Question 1

Which of the following is not the same as the others?
A) "controlling for…"
B) "holding constant…"
C) "keeping constant…"
D) "regressing onto…"

Accepted Answer

The first three terms are used throughout chapter, last term has nothing to do with nested models.

Question 2

In Model 1, Independent Variable A has a statistically significant effect. In Model 2, we add Independent Variable B, which has a statistically significant effect, and the effect of Independent Variable A moves closer to zero and loses its statistical significance. What might we have here?
A) an intervening relationship
B) a dependent relationship
C) an independent relationship
D) a controlling relationship

Accepted Answer

This scenario suggests an intervening relationship, where the introduction of Variable B influences the effect of Variable A on the dependent variable, indicating that Variable B may mediate the relationship between Variable A and the outcome.

Question 3

What symbol is used to signify "this independent variable is not in this model, but it will be introduced in a subsequent model?
A) ---
B) ***
C) >>>
D) :::

Accepted Answer

The symbol "---" is often used in statistical modeling to indicate that an independent variable is not included in the current model but will be introduced in a subsequent model.

Question 4

When you have a first regression model, then a second regression model with an additional independent variable, then a third regression model with yet another additional independent variable, we call these a set of:
A) concurrent models
B) nested models
C) incremental models
D) subsequent models

Accepted Answer

Nested models are a series of models where each model includes all of the variables of the previous model plus one or more additional variables. This allows for the assessment of the incremental value of the new variables added at each step.

Question 5

The smallest number of regression models you need to have nested modeling is:
A) 1
B) 2
C) 3
D) 4

Accepted Answer

Nested modeling requires at least two regression models, where one model has a subset of variables included in the other model. This allows for the comparison of the two models in terms of added explanatory power or significance of the additional variables.

Question 6

We run a regression model using GSS2006 data and find out that the older one is, the higher he/she scores on an index of religiosity (where 0=not religious up to 9=very religious). Then, we hypothesize that, because women outlive men, and because women are typically more religious than men are, part of this age effect is actually due to sex. We run a second model in which we add a variable for sex:
$\begin{array}{lcc}
\text { Independent Variable } & \text { Model 1 } & \text { Model 2 } \
\text { Age (in years) } & .03 * * * & .03 * * * \
\text { Sex }(\mathrm{F}=0, \mathrm{M}=1) & --- & -.93 * * * \
\text { Constant } & 4.23 & 4.65 \
\text { R-Squared } & .04 & .07 \
\mathrm{n} & 2912 & 2912
\end{array}$ Which of the following is the most appropriate interpretation of what is going on here?
A) Sex clearly has a larger effect than age, so our hypothesis is supported.
B) The value of R-Squared rises, so our hypothesis is supported.
C) The effect of age does not change, so our hypothesis is not supported.
D) The constant increases, so our hypothesis is not supported.

Accepted Answer

The effect of age remains the same in both models, indicating that the age effect is not explained by sex, thus not supporting the hypothesis.

Question 7

Here are two models made using GSS2006 data. The dependent variable is hours of television watched per day:
$\begin{array}{lcc}
\text { Independent Variable } & \text { Model 1 } & \text { Model } 2 \
\text { Age (in years) } & .02^{* * *} & .00 \
\text { Working or Retired } & -- & 1.32^{* * *} \
\text { Constant } & 1.70 & 2.51 \
\text { R-Squared } & .03 & .06 \
\mathrm{n} & 1523 & 1523
\end{array}$ Which of the following pairs of people watches the same hours of television?
A) a retired 50-year-old and a retired 80-year-old
B) two 80 year-olds: one who is working, one who is retired
C) two 50-year-olds: one who is working, one who is retired
D) a 20-year-old working person and an 86-year-old retired person

Accepted Answer

The answer of Here are two models made using GSS2006...

Question 8

We hypothetically observe that the higher one's education, the happier one is. We hypothesize that this is actually because of income: people with higher education tend to make higher incomes, and it is these higher incomes, not education, that causes the higher happiness. Here are hypothetical models (using a dependent variable where 0=not at all happy, up to 10=very happy):
$\begin{array}{lcc}
\text { Independent Variable } & \text { Model 1 } & \text { Model } 2 \
\text { Education (in years) } & .35^{* * *} & ? ? ? \
\text { Income (in thousands of dollars) } & --- & .03^{* * *} \
\text { Constant } & .50 & -2.50 \
\text { R-Squared } & .10 & .15 \
\mathrm{n} & 1000 & 1000
\end{array}$ To support the hypothesis, what is the most likely number that would go in the place of the "???" in Model 2?
A) .03
B) .20**
C) .35***
D) .50***

Accepted Answer

The answer of We hypothetically observe that the higher one's...

Question 9

In the "Attitudes toward Inequality" example in the textbook, which group's difference from whites is most robust?
A) blacks
B) others
C) both blacks and others have similarly robust effects
D) neither blacks nor others differ from whites

Accepted Answer

The answer of In the "Attitudes toward Inequality" example in...

Question 10

In the "Attitudes toward Inequality" example in the textbook, what happens to the effect of political party once income is controlled for?
A) it loses its statistical significance
B) it goes down by a lot, but maintains its statistical significance
C) it goes down by a little
D) it goes up

Accepted Answer

The answer of In the "Attitudes toward Inequality" example in...

Question 11

In Model 1, Independent Variable A does not have a statistically significant effect. In Model 2, we add Independent Variable B, which has a statistically significant effect, and the effect of Independent Variable A gets larger and is now statistically significant. Independent Variable B is called:
A) an intervening variable
B) a controlling variable
C) a suppressor variable
D) a detracting variable

Accepted Answer

Based on the given information, the addition of Independent Variable B has led to the effect of Independent Variable A becoming larger and statistically significant. This suggests that Independent Variable B is suppressing the effect of Independent Variable A in Model 1, hence it is called a suppressor variable.

Question 12

The "BMI, Internet Use, and Age" example in the textbook offered an example of a(n) _______________ relationship.
A) intervening
B) suppressor
C) robust
D) dependent

Accepted Answer

The "BMI, Internet Use, and Age" example illustrates a suppressor relationship, where the inclusion of a third variable (age) clarifies or changes the direction or strength of the relationship between the primary variables (BMI and Internet Use).

Question 13

What statistical test do we use to see if a second regression model is better than the first regression model?
A) the chi-square test
B) the t-test
C) the improvement test
D) the F-test

Accepted Answer

The F-test is used to compare the overall fit of two regression models. It tests the null hypothesis that the two models have equal fit against the alternative hypothesis that the second model has better fit than the first model.

Question 14

Someone hypothetically comes to you with some regression results that are confusing him:
$\begin{array}{lcc}
\text { Independent Variable: } & \text { Model 1 } & \text { Model } 2 \
\text { Education (in years) } & 3.47^{* * *} & 3.47 \
\text { Uses a computer at work }(Y=1, N=0) & -- & 7.80 \
\text { Constant } & -20.45 & -24.20 \
\text { R-Squared } & .14 & .14 \
n & 1000 & 100
\end{array}$ He is confused by what happens to the effect of education between Model 1 and Model 2: the value of the slope is the same, but it is no longer statistically significant.
What is the most likely explanation for this?
A) It is not that education has an effect, but that people with more education are more likely to use a computer at work.
B) The value for R-squared stayed the same, so having two independent variables is no better than having one independent variable.
C) The constant in Model 2 decreased by a value greater than the slope for education.
D) The sample size has decreased.

Accepted Answer

The most likely explanation for the change in statistical significance of the education variable between Model 1 and Model 2, while the slope remains the same, is the decrease in sample size from 1000 to 100. A smaller sample size reduces the statistical power of the test, making it harder to detect a statistically significant effect.

Question 15

Ainsworth-Darnell and Downey use which dataset for their study of student grades?
A) The National Longitudinal Study of Adolescent to Adult Health
B) The General Social Survey
C) The National Educational Longitudinal Study
D) The International Study of Student Performance

Accepted Answer

Ainsworth-Darnell and Downey specifically state in their article that they used data from the National Educational Longitudinal Study, which is a large-scale survey that follows students from middle school through early adulthood. The other options listed are other datasets that may be used in sociological research, but they are not the dataset used in this particular study.

Question 16

The explanation that Ainsworth-Darnell and Downey refute with their statistical analysis of racial differences in grades is called:
A) the culture of poverty
B) oppositional culture
C) racial differential association
D) culturally biased testing

Accepted Answer

Ainsworth-Darnell and Downey refute the "oppositional culture" explanation, which suggests that some minority students underperform academically because they perceive academic achievement as being aligned with the culture of the dominant group, and thus reject it to maintain their own cultural identity.

Question 17

In the Ainsworth-Darnell and Downey research on racial differences in grades, which of the following nested stories is not present?
A) where the original independent variable's effect stays the same
B) where the original independent variable's effect decreases, but remains statistically significant
C) where the original independent variable's effect goes away completely
D) where the original independent variable's effect comes back

Accepted Answer

The answer of In the Ainsworth-Darnell and Downey research on...

Question 18

With regard to the expectation that black students have more negative attitudes toward school than white students, Ainsworth-Darnell and Downey provide findings that _________ this expectation.
A) fully support
B) provide some support for
C) refute
D) don't address

Accepted Answer

The question is asking which choice best describes the findings of Ainsworth-Darnell and Downey regarding the expectation that black students have more negative attitudes toward school than white students. Choice C, "refute," is the best choice because the researchers found that black students do not necessarily have more negative attitudes toward school than white students, which goes against the expectation. Choices A and B are incorrect because the findings do not fully or partially support the expectation. Choice D is also incorrect because the question explicitly asks about the researchers' findings, which are addressed in the study.

Question 19

Regarding the interconnections between race and social class in the Ainsworth-Darnell and Downey article about grades, which of the following statements best captures their overall findings:
A) Race and social class are not at all interconnected.
B) Social class fully accounts for racial differences.
C) Race fully accounts for social class differences.
D) Social class partially accounts for racial differences.

Accepted Answer

The Ainsworth-Darnell and Downey article finds that social class partially accounts for racial differences in grades. While race remains a significant factor in academic outcomes, the authors argue that social class plays a role in shaping students' experiences and opportunities within schools. This suggests that both race and social class are interconnected and cannot be fully understood separately.

Question 20

What was Callanan's dependent variable?
A) exposure to television news
B) exposure to crime dramas on television
C) risk of crime
D) fear of crime

Accepted Answer

Callanan's study measured fear of crime, which is the dependent variable. The independent variables were exposure to television news and exposure to crime dramas on television, but the dependent variable is what is being measured and affected by the independent variables.

Quiz 10: Above and Beyond: The Logic of Controlling and The Power of Nested Regression Models

Which of the following is not the same as the others?

In Model 1, Independent Variable A has a statistically significant effect. In Model 2, we add Independent Variable B, which has a statistically significant effect, and the effect of Independent Variable A moves closer to zero and loses its statistical significance. What might we have here?

What symbol is used to signify "this independent variable is not in this model, but it will be introduced in a subsequent model?

When you have a first regression model, then a second regression model with an additional independent variable, then a third regression model with yet another additional independent variable, we call these a set of:

The smallest number of regression models you need to have nested modeling is:

In the "Attitudes toward Inequality" example in the textbook, which group's difference from whites is most robust?

In the "Attitudes toward Inequality" example in the textbook, what happens to the effect of political party once income is controlled for?

In Model 1, Independent Variable A does not have a statistically significant effect. In Model 2, we add Independent Variable B, which has a statistically significant effect, and the effect of Independent Variable A gets larger and is now statistically significant. Independent Variable B is called:

The "BMI, Internet Use, and Age" example in the textbook offered an example of a(n) _______________ relationship.

What statistical test do we use to see if a second regression model is better than the first regression model?

Ainsworth-Darnell and Downey use which dataset for their study of student grades?

The explanation that Ainsworth-Darnell and Downey refute with their statistical analysis of racial differences in grades is called:

In the Ainsworth-Darnell and Downey research on racial differences in grades, which of the following nested stories is not present?

With regard to the expectation that black students have more negative attitudes toward school than white students, Ainsworth-Darnell and Downey provide findings that _________ this expectation.

Regarding the interconnections between race and social class in the Ainsworth-Darnell and Downey article about grades, which of the following statements best captures their overall findings:

What was Callanan's dependent variable?