Week Four

Question 1

Demonstrate that 2 variables are
related or change together.

Answer 1

Association

Question 2

Determine that the relationship it is
not due to variation in a third

Answer 2

Non-spuriousness

Question 3

Demonstrate that 1 variable
occurs before another variable.

Answer 3

Temporal Ordering

Question 4

Correlation describes a statistical association
between two or more variables, meaning they
tend to change together.
Spuriousness
 When an apparent relationship between two
variables is the result of third variable
(confounding variable) influencing both or
one of them.
 For instance, ice cream sales are
correlated with crime rates.
 Best ways to assess spuriousness: use
control variables, direct observation,
longitudinal design.

Answer 4

Non-Spuriousness

Question 5

A variable that is held constant so that the relationship between
two (or more) other variables can be assessed.
Ways to show that the dependent variable changes due to the
independent variable.

Answer 5

Control Variable

Question 6

Data collected from at one point in time.

Answer 6

Cross-Sectional

Question 7

Data collected over multiple points in time

Answer 7

Longitudinal

Question 8

 Longitudinal study in which data are collected from the same
subjects at multiple time points.
 Observes same set of individuals changing over time as
opposed to groups.

Answer 8

Panel design (fixed sample)

Question 9

 Type of longitudinal study design in which data are collected
from different members within a particular subpopulation (cohort)
at multiple time points.

Answer 9

Cohort design (AKA event-based)

Question 10

Relatively inexpensive
No need to recontact subjects
Can be quickly implemented to address current events
and hot-button issues

Answer 10

Cross-Sectional research pros

Question 11

Cannot assess causal ordering

Answer 11

Cross-Sectional research cons

Question 12

Can assess causal ordering and change over time

Answer 12

Longitudinal Research Pros

Question 13

Attrition, or the loss of sample members over time, usually
to death or dropout, in panel studies
Costs are higher.

Answer 13

Longitudinal Research Cons

Question 14

When a relationship has already been established, but a
third variable explaining the relationship has not been
discovered
For instance,
X—>Y
How does “Z” influence this relationship?
X->Z->Y

Answer 14

Introduce new relationships

Question 15

What is the order of the relationship?
“Flip relationship on it’s head”
Does the independent variable cause the dependent variable, or
is it the other way around?
Often called “selection effects”

Answer 15

Reverse causation