Introduction

From a statistical point of view, the objectives of this course are to understand the relationship between variables of different types: when the two variables are quantitative, when a variable is qualitative and the other one is quantitative, when both variables are qualitative. Then, we go for multivariate exploratory analyses such as Principal Component Analysis (PCA), Correspondence Analysis (CA), and Multiple Correspondence Analysis (MCA).

From an experimental point of view, the objectives of the course are to know how to express properly a problem (an issue), how to translate this problem in terms of research question: in other words, what would be the experiment that would give me insights, in order to provide solutions for solving my problem. Within this course, the experiment should lead to a survey, to the construction of a multivariate measure that should differentiate the respondents of the survey.

From a practical point of view, I want my students to know how to use a software such as LimeSurvey, to know how to do reproducible research with Rstudio and Rmarkdown, to know how to analyse with R and the FactoMineR package.

In this course, examples are very perception oriented.

Three is the magic number

3 new powers: an introduction to textual analysis, the Q-methodology, the notion of reproducibility.

3 measures of the relationship between variables: correlation coefficient, etha square, chi-square (and you should know how to visualize that).

3 acronyms you should get tattooed on the fist: PCA, CA, MCA (clustering is just the gift wrap).

PCA for Principal Component Analysis

PCA for Principal Component Analysis

CA for Correspondence Analysis

CA for Correspondence Analysis

Syllabus

Course 1 (10/03)

  1. Introduction to the notion of measure
  2. Relationship between two quantitative variables
  3. Phrasing properly the research question : How would you define the overall quality of a product?
  4. Sociology is important
  5. Your data

Course 2 (17/03)

  1. Distance between two individuals
  2. Distance between two variables
  3. An introduction to Principal Component Analysis?

Course 3 (22/03)

  1. The notion of stimuli and open-ended question
  2. Who’s the man behind the beard?
  3. An introduction to the analysis of textual data (open-ended question) with the Chi-square test

Course 4 (23/03)

  1. The independence model with 2 qualitative variables
  2. Relationship between two quantitative variables
  3. An introduction to Correspondence Analysis
  4. Tell me what kind of music you are listening to and I will tell you who you are

Course 5 (02/04)

  1. An introduction to survey data: what kind of tea do you drink?
  2. What can I expect from the analysis of survey data?
  3. An introduction to Multiple Correspondence Analysis

Lab (06-07-09/04)

  1. An introduction to the Q-methodology: how would you define the notion of healthy lifestyle?
  2. Representing a qualitative variable
  3. Presentation of my first R markdown file
  4. Manipulating qualitative data