POLITICAL ANALYSIS, 2021, TEST 2

1. (20 points) What is the MODE and the MEDIAN for each of the following distributions (don’t knows are omitted from these polls and distributed across the other categories). You may just write the one word answers, Mode and Median next to the appropriate categories. You should have two answers for each question (A-J), so each answer is worth 1 point.

CBS News Poll. May 29-June 2, 2020. N=1,309 adults nationwide. Margin of error ± 3.1.

"Which of these comes closest to your view? Abortion should be generally available to those who want it. Abortion should be available, but under stricter limits than it is now. OR, Abortion should not be permitted."

Abortions generally available = 44%

Available but stricter limits = 32%

Not permit abortions = 24%

ABC News/Washington Post Poll. Oct. 6-9, 2020. N=879 registered voters nationwide. Margin of error ± 3.5.

"How do you feel about the possibility that you or someone in your immediate family might catch the coronavirus: very worried, somewhat worried, not too worried, or not worried at all?" (Note: already caught it is placed under not worried category.)

Very worried = 29%

Somewhat worried = 36%

Not too worried = 15%

Not worried at all = 20%

Kaiser Family Foundation. July 14-19, 2020. N=1,313 adults nationwide. Margin of error ± 3.

"Overall, how would you rate the federal government's response to the current coronavirus outbreak: excellent, good, fair, or poor?"

Excellent = 7%

Good = 20%

Fair = 24%

Poor = 49%

The Hill-HarrisX poll was conducted online among 945 registered voters. It has a margin of error of 3.19 percentage points. Conducted Jan. 28-29, 2021.

“Do you approve or disapprove of the job Joe Biden is doing as President of the United States?”

Approve = 61%

Disapprove = 39%

New York Times/Siena College Poll. June 17-22, 2020. N=1,337 registered voters nationwide. Margin of error ± 3.

"I'm going to read a few names of people or groups in public life and I'd like you to tell me whether you have a very favorable, somewhat favorable, somewhat unfavorable, or very unfavorable impression of each. ... The police."

Very favorable = 45%

Somewhat favorable = 30%

Somewhat unfavorable = 15%

Very unfavorable = 10%

Washington Post-Schar School Poll. June 2-7, 2020. N=1,006 adults nationwide. Margin of error ± 3.5.

"As you may have heard, a black man named George Floyd was killed in police custody in Minnesota after three white police officers kneeled on his neck and body. Four officers were fired and have been charged in Floyd's death, including one who was charged with murder. ..." [Interviews conducted June 2 ended with: "Four officers were fired and one was charged with third-degree murder and manslaughter."]

"Do you think police are using too much force when dealing with protesters in cities across the country, are not using enough force, or are they handling this about right?"

Too much force used = 45%

Handling this about right = 40%

Not using enough force = 15%

NBC News/Wall Street Journal Poll conducted by Hart Research Associates (D) and Public Opinion Strategies (R). Aug. 10-14, 2019. N=1,000 adults nationwide. Margin of error ± 3.1.

"I'm going to read you some actions that Congress might take related to guns. Would you strongly support, somewhat support, somewhat oppose, or strongly oppose this action? ..."

"Banning the sale of selected semi-automatic firearms referred to as assault weapons"

Strongly support = 49%

Somewhat support = 14%

Somewhat oppose = 9%

Strongly oppose = 28%

Same poll and introductory statement as in G above.

"Banning the sale of handguns."

Strongly support = 13%

Somewhat support = 12%

Somewhat oppose = 17%

Strongly oppose = 58%

NBC News/Wall Street Journal Poll conducted by Hart Research Associates (D) and Public Opinion Strategies (R). March 11-13, 2020. N=900 registered voters nationwide. Margin of error ± 3.3.

"Now as you may know, Barack Obama's health care plan was signed into law in 2010. From what you have heard about the health care law, do you think it is a good idea or a bad idea? If you do not have an opinion either way, please just say so." (No opinions are rated as mixed views to produce an ordinal level variable.)

Good idea = 42%

Mixed views = 23%

Bad idea = 35%

CBS News Poll. May 29-June 2, 2020. N=1,309 adults nationwide. Margin of error ± 3.1.

"How much discrimination do you think there is against each of these groups in society today: a lot, some, not much, or none at all? ...What about African Americans?"

A lot = 50%

Some = 31%

Not much or none = 17%

No Opinion = 2%

2. (8 points) This question tests your knowledge of the statistic MEAN. It uses data from the July 7-9, 2020 Morning Consult asked 1,993 voters to place Biden and President Donald Trump on an ideological scale of one to seven, with one meaning “very liberal,” four meaning “moderate” and seven meaning “very conservative.” (Note: I gave plausible names to the other four categories so that you can answer the question correctly. The poll asks about the voter’s ideal candidate, which I interpret as their own ideology.)

1- Very Liberal

2- Liberal

3- Slightly Liberal

4- Moderate

5- Slightly Conservative

6- Conservative

7- Very Conservative

The self-placement of the average American voter gave a mean value of 4.2

The average American’s perception of Joe Biden gave a mean of 2.8

The average American’s perception of Donald Trump gave a mean of 5.7

Now answer the following 5 questions:

A) What is the mean in words of the average person’s perception of Joe Biden’s ideology?

B) What is the mean in words of the average person’s perception of Donald Trump’s ideology?

C) What is the mean in words of the average person’s own ideology?

D) Generally, was the average American ideologically closer to Joe Biden or to Donald Trump (in terms of their perceptions)?

3. (12 points) What is the RANGE for each of the following scales:

A) Adjusted net national income per capita ranged from a high of $66,019 in Norway to a low of $206 in Burundi. (2011 data, https://data.worldbank.org/indicator/NY.ADJ.NNTY.PC.CD)

B) Incarceration rate per 100,000 population ranges from a low of 6 in San Marino to a high of 655 in the United States. (https://en.wikipedia.org/wiki/List_of_countries_by_incarceration_rate)

C) Life expectancy ranges from a high of 85 in Japan to a low of 54 in the Central African Republic. (https://www.worldometers.info/demographics/life-expectancy/#countries-ranked-by-life-expectancy)

D) Average years of schooling of adults ranged from a high of 12 for the United States to a low of 1 for the countries of Niger, Mali, and Guinea-Bissau. (https://www.nationmaster.com/country-info/stats/Education/Average-years-of-schooling-of-adults)

E) Quality of life index ranges from a high of 191 for Switzerland to a low of 52 for Nigeria. (https://www.numbeo.com/quality-of-life/rankings_by_country.jsp)

F) Happiness ranged from a high of 7.8 for Finland to a low of 2.6 for Afghanistan. (https://worldhappiness.report/ed/2020/social-environments-for-world-happiness/)

4. (10 points) Using the statistic of VARIANCE, are whites and blacks most united on each of the following factors. These are data from the General Social Survey, cumulative file from 1972 thru 2018, so they are recent historical in nature. (https://sda.berkeley.edu/)

A- Years of education completed.

White variance = 9.22

Black variance = 9.98

B- Abortion attitudes.

White variance = .27

Black variance = .38

C- Afraid to walk in neighborhood at night.

White variance = .23

Black variance = .25

D- Confidence in Congress.

White variance = .40

Black variance = .41

E- Pride in America’s history.

White variance = .44

Black variance = .95

5. (15 points) A researcher is studying how gender differences may affect political attitudes or one’s behavior or experiences. Using the national 2018 General Social Survey poll, she generates the following five tables based on testing the relationship between sex and an issue attitude or behavior. Columns total 100% down each column. For each of these tables, write the hypothesis showing the direction of the relationship between the two variables.

A. Sex and Attitude toward the Death Penalty for Murder

Death Penalty Attitude	Male	Female
For	67%	60%
Against	33%	40%

B. Sex and Permit Free Speech for Racists

Permit Free Speech for Racists	Male	Female
Yes	65%	50%
No	35%	50%

C. Sex and Spending on National Health Care Is:

Current health care spending is:	Male	Female
Too Little	67%	77%
About Right or Too Much	33%	23%

D. Sex and Frequency of Reading Newspaper

Frequency of Newspaper Reading	Male	Female
Every Day	21%	16%
Middle response	38%	35%
Never	41%	49%

E. Sex and Ever Threatened or Harassed?

	Male	Female
Yes	40%	36%
No	60%	64%

6. (10 points) A researcher did a study of the correlates of support for nation government spending on health care, using the 2018 General Social Survey. She derived the following GAMMA values depicting the strength of the relationships between each demographic variable and health care spending support.

Those gamma values are:

Income and health care spending support = .07

Ideology and health care spending support = .24

Race and health care spending support = -.21

Age and health care spending support = -.01

Party Identification and health care spending support = .32

A. Which of these five factors is MOST important in shaping health care spending support?

B. Which of these five factors is SECOND MOST important in shaping health care spending support?

C. Which of these five factors is THIRD MOST important in shaping health care spending support?

D. Which of these five factors is FOURTH MOST important in shaping health care spending support?

E. Which of these five factors is LEAST important in shaping health care spending support?

7. (25 points) Please study the following three bivariate tables and four multivariate tables, and answer each of the lettered questions. This info is drawn from the 2018 General Social Survey.

A 5% difference or higher constitutes statistical significance. They bear on the model (Age is the earliest variable, Church Attendance is the Dependent Variable, and Marital Status is the intervening variable):

MARITAL STATUS

(married)

Age Church Attendance

(old) (high)

BIVARIATE RELATION BETWEEN AGE AND CHURCH ATTENDANCE:

Church Attendance	18-35	36-65	66 and older
Yearly or less	58%	45%	40%
Sometimes	23%	27%	21%
Weekly or more	19%	28%	39%
N Size	(672)	(1159)	(495)

A. What is the direction of the relationship between age and church attendance?

BIVARIATE RELATION BETWEEN AGE AND MARITAL STATUS:

Marital Status	18-35	36-65	66 and older
Married	31%	60%	51%
Not Married	69%	40%	49%
N Size	(672)	(1167)	(500)

B. What is the direction of the relationship between age and marital status?

BIVARIATE RELATION BETWEEN MARITAL STATUS AND CHURCH ATTENDANCE:

Church Attendance	Married	Not Married
Yearly or less	43%	53%
Sometimes	25%	25%
Weekly or more	32%	22%
N Size	(989)	(1342)

C. What is the direction of the relationship between marital status and church attendance?

RELATIONSHIP BETWEEN AGE AND CHURCH ATTENDANCE, AMONG THE MARRIED

Church Attendance	18-35	36-65	66 and older
Yearly or less	52%	41%	42%
Sometimes	21%	27%	19%
Weekly or more	27%	32%	39%
N Size	(206)	(585)	(194)

D. Among those who are married, is there any relationship between age and church attendance? Yes, or no? If so, what is the direction of the relationship?

RELATIONSHIP BETWEEN AGE AND CHURCH ATTENDANCE, AMONG THOSE NOT MARRIED

Church Attendance	18-35	36-65	66 and older
Yearly or less	60%	52%	39%
Sometimes	25%	26%	23%
Weekly or more	15%	22%	38%
N Size	(466)	(573)	(301)

E. Among those who are unmarried, is there any relationship between age and church attendance? Yes, or no? If so, what is the direction of the relationship?

RELATIONSHIP BETWEEN MARITAL STATUS AND CHURCH ATTENDANCE, AMONG ADULTS 18-35

Church Attendance	Married	Not Married
Yearly or less	52%	60%
Sometimes	21%	25%
Weekly or more	27%	15%
N Size	(206)	(466)

F. Among adults 18-35 years old, is there any relationship between marital status and church attendance? Yes, or no? If so, what is the direction of the relationship?

RELATIONSHIP BETWEEN MARITAL STATUS AND CHURCH ATTENDANCE, AMONG ADULTS 66 AND OLDER

Church Attendance	Married	Not Married
Yearly or less	42%	39%
Sometimes	19%	23%
Weekly or more	39%	38%
N Size	(194)	(301)

G. Among adults 66 and older, is there any relationship between marital status and church attendance? Yes, or no? If so, what is the direction of the relationship?

H. So what factor consistently has a statistically significant effect on church attendance? Is it age, marital status, or both? Just circle the correct response.