897977

Course

Statistics for the behavioural sciences (runs annually)

Faculty

Laura Winther Balling, associate professor of experimental psycholinguistics, Department of Management, Society and Communication, CBS

Søren Feodor Nielsen, professor of statistics, Department of Finance, CBS

Course Coordinator

Laura Winther Balling

Prerequisites

The course is for anyone who is interested in doing quantitative research and analyse the results statistically. We assume no prior knowledge of statistics or mathematics, but we do assume that participants have worked seriously with the preparation for the course and are ready for a steep learning curve. Participants who have data from their own projects may bring them for discussion and analysis during the course, but this is not a requirement.

The starting point is questions and datasets from linguistics and experimental psychology, but a background in these fields is not assumed. The course is open for researchers from all fields, but will be of particular relevance for those who in some way study human behaviour due to our focus on mixed effects modelling.

The last three days of the course focus on multi-variate analyses and may be attended in isolation by participants who have completed the course Applied Quantitative Methods for Non-quantitative Doctoral Researchers in Organization and Management Studies http://ssl1.peoplexs.com/Peoplexs22/CandidatesPortalNoLogin/Vacancy.cfm?PortalID=15442&VacatureID=878445 or similar, provided the participants have familiarised themselves with the R software environment in advance (instructions on how to do this will be provided). There is no such prerequisite for participants that attend the whole course.

Participants should submit a 1-2 page summary of their PhD (or other research project) by November 20. The summary should be written so that it is understandable to readers who are unfamiliar with the research field. The summary should give a general idea of the aim of research project and the use of quantitative methods and analyses. We advise participants to read chapter 1 of Gries (2009) before preparing their project description.

A precondition for receiving the course diploma for the full 5 ECTS course is that the student attends the whole course; those choosing to attend the last three days will get a diploma for 3 ECTS if they attend all three days.

Aim

The aim of the course is to enable participants to conduct sound quantitative research and statistical analyses, both in their PhD-projects and in their further careers inside and outside of academia. We start with the basics of quantitative research design and statistical testing, but quickly move on to the type of multivariate methods that are necessary to analyse real datasets with many variables and complex dependencies. We focus on the practical applications of statistics and include a substantial amount of hands-on examples and exercises.

We use the open-source programme R which is a powerful tool that does not require an institutional license, but is freely available to all users.

Throughout the course, we link the methods covered to participants’ own research projects. To further strengthen this link, participants will have the chance to hand in a final report (maximally 10 pages) in which they show how the methods from the course could be used to address a research problem within their own research area. The final report must describe the problem the participant wants to examine, the type of conclusions the participant would like to draw, and the data collection and statistical analysis that is needed in order to support the conclusions. The deadline for submitting the report will be approximately 4 weeks after the course; more information will be provided at the beginning of the course. Thorough feedback will be provided, orally or in writing. Approval of the report will give the participant an additional 1 ECTS.

When registering, students need to decide whether to hand in a paper or not.

Course content

The course covers the following topics

1. Key aspects of quantitative research design, including research questions and hypotheses, sample and population, the types and roles of variables in the research design. We emphasise the importance of considering statistics from the earliest phases of research design.

2. Loading and manipulating data in the R environment prior to analysis.

3. Fundamentals of statistical testing: null and alternative hypotheses, test statistics and p-values.

4. Selected simple statistical tests, the exact choice depending on the participants’ needs and backgrounds as expressed in a project description to be handed in before the course (see below).

5. Multiple regression with a special emphasis on mixed-effects models which are ideal for research that involves multiple responses from multiple individuals.

Teaching style

Each morning and afternoon session consists of a lecture, with the possibility for questions and discussion, followed by exercises related to the topic of the lecture. Throughout the course, participants are encouraged to relate their learning to their own projects and to discuss their own research with each other and with the lecturers, who are both present throughout the course. There is a strong emphasis on practical applications and real-life data.

Lecture plan

Monday, December 11
10.00 - 12.30	Welcome Lecture 1 (LWB): Introduction to quantitative research design
12.30 - 13.30	Lunch
13.30 -16.00	Lecture 2 (SFN): Introduction to statistical thinking and testing hypotheses Exercises on data manipulation and visualization
Tuesday, December 12
9.00 - 12.00	Lecture 3 (LWB): Data types and data manipulation Workshop: Approaches in own research
12.00 - 13.00	Lunch
13.00 - 16.00	Lecture 4 (SFN): Test theory and basic statistical tests Exercises on basic statistical tests
Wednesday, December 13
9.00 - 12.00	Lecture 5 (SFN): Linear regression models Exercises on linear regression
12.00 - 13.00	Lunch
13.00 - 16.00	Lecture 6 (LWB): Example of mixed-effects analysis I Exercises on linear regression
Thursday, December 14
9.00 - 12.00	Lecture 7 (SFN): Generalized linear models Exercises on mixed-effects models
12.00 - 13.00	Lunch
13.00 - 16.00	Lecture 8 (LWB): Example of mixed-effects analysis II Exercises on mixed-effects models
Friday, December 15
9.00 - 12.00	Lecture 9 (SFN): Analysis methods for non-normal data Working with especially problematic and/or important issues from previous days
12.00 - 13.00	Lunch
13.00 - 15.00	Workshop: Analysis options for own research
15.00 - 16.00	Summing up & evaluation of whole course

Learning objectives

At the end of the course, participants will have achieved

- An understanding of quantitative research design

- Familiarity with R and relative ease of working with large sets of quantitative data

- The ability to explore quantitative data using appropriate graphics and descriptive statistics

- The ability to choose and apply statistical tests that are appropriate for the data and the problem to be solved.

- An understanding of the possibilities offered by regression designs and mixed-effects models and the ability to use these methods

- The ability to understand and evaluate the use of quantitative data and statistics by researchers and practitioners in their field(s).

- The ability to continue learning statistics.

Exam

N/A

Other

The course is primarily for PhD students but other researchers are welcome if there are places available. The course is not open to masters students.

Start date

11/12/2017

End date

15/12/2017

Level

PhD

ECTS

5 ECTS for the course, 1 additional ECTS for those participants who choose to hand in a final paper (see above). 3 ECTS for participant attending only the final three days of the course (see above).

Language

English

Course Literature

The main course book will be R. Harald Baayen, Analyzing Linguistic Data. A Practical Introduction to Statistics for R (Cambridge UP, 2008), and participants should bring a copy of this book. In addition, participants should read chapter 1 of Stephan Th. Gries’ Statistics for Linguistics with R. A Practical Introduction (De Gruyter, 2009), a copy of which will be provided after registration, along with details on what to focus on in the reading.Participants should bring a laptop with the following programmes and packages installed: - the R language for statistical computing, from www.r-project.org- the R Studio environment from www.rstudio.com which is the interface we recommend for working in R. There are short videos on the website to familiarise you with it. - The R-packages rms, languageR, Epi, ltm, psy, lmerTest, car (and packages that are automatically installed with these, if you check “install dependencies” in RStudio)Before the course, participants should familiarise themselves with the R language by typing all the code in the first two chapters in Baayen’s book in their own version of R/RStudio, and by working through the R code school that can be found here http://tryr.codeschool.com/levels/1/challenges/1.

Fee

DKK 6,500 for the full course, DKK 7,800 if also handing in a paper. DKK 3,900 DKK for the reduced course, DKK 5,200 if also handing in a paper. The fee covers the course and coffee/tea

Minimum number of participants

Maximum number of participants

Location

Copenhagen Business School
Dalgas Have 15
2000 Frederiksberg
Room: 1Ø.020 (1st floor)

Contact information

For practical enquiries, please contact
The PhD Support
Katja Høeg Tingleff
Tel.: +45 38 15 28 39
E-mail: kht.research@cbs.dk

For enquiries about the contents, please contact course coordinator Laura Winther Balling, lwb.msc@cbs.dk

Registration deadline

30/10/2017

When registering, students need to decide whether to hand in a paper or not.

Please note that your registration is binding after the registration deadline.

In case we receive more registrations for the course than we have places, the registrations will be prioritized in the the following order: Students from Doctoral School of Organisation and Management Studies (OMS), students from other CBS PhD schools, students from other institutions than CBS.

Top