Math 124: Paper 2Due: Friday, November 20th on iLearn

Data analysis can be roughly divided into two large categories: confirmatory data analysis (CDA) and exploratorydata analysis (EDA). CDA is focused on testing theories. Does a set of data conform to the outcomes specified bysome theory? For example, the general theory of relativity predicts that time on a satellite orbiting the earth will notbe synchronous with time passing on earth because the time bending influence of gravity on the satellite is a quarterof what it is on earth. If we compare time on 40 atomic clocks aboard satellites to 40 atomic clocks on earth, weexpect the mean time of the first 40 to be different that the latter 40. We could use a t test to critically examine theseresults. That makes a t test a CDA statistics.

In contrast, EDA does not start with a theory. Instead, it explores data to find patterns and then suggests theoreticalnotions that might be tested with CDA (on different data). EDA now goes by the term ‘data mining’ and its utilityfor marketing, consumer profiling, understanding linguistic shifts, and so on has been amply demonstrated over thepast couple decades. This paper requires you to engage in a simple version of EDA.

Unlike paper one, there is no central question for you to answer in this paper. Rather, I want you to explorethe connection between political ideology and each of five demographic variables often thought to be importantdeterminants of social attitudes – gender, income, race, and education. For this paper, you must produce:

• good univariate description of the measure of political ideology (IDEOCONSIST),
• a table summarizing central tendency and dispersion statistics for political ideology disaggregated by values ofthe other five variables (AGE, SEX, INC, RACIAL, EDUC),
• t tests or Wilcoxon tests of differences between the two values of dichotomous variables,
• t tests or Wilcoxon tests of differences between selected values of categorical variables with more than two values,
• appropriate correlation and regression statistics for differences over values of measurement data,
• appropriate tables or graphics presenting these bivariate statistics, and
• accompanying text turning all of these elements into a coherent description of variation in political ideology bydemographic category.I’ve pushed back the due date a bit to give us more time to cover things. I’m also going to try to touch on thesedata as we learn some of the techniques I ask you to produce in the paper. I hope this gives you more opportunity toconceptualize the paper, something that I think was a problem with paper one.The following pages contain descriptions of the variables I’ve included in our dataset, but the this is just a small subsetfrom a full social survey. This larger dataset was produced by the Pew Research Center. Pew publications relatedto this dataset and a link to the full data can be found at: https://www.people-press.org/dataset/2014-political-polarization-survey/

1

Codebook

This codebook describes the five variables I’d like you to use for this paper. You need to understand the data wellin order to work with them easily. This is especially true for the variable measuring political ideology. That one issomewhat complicated in its composition but is key to this paper. All this is to say that you should ask plenty ofquestions about these data before tackling the analysis and writing of the paper.

AGE — age of respondent in years

SEX — gender identity of respondent

1. Man
2. Woman

INC — income of respondent’s household

1. Less than $10000
2. $10,000 to under $20,0003. $20,000 to under $30,0004. $30,000 to under $40,0005. $40,000 to under $50,0006. $50,000 to under $75,0007. $75,000 to under $100,0008. $100,000 to under $150,0009. $150,000 to under $250,000

10. $250,000 or more
RACIAL — ethnoracial identity of respondent

1. African American2. Asian
3. Latinx
4. Mixed

5. Some other race6. White

EDUC — formal educational attainment of respondent

1. Less than high school (Grades 1-8 or no formal schooling)
2. High school incomplete (Grades 9-11 or Grade 12 with NO diploma)
3. High school graduate (Grade 12 with diploma or GED certificate)
4. Some college, no degree (includes community college)
5. Two year associate degree from a college or university
6. Four year college or university degree/Bachelor’s degree (e.g., BS, BA, AB)
7. Some postgraduate or professional schooling, no postgraduate degree (e.g. some graduate school)
8. Postgraduate or professional degree, including master’s, doctorate, medical or law degree (e.g., MA, MS, PhD,MD, JD)

IDEOCONSIST This is a ratio-level measure of liberal to conservative ideology. It ranges from -10 (most liberal inattitude) to 10 (most conservative in attitude). Because this is the primary variable I want you to explore, it’s usefulto know a bit about how this index was created. Essentially, ten questions asked of all respondents and they were giventwo choices for response – a liberal answer and a conservative answer. Liberal answers were each given the value -1.Conservative answers were each given the value 1. If I respondent answered all 10 questions with the liberal response,

2

they would score a -10 on the index. A completely consistent conservative would score a 10. Respondent who gota 0 had half liberal and half conservative answers. Respondents also sometimes did not answer a question or circledboth responses. In such cases, the respondent was given a 0 on this item. The questions used for this index are givenverbatim below.

I’m going to read you some pairs of statements that will help us understand how you feel about a number of things.As I read each pair, tell me whether the FIRST statement or the SECOND statement comes closer to your own views— even if neither is exactly right. The first pair is [READ AND RANDOMIZE PAIRS BUT NOT STATEMENTSWITHIN EACH PAIR]. Next, [NEXT PAIR] [IF NECESSARY: “Which statement comes closer to your views, evenif neither is exactly right?”]

1. Government is almost always wasteful and inefficient [OR] Government often does a better job than people giveit credit for
2. Government regulation of business is necessary to protect the public interest [OR] Government regulation ofbusiness usually does more harm than good
3. Poor people today have it easy because they can get government benefits without doing anything in return [OR]Poor people have hard lives because government benefits don’t go far enough to help them live decently
4. The government should do more to help needy Americans, even if it means going deeper into debt [OR] Thegovernment today can’t afford to do much more to help the needy
5. Racial discrimination is the main reason why many black people can’t get ahead these days [OR] Blacks whocan’t get ahead in this country are mostly responsible for their own condition
6. Immigrants today strengthen our country because of their hard work and talents [OR] Immigrants today are aburden on our country because they take our jobs, housing and health care
7. The best way to ensure peace is through military strength [OR] Good diplomacy is the best way to ensure peace
8. Business corporations make too much profit [OR] Most corporations make a fair and reasonable amount of profit

Now I’m going to read a few more pairs of statements. Again, just tell me whether the FIRST statement or theSECOND statement comes closer to your own views — even if neither is exactly right. Next, [NEXT PAIR] [IFNECESSARY: “Which statement comes closer to your views, even if neither is exactly right?”]

1. Stricter environmental laws and regulations cost too many jobs and hurt the economy [OR] Stricter environmentallaws and regulations are worth the cost
2. Homosexuality should be accepted by society [OR] Homosexuality should be discouraged by society