Vocab Words Flashcards

(54 cards)

0
Q

Data Set Types

A

1) Cross-sectional
2) Panel
3) Cross-sectional time series
4) Time series

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
1
Q

Types of Variables

A

1) Independent Variable (X)
2) Dependent Variable (Y)
3) Confounding Variable (Z)
4) Intervening Variable (W)
5) Moderating Variable (P)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Panel Data Set

A

The same cases observed over multiple time periods

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Pure control vs placebo control

A

Pure control group gets nothing while the placebo group gets placebo

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Induced state failure

A

Internal validity problem in doing treatment where you don’t get desired state, or a different state than intended

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Unsuccessful administration of X

A

Internal validity problem with treatment. All individuals assigned to treatment don’t get X.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Diffusion of X

A

Internal validity problem with treatment. Some members of control group get X.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Criteria for establishing causality

A

Rule out!

1) Fluke
2) Spurious
3) Pre-existing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Time series data set

A

One entity observed over multiple time periods

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Spurious scenarios

A

1) Common cause

2) Correlated cause

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Demand effects

A

Internal validity problem with subjects. When subject tries to please. This can be solved with deception or by doing the study blind.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Internal validity problem with researcher or outsiders

A

1) Experimenters bias

2) Compensatory equalization

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Compensatory equalization

A

Internal validity problem with outsider or researcher when compensation is given to control group for lack of treatment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

3 parts of external validity

A

1) Subjects
2) Context: time and place
3) Circumstances under which X is administered

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Problems with X (external validity)

A

1) Artificial X
2) Unrealistic X
3) Limited realization of X
A) Limited range
B) Limited sample

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Limited realization of X: limited range

A

External validity problem with X. When you have a continuous X that you only test a small range of

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Experimental artifacts

A

Internal validity problem with treatment. Things thought to be irrelevant differences between treatment and control groups. This is built into design.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

A rich theory

A

Has many hypothesis you can generate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Experimenters bias

A

Internal validity problem with researcher. You are inclined to see what you expect. Do a double blind test to counter.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Spillover effects

A

Internal validity problem with subjects. When treatment group influence control groups dependent variable. The control group is contaminated.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Resentful demoralization

A

Internal validity problem with subjects. Control group is demoralized for not receiving X, can be countered by doing experiment blind.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Compensatory rivalry

A

Internal validity problem with subject. The central group tries extra hard to compete with treatment group. Doing blind test helps with this.

22
Q

Differential morality

A

Internal validity problem with subjects. People drop out of treatment and/or control group but not at random. This hurts the expectation of equivalence.

23
Q

Reason for X Y correlation

A

1) X had a causal effect on Y
2) Coincidence
3) Spuriousness
4) Correlation existed before X was introduced

24
Cross sectional time series
Multiple cases observed over multiple periods of time. Same variables though.
25
What makes an experiment
Control over who gets X Control over what X consists if and the circumstances if administration.
26
CEPO
Controlled Experiment Post-test Only. Matched by potential confounding variables and observed post test.
27
External validity
The degree to which one can generalize from 1) A study's subjects 2) Context 3) X 4) Circumstances in which X is administered to the real world
28
What effects internal validity
Architecture and design/research procedure
29
Internal validity
The degree to which one can make valued causal influences about the effect of X on Y
30
Static control group
Two groups, one treatment, one control, one time. X O O Only gives correlation
31
Causal heterogeneity
Identifying a moderating variable (P) that divides your cases into subject where CE is expected to vary
32
CREPO
Controlled Randomized Experiment Post-test Only. | Random selection of who gets X.
33
Artificial and unreal X
Experimental validity problem with X. Artificial is fake, wouldn't happen. Unrealistic is either too large or too small to generalize to real life.
34
Compound X
Internal validity problem with treatment. When X is more than one thing differing treatment and control groups by two or more X variables
35
Induced state treatments
Where your treatment induces your X
36
Factorial treatment design
You can construct many treatment conditions by simultaneously manipulating 2 or more variables
37
Assignment procedure in experiments
1) Matching 2) Random assignments 3) Random assignments with blocking
38
A general theory
Having a wide range of situations it can be applied to
39
Cross-sectional data set
Data characteristics that are gathered at one point in time and time is not available
40
Case
The entity being studied
41
Intercession History
Internal validity problem with treatment. When something unplanned happens with one group and not other, that could affect the Y variable.
42
Is the causal effect of X on Y observable
No, this is the fundamental problem of causal inference
43
CREPP
Controlled Randomized Experiment Pre Post-test: | Gives replication and gives evidence to rule out pre-existing conditions
44
Internal validity problem with Subjects
1) Too few subjects 2) Differential morality 3) Demand effects 4) Compensatory Rivalry 5) Resentful demoralization 6) Spillover effects
45
Internal validity problem with in-doing treatment
1) Compound X 2) Induced state failure 3) Unsuccessful admin of treatment 4) Diffusion of X 5) Intercession history 6) Experimental artifacts
46
Continuous X treatment
X that has many levels of treatment
47
What does treatment design entail
1) Continuous X 2) Pure control vs placebo control 3) Factorial treatment 4) Induced state
48
Design architecture
1) Assignment procedure 2) Number if groups 3) Number of observations on Y over time
49
Blocking
Grouping based on potential variables and/or pre-treatment of Y
50
Criteria for good theory
1) Logically coherent 2) Accurate (hypothesis supported by evidence) 3) General 4) Rich 5) Parsimonious
51
Theory
Simplified representation of a phenomena or set of, that emerges through deductive or inductive reasoning, explains and has descriptive components to generate hypothesis that is tested empirically through data.
52
Limited realization of X: Limited Sample
External validity problem with X. When Y could be categorized by lots of examples and yet you only use a few
53
External validity problems are usually based on
procedure, not architecture