Inquiries and Data

Question 1

What is an Inquiry?

Answer 1

Inquiry formulates research questions within the context of our model. Can be causal or descriptive.

Question 2

What are three types of Inquiries?

Answer 2

Descriptive, Causal, Counterfactual

Question 3

What is a descriptive inquiry?

Answer 3

Descriptive: Describes a phenomenon (eg, how many eggs does a chicken lay?)

Question 4

What is a Causal Inquiry?

Answer 4

Asks about cause and effect relationships
(eg, how does age of a chicken affect how many eggs it lays?)

Question 5

What is a Counterfactual Inquiry?

Answer 5

Explores how something would be different under different conditions
(eg, how many eggs would a chicken lay if it was free range instead of captive?)

Question 6

What are the Elements of an Inquiry?

Answer 6

1. Units of analysis
2. Outcomes
3. Treatment conditions
4. Summary Functions

Question 7

What are Units of Analysis?

Answer 7

People, places, or things studied (e.g., individuals, schools).

Question 8

What are Population of units?

Answer 8

All possible units.

Question 9

What are treated units?

Answer 9

Units receiving the treatment.

Question 10

What are untreated units?

Answer 10

Units not receiving the treatment.

Question 11

What are complier units?

Answer 11

Units following the assigned treatment (taking it if assigned, not taking it if not assigned).

Question 12

What are treatment conditions?

Answer 12

What is being manipulated.
- Descriptive inquiries often focus on a single condition.
- Causal inquiries compare at least two conditions.

Question 13

What are summary functions?

Answer 13

How the inquiry is calculated (e.g., average, slope).

Question 14

What is the estimand?

Answer 14

Part of the Inquiry: The estimand represents the theoretical answer to the Inquiry (research question derived from our model)

Question 15

What is Data Strategy?

Answer 15

Data strategy focuses on measuring and operationalizing elements of our model and inquiry.

Question 16

What are three key elements of a Data Strategy?

Answer 16

• Sampling:How will you select the units (people, places, or things) that you will study?
• Treatment assignment: How will you assign treatments (if any) to the units in your sample?
• Measurement: How will you measure the outcomes (variables) that you are interested in?

Question 17

What is a dataset?

Answer 17

The data strategy (D) is the process of collecting data, and the dataset (d) is the resulting collection of data. When you apply a data strategy to the real world (m), you produce a dataset.

Question 18

What are 4 threats to Data Strategies?

Answer 18

1. Noncompliance
2. Attrition
3. Excludability Violations
4. Interferance

Question 19

What is Noncompliance and how do you address it?

Answer 19

This occurs when the units in your sample do not follow the treatment assignment that you gave them.

You can try to mitigate noncompliance by using strong encouragements to participate in the study

Question 20

What is Attrition and how do you address it?

Answer 20

This occurs when you lose data from your sample over time.

You can try to reduce attrition by improving your data collection procedures

Question 21

What is Excludability Violation and how do you address it?

Answer 21

This occurs when the assumptions that you make about how the variables in your study are related are not actually true.

You can try to avoid excludability violations by carefully considering the assumptions of your research design and by collecting data that will allow you to test those assumptions.

Question 22

What is Interference and how do you address it?

Answer 22

This occurs when the outcomes of the units in your sample are affected by the treatments or measurements that are given to other units in the sample.

You can try to avoid interference by using a design that controls for the effects of other units in the sample.

Question 23

In relation to Causal Inquiries, what is ATE?

Answer 23

Average Treatment Effect (ATE): The average difference in outcomes between treated and untreated units.

Question 24

In relation to Causal Inquiries, what is ATT?

Answer 24

Average Treatment Effect on the Treated (ATT): The average effect of treatment for those who actually received it.

Question 25

In relation to Causal Inquiries, what is ATU?

Answer 25

Average Treatment Effect on the Untreated (ATU): The average effect of treatment for those who did not receive it (impossible to observe directly).

Question 26

What is Random Sampling?

Answer 26

Every unit has an equal chance of being selected

(e.g., simple random sampling, stratified random sampling, cluster sampling, multistage sampling).

Question 27

What is Non-Random sampling?

Answer 27

Non-probability sampling is a method of selecting units from a population using a subjective method

(e.g., convenience sampling, purposive sampling, quota sampling, respondent-driven sampling).

Question 28

What is random (treatment) assignment?

Answer 28

Units randomly assigned to treatment or control groups.

Question 29

What is non random (treatment) assignment?

Answer 29

Units assigned based on non-random factors.

Question 30

What treatments are in two arm designs?

Answer 30

Two treatment conditions (e.g., treatment vs. control).

Question 31

What treatments are in multi-arm designs?

Answer 31

Units can receive one of multiple treatments.

Question 32

What treatments are in factorial designs?

Answer 32

Units can receive combinations of multiple treatments

Question 33

What treatments are in over time designs?

Answer 33

Treatments applied at different points in time.