Module 2: Learning Objectives

Question 1

List 4 goals of ideal sampling design

Answer 1

1) All sampling units selectable
2) All samples possible
3) Selection is unbiased
4) Selection is independent

Question 2

Define bias

Answer 2

Over/under estimate of some value from an average sample compared to stat population

Question 3

Define sampling independance

Answer 3

When selection of one sampling unit does not influence another’s probability of being picked

Question 4

What is the main goal of observational studies?

Answer 4

Characterize something about statistical population

Question 5

Define response variable

Answer 5

Response we are interested in (ex. lung cancer)

Question 6

Define explanatory variable

Answer 6

Factor we investigate (ex. smoking)

Question 7

Define cofounding variables

Answer 7

Unobserved variables that affect response variable (ex. age)

Question 8

Define spurious

Answer 8

When relationship between response and explanatory is driven by cofounding

Question 9

Can you find causation between response and explanatory un observational studies?

Answer 9

No just association.

Question 10

Define observational study

Answer 10

Researcher does not have any control over explanatory variable

Question 11

List the five types of observational studies

Answer 11

1) Simple random survey
2) Stratified survey
3) Cluster survey
4) Case-control survey
5) Cohort Survey

Question 12

Define simple random survey

Answer 12

Identify each sampling unit in stat population than randomly select (ex. self employed or not)

Question 13

Define stratified survey

Answer 13

Used when subgroups in stat population (ex. different providences) break into stratas and sample stratas

Question 14

Define cluster survey

Answer 14

Observes groups in one or two stage designs (ex. one stage is school, two stage is 30 students in school)

Question 15

Define case-control survey

Answer 15

Compares data between two groups (ex. case is people with heart conditions, control is not

Question 16

Define cohort survey

Answer 16

Follows sampling units over time to observe outcome (did they get heart disease)

Question 17

Define retrospective

Answer 17

Outcome is known (ex. case-control)

Question 18

Define prosepctive

Answer 18

Outcome is not known (ex. cohort)

Question 19

Define cross-sectional

Answer 19

Response variable at specific time (ex. after flu shot)

Question 20

Define longitudinal

Answer 20

Response variable at multiple times (ex. after flu shot and after a week)

Question 21

What is an experimental study?

Answer 21

Study the effect of one or more manipulated variable on one or two response variables?

Question 22

What is a factor?

Answer 22

A manipulated variable

Question 23

Two differences between experimental and observational studies

Answer 23

1) Explanatory variable is manipulated by researcher

2) sampling units are randomly assigned to each level in factor

Question 24

What is replication?

What is pseudo replication?

Answer 24

Idea that experiment is repeatable

Error in design of experiment where observation units are studied instead of sampling

Question 25

List the 5 common experimental treatment designs and what they are

Answer 25

1) Control treatment: to compare everything against
2) Blocking: control variation (ex. sidewalk and driveway when using chemicals on lawn)
3) Blinded: people don’t know what treatment they are having
4) Placebo: fake medical trials
5) Sham treatment: (focuses on delivery)

Question 26

Levels vs Factors

Answer 26

Factor is drug

Levels is sub of drugs