and express these verbally;
  • to be able to produce tabular and graphical output from the chosen methods, including just the essential results.

General competence

  • to dominate the statistical methods important to the student's research;
  • to understand the usefulness of the quantitative approach to research;
  • to be familiar with at least one statistical package and be introduced to the R language.

Course prerequisites:
Some basic statistical knowledge is required, for example knowledge of basic statistics such as mean, variance, standard deviation, correlation and simple linear regression. It is highly recommended as well to have followed the course Applied Multivariate Methods prior to this one. Familiarity with R is not a prerequisite. In case you wish to follow the demonstrations of the R language and/or JASP on your own computer, downloading and installing R (http://cran.r-project.org), Rstudio (http://rstudio.org) and JASP (https://jasp-stats.org/) prior to the course is strongly suggested. Participants are encouraged to play around with the software before the course in order to get a feeling for it. There are many video tutorials online, for example for R: XXXX and for JASP: XXXX .

Reading list:
A copy of the companion workshop notes will be provided to the participants:
 Greenacre, M. and Nenadić, O. (2015): “Mathematical Tools for Social Scientists: An Introduction with R”, forthcoming.

Optional literature:
 Sydsæter, K., Hammond, P. and Strom, A. (2012): “Essential Mathematics for Economic Analysis”, 4th ed., Pearson.
 Sydsæter, K., Hammond, P., Seierstad, A. and Strom, A. (2008): “Further Mathematics for Economic Analysis”, 2nd ed., Pearson.
 Dalgaard, P. (2008): “Introductory Statistics with R”, Springer.

Preliminary timetable

Monday 16.4.2018

9.00 – 12.00
Univariate and bivariate statistics

13.00 –16:00

  • Univariate summary measures: the difference between standard deviation and standard error
  • Intervals of dispersion and onfidence intervals; bootstrapping
  • Reporting and graphing univariate results
  • Bivariate summary measures 
  • Parametric vs. nonparametric correlation
  • Tests of difference; permuation tests

Tuesday 17.4.2018

9.00 – 12.00
Multivariate analysis 1

13.00 –16:00

  • Distinction between functional (supervised learning) and structural (unsupervised learning) methods
  • Regression analysis and the general linear model (including ANOVA)
  • Generalized linear models (including Poisson regression and logistic regression)

 
Wednesday 18.4.2018

9.00 – 12.00  Multivariate analysis 2

13.00 –16:00

  • Forming groups of multivariate observations: the concept of distance
  • Hierarchical cluster analysis for small to medium data sets
  • Nonhierarchical cluster analysis for large data sets
  • Interpretation of clusters
  • Visualization of clusters
     

Thursday 19.4.2018

9.00 – 12.00 Multivariate analysis 3

13.00 –16:00

  • Inferring dimensions in multivariate observations: the concept of dimension reduction
  • Multidimensional scaling
  • Principal component analysis
  • Correspondence analysis
  • Multiple correspondence analysis
  • Visualization of results

Friday 20.4.2018

9.00 – 12.00 Modern statistical methods
 
13.00 –16:00

  • Generalized additive modelling
  • Regression and classification trees
  • Recommender systems
  • Stuctural equation modelling
  • Network analysis
  • General summary and discussion 

Exam and evaluation:
Participation in lectures and an application of a method taught in the course. Paper graded: pass/non pass.

Language of education:
English

Course presenters:
Michael Greenacre is professor of Statistics at the Universitat Pompeu Fabra, Barcelona, specializing in multivariate data analysis, principally in the social and environmental sciences. Apart from more than 60 published articles in international journals (see http://www.researcherid.com/rid/D-7401-2013) he has written five books on correspondence analysis and related methods and co-edited four books (with Jörg Blasius) on data visualization. He has given short courses in 15 countries in Europe, north and south America, Africa and Australia.

Oleg Nenadić is presently Akademischer Rat at the University of Göttingen and has been a visiting professor at the universities of Göttingen and Erfurt. His interest lies in Applied Statistics across various disciplines such as Economics, Social Sciences and Biology, and his main focus is on Computational Statistics with the statistical software environment R (http://www.r-project.org). He is author of a number of R-packages and has given workshops on R on four continents.


Application form (Pdf)

Deadline for registration:
March 5, 2018.