BiVariate Analysis

Question 1

Which resources can be used to explore a pair of two nominal/categorical variables?

Answer 1

Paneled Bar Graph
Chi-Square Test
Cross-Tabs
Paneled Pie Graph
Clustered Bar Graph

Question 2

Which resources can be used to explore the connection between a Scale (Metric) variable and a Categorical/Nominal one?

Answer 2

Dependent Samples Means t-Test
Mean Graph
ANOVA
Paneled Histogram
Post - Hoc ANOVA Tests
Box-Plot Graph
Independent Samples Means t-Test
(Descriptives) Explore Procedure

Question 3

Which resources can be used to explore a pair of two Scale variables?:

Answer 3

Scatter Diagram

Linear Correlation Coefficient

Question 4

ANOVA procedure. It is usefull to explore the relationship between a scale variable and a categorical variable

Answer 4

TRUE

Question 5

ANOVA test is sometimes followed by a Post - Hoc Test (Bonferroni, LSD….)

Answer 5

TRUE

Question 6

The Null Hypothesis is that the mean of the scale variable is the same in each category (of the categorical variable)

Answer 6

TRUE

Question 7

ANOVA is the usual inferential test that follows an explore procedure

Answer 7

TRUE

Question 8

ANOVA is COMMONLY used ONLY when the scale variable is the dependent variable and the categorical one is suppose to be the independent/factor

Answer 8

TRUE

Question 9

ANOVA is the typical test we sometime use jointly with a Box-Plot Graph

Answer 9

TRUE

Question 10

A correlation coefficient with a value of 0.8 and a test p-value of 0.15 means:

Answer 10

A strong but statistically NON-significant positive correlation

Question 11

What type of GRAPH should I use to explore insurance price (in €) by time as customer (in years)?

Answer 11

Scatter

Question 12

Which is the MOST IMPORTANT LIMITATION of bivariate analysis when we are interested in the exploration of variables relationships:

Answer 12

We could be missing some covariates (third variables) of interest to understand relationships

Question 13

In the context of bivariate analysis, a Chi – Square test is:

Answer 13

An statistical way of checking bivariate relationship for two categorical variables when using sample data
A statistical option of SPSS Frequencies procedure
An option for a Clustered Bar Graph

Question 14

Which of the following features of an scale variable can be initially explored/represented using a Box – Plot graph?

Answer 14

Median
Presence of outliers
Dispersion

Question 15

We take a sample of 1000 people and we statistically compare the average salary (in weekly $) by being self employed (Yes/No). What statistical test should I use?

Answer 15

T- tests for independent samples

Question 16

SIMPSON’S PARADX paradox is

Answer 16

About the risk of omitting confounding variables when exploring a causal relationship