Links

F. Husson website





Book Exploratory Multivariate Analysis Using R

Outline

Introduction

Principal Component Analysis

Correspondence Analysis

Multiple Correspondence Analysis

Clustering

Multiple Factor Analysis

To conclude

Forum

Quiz 2 on Mutiple Correspondence Analysis: Visualizing the point cloud of individuals

For each question, tick the correct answer or answers.

Q1) The distance between the points representing the i-th and i’-th individuals:
is smaller the more the two individuals share the same categories
is larger the less the two individuals share the same categories
is larger the more the two individuals are in different categories, which are common categories
is larger the more the two individuals are in different categories, which are rare categories
an individual who is mostly in common categories will be close to the center of gravity
an individual who is mostly in rare categories will be close to the center of gravity

Q2) For a table with J variables and a total of K categories, the total inertia of the point cloud of individuals
only depends on the number of variables and number of categories
is large if there are a lot of rare categories
is large if there are few rare categories
is large if there are many individuals

Q3) In the individuals plot, we can project each category of each variable as the barycenter of the individuals in that category;
we can therefore interpret the individuals plot using the positions of the categories
an individual close to a category must be in that category
a category can have a larger (absolute) coordinate value than all individuals for a given axis

Q4) We surveyed 1000 people on 10 questions, each with three possible responses. The data, put in a table T with 1000 rows and 10 columns, is then subjected to an MCA. What is the total inertia of the point cloud of individuals?
10
9
2
30

Q5) Suppose we calculate the square of the correlation ratio between a factor (the variable of the coordinate values of individuals in a certain dimension) and a qualitative variable. Which of the following is/are true?
if the correlation ratio is close to 1, the qualitative variable is highly associated with that axis and characterizes it well
if the correlation ratio is close to 1, all categories of that qualitative variable have the same coordinate value
if the correlation ratio is close to 0, all categories of that qualitative variable are in the center of the cloud
a variable with 2 categories can have a correlation ratio close to 1 with two axes
a variable with 3 categories can have a correlation ratio close to 1 with two axes

Score =
Correct answers: