Bookmark and Share

Advanced Research Design and Applied Data Analysis for Education and Social Science


Graduate School, Arts at Aarhus University


Description:

The course is primer in applied quantitative data analyses that emphasizes on the connection of data analyses and research design.

Aim:

This course aims to provide early stage PhD students with the skills to choose and apply state of the art quantitative social science research methods. The course will help to choose research designs and analysis methods which are frequently used in publications in international journals. It is primarily directed to students with the interest in a quantitative PhD-project in the early phase of their PhD studies, but is also directed to all other students who wish to improve their understanding of published quantitative social research.

Content:

Block 1: (Day 1 and 2)

Part A: Research Design in Quantitative Social Research

a. Challenges to descriptive and causal evidence in the social sciences

b. Specific problems of causal interpretation and generalizations in social sciences

c. Challenges to systematic development of social science knowledge (covers e.g. publication bias and p-hacking, replication and replicability, strengths and weaknesses of peer-review, meta-analysis)

Part B: Introduction to multivariate data analysis A: Regression models

This session repeats the basics of regression analysis, discusses assumptions of linear models.

Block 2 (Day 2)

Part A: Nonlinear and other relevant regression models such as binary, ordinal, multinomial or count data models.

Part B: Causal inference I (Day 2)

This part focusses on different strategies of addressing causal research questions with observational data. It will give an overview on the strengths and weaknesses of the different models and give short examples. The main methods to be covered are ‘propensity score matching’, ‘panel data analysis’,

Block 3: (Day 3 and 4)

Multilevel modelling

This block introduces and discusses different strategies for taking into account clustered data structure, for example students in schools, classrooms or countries. We will introduce possible solutions for quantitative analyses, in particular multilevel models and multilevel-mediation analyses.

Block 4: (Day 4)

Part A: ‘quasi experiments and instruments’,

Part B:  ‘selection models’ and ‘regression discontinuity design’.

Target group:

First year PhD students who want to do a quantitative PhD project

More advanced PhD students who want to improve their understanding of quantitative literature

The course requires a basic understanding of descriptive and inference statistics. If students have no prior knowledge of quantitative methods they can contact David Reimer/Felix Weiss in order to receive recommendations for preparatory reading material.

Language:

English

Form:

Seminar

ECTS:

4

Lecturers:

Felix Weiss fewe@edu.au.dk

David Reimer dare@edu.au.dk

Gregory Palardy: gregory.palardy@ucr.edu

Dates and time:

18 September 2017

2 October 2017

6-7 November 2017

Times: 9-12 and 13-16

Venue:

DPU, Campus Aarhus, Katrinebjerg, building 5335, room 091, Finlandsgade 21, 8200 Aarhus N

Application / registration:

Please apply via https://auws.au.dk/advancedresearchdesign no later than 14 September 2017.


Back

Course dates
18 September 2017 - 07 November 2017
Lecturer
Gregory Palardy and others
Place/Venue
Aarhus University
City
Aarhus N
ECTS
4 points
Link
https://auws.au.dk/advancedresearchdesign

If you have any questions about this site, please contact Danske Universiteter via mail: dkuni @ dkuni.dk