Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

. > pols > Flashcards

pols Flashcards

1
Q

Measurement Error

A

Any type of error that’s introduced when measuring a variable that does not capture the concept trying to be measured

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

Validity

A

Mismatch between the measure and the concept as you it was defined.

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

Processing Error

A

Any error introduced between the response given by a survey respondent and what you
record

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

three major types of representational errors in surveys

A

coverage, sampling, non-response

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

Coverage Error

A

Errors due to a mismatch between the target population and the sampling frame.

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

Sampling Error

A

Sampling errors are errors introduced by collecting a sample rather than studying the full
population

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

Non-Response Error

A

Once individuals are sampled, those that don’t respond but were sampled are non-response error

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

Individuals without internet access will be a form of —- in an online survey

A

coverage error

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

—- can often be decreased by recontacting those who did not answer the
survey the first time you contacted them

A

Non-response error

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

Sensitive survey questions tend to see — levels of item non-response

A

higher

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

A census is any survey of —

A

the entire population

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

A census survey does not have sampling error (T/F)

A

true

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

Double Barreled Questions (Definition)

A

Questions in surveys that require respondents to report two (or more) attitudes at once

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

Unbalanced/Leading Questions (Definition)

A

These are questions that push (or lead) people towards particular responses

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

If measures of two variables X and Y are biased, it can result in changes to both the direction and strength of their correlation

A

true

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

A correlation of r = -1 indicates a weaker relationship between variables X and Y than a
correlation of r = 1

A

false

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

In a scatter plot showing a strong correlation between two variables X and Y, the data
points will generally align more closely with a straight line

A

true

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

Correlations can only be calculated for — variables.

A

quantitative

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

explain the fundamental
problem of causal inference

A

You can’t observe the unobservable potential outcome.

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

For each individual, we can observe multiple potential outcomes at the same time

A

false

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

The concept of —– is used to refer to the outcomes that would have occurred under different treatments

A

counterfactuals

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

The existence of —- implies the potential outcomes for individuals
with different treatment levels (X) are inherently unequal

A

confounding variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Establishing a correlation between two variables X and Y is adequate to confirm there is a causal link between X and Y, even if you don’t know the causal direction

A

false

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

In — experiments, participants can’t select whether they would like to be
in the treatment or control groups

A

between-groups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

—- in experiments ensures that there is no association between confounding variables and the treatment being studied

A

Randomization

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

— experiments compare the outcomes of the same individuals before
and after receiving a treatment

A

Within-individuals

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

One downside of running — experiments is that participants tend to be very aware they
are participating in a research study

A

lab

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

A researcher wants to study the causal effects of sex (i.e., whether someone is Male or
Female) on support for over-the-counter birth control availability. It is not possible to
randomly assign sex in an experiment. What is a possible way of testing the causal effects
of sex on support for OTC birth control in a between-groups experiment? (1-3 sentences).

A

Priming experiment.
We could ask individuals in a treatment group: “As a man/woman, do you support or
oppose OTC birth control?”
We could ask individuals in a control group: “Do you support or oppose OTC birth
control.”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

In linear regression, the — involves minimizing the sum of
the squared distances between the observed data and predicted values

A

ordinary least squares method

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

In simple linear regression, the slope terms represent the

A

predicted change in Y for a
one-unit increase in X

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

In bivariate linear regression, the sign of the slope of X on Y can be different from the
sign of the correlation between X and Y.

A

False

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

In multiple linear regression, the sign of the slope of X on Y can be different from the
sign of the correlation between X and Y

A

True

33
Q

In multiple regression, — reflect the isolated impact of each independent variable on
the dependent variable, controlling for confounding variables in the model.

A

slopes

34
Q

Categorical variables can be directly included in linear regression models.

A

False

35
Q

cross sectional data

A

observing diff ppl at 1 point in time, then comparing y across ppl w diff levels of x

36
Q

logitudinal data

A

overtime observing ppl and comparing y across ppl at diff levels of x — ex (when ppls income increases (decreases), support for taxes goes up (down)

37
Q

repeated cross section

A

observing ppl at diff time periods comparisons made across groups over time

38
Q

between group experiments are —- comparisons

A

cross-sectional

39
Q

within individuals experiments are — experiments

A

longitudinal

40
Q

linear regression is a

A

summary and a simple approximation

41
Q

formula for linear regression

A

y=a+Bx (y=mx+b)

42
Q

normative

A

what ought to be (should, would)

43
Q

empirical

A

what is (are, is , do)

44
Q

a theory proposes a relationship between —-

A

concepts

45
Q

a hypothesis proposes a relationship between —

A

variables

46
Q

mode

A

most

47
Q

mean

A

add and divide by how many numbers

48
Q

formula for sample variance

A
49
Q

formula for sample standard deviation

A
50
Q

Mode, Median, Mean table

A

nominal-yes no no, ordinal- yes yes no, quantitative-yes yes yes

51
Q

in a left skewed, bell shaped distribution, the mean is —- than the median

A

smaller

52
Q

outlier values tend to affect the position of the median more than the mean (T/F)

A

False

53
Q

the mode of a variable is always located at its 99.9th percentile (T/F)

A

False

54
Q

levels of distribution

A

68, 95, 99.7

55
Q

the empirical rule

A
56
Q

can the empirical rule be applied to normal distributions

A

yes

57
Q

increasing sample size decreases —

A

sampling error

58
Q

Ordinary Least Square is used to

A

find a line of best fit, minimizes the squared distance between line and the data

59
Q

info between line of best fit and data are called

A

residuals

60
Q

confounding variables

A

things that will influence outcome

61
Q

to address confounding variables what should you add?

A

control variables

62
Q

in multiple linear regression we are addressing the — by including z

A

confounds

63
Q

all variables in linear regression must be

A

quantitative

64
Q

we can include categorical variables in the regression by

A

transforming them into dummy variables. 1=in the category, 0=not in the category

65
Q

how to find the baseline category

A

leave 1 dummy variable out

66
Q

randomization works best —

A

in larger samples

67
Q

Selection Effects

A

When people can opt into levels of X and do
so as a function of unobserved potential outcomes

68
Q

Categorical variables must be analyzed with –

A

cross-
tabulations

69
Q

Quantitative variables should be analyzed with —

A

scatter plots
and correlations

70
Q

We use cross tabs to–

A

calculate the percentages

71
Q

When we have two quantitative variables, we can visualize their
relationship using a scatter plot because—

A

Quantitative data usually have too many unique outcomes for a
cross-tabulation to be meaningful

72
Q

When we have two quantitative variables, we can assess their
association (relationship) using –

A

correlation coefficients

73
Q

If the individual item biases are equivalently distributed, the
correlation is —

A

unbiased

74
Q

Confounding

A

When a third variable (Z) causes both X and Y

75
Q

Reverse Causality:

A

Sometimes, things you hypothesize are causes are effects! (y is the cause and x is the effect)

76
Q

Reciprocal Causality

A

A special case of reverse causality
where X and Y are both causes and effects (x and y both causes)

77
Q

Adjustment Error

A

Some surveys
“weight” to better represent the population

78
Q

Two issues in survey taking

A

– Satisficing: Not all respondents will put full
effort into answering your survey questions.
– Dishonesty: Some respondents will not give
their true views, often due to social pressures

. (2 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions