Second section: post paper proposals here
Your paper proposal should follow this format:
1) Your name
2) Your main thesis or what you want to write about
3) The main relationship that you want to examine
Here is an example:
I am interested in whether poor people turned out during the Depression.
My main relationship I want to examine is whether poverty is positively related to turnout in a Depression year. I will compare this to poverty and turnout in a non-Depression year. I will also control for how much a respondent liked FDR.
1) Your name
2) Your main thesis or what you want to write about
3) The main relationship that you want to examine
Here is an example:
I am interested in whether poor people turned out during the Depression.
My main relationship I want to examine is whether poverty is positively related to turnout in a Depression year. I will compare this to poverty and turnout in a non-Depression year. I will also control for how much a respondent liked FDR.
posted by Paul Gronke at 11/06/2006 10:53:00 PM
45 Comments:
my name is sabok
I would like to write about women, education, employment and voting. Ideally, i would like to see whether whether the voting turnout of a male spouse affects whether a female spouse votes. then i would like to control for education and employment (using the variable v3260a for hours worked) to see if these affect political participation among married women. the relationship i want to examine is whether women make choices to participate politically independent of thier husbands. predictably, i will argue that women who have a higher degree of independence (indicated by level of education and hours worked) will turnout more.
the relationship i would like to examine is whether marital status affects women's turnout.
my name is mcgarry
I would like to see whether care, strength of preference, or political activity has a greater effect on turnout, using the variables pres approval, care who, wins, will race be close, interest in campaigns, strength of preference for candidate, and the pol210 index on political activity.
I would like to write about predictions about the election and voting turnout. I want to study whether it matters who the respondent thinks is going to win the election in terms of determining their likelihood to vote. I also want to confirm that the respondents who thought the election would be close were more likely to vote than those who thought the election would not be close. I can also separate this by male and female and see if there is any correllation between gender, predictions, and voting turnout.
-Nick
I would like to see if there is a correlation between authoritarian values (p2102-child) and patriotism (p210-patriotism) by gender and how this affects voting turnout (p210-didvote). I plan to compare the results of overall turnout by gender, by authoritarianism, and by patriotism (if no major correlation exists between the latter and the former) to see if an overall trend exists between the three.
-Lisa
from Carmina
My proposed study is on the relationship between voter perception on economic conditions and voter turnout. Moreover, I am interested in the discrepency of this relationship between party lines. (i.e. do republicans/democrats turn out more/less during perceived economic highs/lows?). I would also like to take into account the economic status of the respondant observing: 1) whether R invested in the stock market and 2) whether R can afford health care. I will measure economic preception with variables such as: "national economy better or worse in the last year", "will national economy be better or worse in the next year", or "unemployment better or worse in the last year".
Patricia Snarski
I want to write about American citizens who identify with another ethnicity or whose parents are not from the United States, especially those who identify themselves as members from the coalition of the willing. Then I want to look at how they feel about the Iraq war in 2004 and for who they voted for.
Essentially I want to look at people who don't consider themselves purely "American" and understand their voting habits compared to their other ethnicity's nation policy on the same subject.
Mark M. Jaycox
I want to write about the relationship between unemployment and voter turnout-if turnout is in favor of the democrats or in favor of the party that is not in power of the executive.
The main relationship I want to examine is if the unemployment rate increase the rate of turnout for the party that is not in power of the Executive (or should I see if unemployment increases voter turnout for the democrat party in mid-term and presidential elections)
Do you have any suggestions as to whether I should do democrats or the party that is not in power of the Executive?
*increases
I would to study the voting trends of religiously active individuals in the United States.
I would consider whether active Christians in America participate more or less than people who never go to church. Furthermore, I would like to see whether or not socioeconomic class has identification as religiously active, and what this explaines about turnout.
I’m interested in whether an individual's concern regarding government medical insurance policy influences voter turnout.
I want to examine how interest in govt medical insurance varies across different age groups and establish whether this is a primary factor explaining why the most vulnerable age group in this respect (senior citizens) usually has the highest voter turnout during elections. I’ll also find out whether citizens who cannot afford health insurance and are directly affected by this policy turn up to vote.
my name is Jordan Kinley
I would like to write about whether absolute satisfaction or complete disastisfaction affects a person's turnout. I would ideally like to determine whether people who agree with the state of the country are more likely to vote than those who do not agree. The two variables I will use to determine the person's satisfaction deal with their perception of the state of the economy(v3097) and the state of the moral climate(v3291). Ideally I would like to see if on the whole, satisfaction with the overall state of the country in these two variables makes a person more likely to vote or the contrary. Secondly I would like to judge the difference in likelihood between the variables and venture reasonable explanations for the differences (i.e. does satisfaction with the economy make it more likely for some one to turnout than the moral climate).
I want to compare political activism and turnout.
Which candidate are activists most likely to favor? How patriotic are activists? Are they more likely to be pessimistic about how women/blacks are treated in society. Essentially, I will try to find if political activists are more or less likely to have enough faith in our government to vote. In other words, does voiceing your opinion through protest translate into valueing the power of your vote.
my name is walsh
I would like to see if there is a positive correlation between low-income, white males and voter turnout for Bush. As well, I would like to compare this to their feelings about Bush and their perceptions on the economy and job lost in the US. Are they voting in their economic self-interest? Or is their vote for Bush the result of an emotional attachment to what Bush represents, the "every man"? I would also compare these findings to the 2000 election votes for Bush.
I'd like to study opinions about foreign policy goals compared to opinions on the efficacy of the Bush administration's war on terror.
Specifically, v3215 questions whether R feels that the U.S. has become more or less secure since 2000, v3112 asks if the U.S. position in the world has grown stronger or weakerand v3135 asks if the war in Iraq has decreased or increased the threat of terrorism.
I'd like to compare those to the importance of being American (v3209) to determine R's fundamental opinions of the importance of nationality and American foreign power. What should be the primary goal of a U.S. international presence, and how do those opinions change voting habits/preferences?
My proposed topic would investigate the voting characteristics of 18-24 year old voters in the 2004 election.
Much was made of non-partisan efforts to mobilize this demographic of younger voters, though predictions of increased turnout among 18-24 year olds turned out to be gross overestimates. My paper will examine the characteristics of members of this demographic who DID vote. Specifically, I will investigate turnout--to an extent--in terms of education and income levels, though the thrust of my analysis will come in the percieved political efficacy and previous non-voting participation of those 18-24 year olds who did (and did not) vote.
Leo here. I would like to compare the impotance of party preferences and information (knowledge, attention to campaigns) in congressional elections for working class/lower income voters. Which one has the strongest effect? Do low income voters prefer vote a certain way because they feel themselves well informed about their choices or do they default to other criteria, such as "uninformed" party preference or to the same party as their choice for presidential candidate?
I am interested in writing about how an individual’s views on abortion might affect voter turnout. I would also like to examine any possible correlation between Christian voter turnout and pro-life opinions. I will then compare any difference between Christian turn out related to views on abortion between the 2000 and 2004 elections.
-Katrina Brinkerhoff
Sabok:
I don't think you can examine voting between spouses. They dont ask about a spouse's voting behavior, only your own.
But don't give up! It may be possible to examine married women vs. non-married women, or married women who work outside of the home vs. married women who are homemakers.
I think you have a good idea there.
McGarry,
We talked about this. This will work just fine. THe trick here is going to be combining some of these "care who wins" and "ideological extremity" or feeling thermometer extremity measures into something that you can use.
Post up requests on the variable request thread.
Nick
Good and straightforward. I like how you frame the question--you can link this to external media coverage about the election outcome, and use that to motivate the paper.
I think you have a good sense of what items you want to use.
Lisa (Manson?):
Sounds good. Be careful with your language--you may want to correlate authoritarianism, patriotism, and gender, but ultimately you want to see how they all affect turnout.
So you may want to start by compring turnout levels among authoritarians and non-authoritarians, and men / women, then adding all into the regression.
Carmina (aka sensation):
Good theory but one problem: you can't really say anything about highs and lows since we have one fixed natioanl economy in 2004.
However, you can examine differences in the *perceptions* of the economy and turnout.
Note that at least the way you present it, you are theorizing an interaction between party and economic perceptions (that is, being a Republican *and* thinking the economy is doing well = high turnout).
This implies:
turnout=economy + party + economy*party
Patricia
Ouch! That is a great idea which I am almost certain is impossible. You'd need to examine ethnic identity and relate this to the members of the "coalition of the willing."
I don't think the NES has a measure of ethnic background (I could be wrong about this).
AHA! V3303. Ok this is feasible but it could be hard to code. We can help you only if you get us a list of what codes on V3303 you want to be 1-member of coalition, 0-not.
Mark J,
Classic economics and politics argument. Do you penalize the party that is seen as hurting unemployed people or do you penalize the party in power?
Tough choice but since you are doing turnout only, it is easy: you just examine unemp status -> turnout. You might also consider measures of personal and national economic well-being.
Andy House,
Nice idea and should be not too difficult to carry off. We need some guidance as to what you mean by religiously active. Belief in biblical inerrancy? Attendace at chuch? Member of certain denominations?
Controlling by SES is a great idea. Income + education?
Tinashe
Good and straightforward. You need some other controls in there--how about income? You can use that as a surrogate for affordability of health care. Maybe employment status also?
Jordan,
nice idea. How do you want to code these? Do you want us to add them up in such as way as to make a "very unhappy and depressed" scale? Or just a 0=super unhappy, 1=not super unhappy?
And what theory of turnout does this implicate? Not very rat choice, is it? Sort of a moral/state of times turnout model.
Emily,
Keep in mind that you are studying turnout at some point in the paper. But I think a paper that profiles activists, seeing what attitudes and behaviors they have, and ultimately leading this to a turnout examination, is a nice idea.
Relate this to the literature we have read on activism, obviously Rosenstone and Hansen mainly, but also stuff in SB.
Kitty,
Ok, this is fine. I think you are going to want a series of dummy variables (low income / not low income and white/not white).
I think the first thing you want to do is examine any voting differences in this group. Then why? Do they differ on economy? Something else?
Angelique,
Nicely put. Keep in mind that you need to examine turnout, not voting preferences. So essentially does "safety" or perceptions of fear impel people to turn out to vote or not. Presumably this translates in some way to a sense of costs/benefits?
Matt,
Nicely done. So you look at this group first, looking at vote/not vote. Then you will be looking for statistically significant differences (either cross tabs or diff means) on a variety of demog and attitudinal measures.
Simple, straightforward. Move on!
Leo,
One note: you are looking at turnout and not vote preference. Otherwise, what you describe is fine. We know that party preferences (But do you mean strength of prefernce or direction?) and interest matter. Do they matter across the board? More so among low income vs. high income?
The way to do the latter by the way is to figure out your regression and then partition the sample into income categories (maybe 2 or 3). Then estimate separate regressions and compare the size of the coeffs.
Katrina,
Good to go. main issue is going to be how you want us to code abortion. Keep it in a scale? Into four dummy variables? Think about this.
Also, how to operationalize religious sentiments?
Shamil reporting in,
I would like adress the voting turnout of families with children in the houshold, versus those with just adults. Specifically, I want to see whether there is a similar difference in voter turnout (between families with children and those without) between voters who associate themselves with either the Republican or Democratic parties. Another variable I would like to use is relative levels of income or education.
Shamil,
Looking good. Why do you think number of kids will affect turnout? Not enough time? More investment in the system? What is the direction of your hypothesis?
I would like write about the military and education. More specifically, does military service affect turnout for people who have the same amount of education? I would also like to examine what party people who had military service turned out for vs. non military who had the same education as well as military with different schooling.
pope
I would like to examine the correlation between the general level of satisfaction with the current state of American affairs and voter turnout and then compare these relationships in various levels of expressed patriotism. It seems obvious that people who have a greater love of America are more likely to vote, but how does the level of satisfaction affect participation. Being able to analyse this relationship and then relate it to patriotism seems liable to yield interesting results. Are happy patriots more or less likely to vote than unhappy patriots. Is the relationship the same in people who are more disenfranchised (less emotionally invested in the system). I am not sure what the best way to go about this.
The main relationship of interest is that which occurs between the level of investment in the system and participation, but this investment occurs both based on current issues and belief in the power of the American system, so by examining both types of investment one can better comprehend the correlation.
my name is Evan B.
I'd like to examine the relationship between access/attention to media and voter turnout, especially regarding prevailing concepts of a "liberal" media and TV viewership as relating to R's vote for president.
I'd particularly be interested to see if there were certain kinds of TV programs that affected turnout more than others, or if the turnout according to the type of program varied according to R's political preference or concerns.
GM Bennett
I would like to examine the effects of religious participation on voter turnout. Do voters who are participate regularly in religious rituals vote more often, less often, or just as often as those who do not? Is there a relationship between religious activity and voter turnout? Part of answering this question will be testing the ways in which religion affects party preference and participation, and whether that has any effect on their voting practices.
I am interested in the interaction of media/information access and intensity of partisanship as determinants of voter turnout. Are high information voters without strict party preferences a high turnout group? Are low information voters with strict party preferences mobilized? High/low and strict/non-strict would be examined in finer gradation. It would also be interesting to see whether, as discussed earlier in the course, holders of extreme fringe positions who regard the candidates as distant from their own position have low turnout rates. I'd be curious to see whether distance-from-center would have a bell-shaped impact on turnout: first increasing turnout (starting from the center) as preference for one canidate became more clear, then reducing turnout as voters began to see the canidates as similarly distant from their positions.
-Ben
Ben
all doable but don't get too complicated. A few thoughts.
- to use party strength, "fold" the PID variable. You can either create a new variable by using the recode faciltiy, or always use PID this way:
v3116(r:0=3;1=2,4;2=1,5;3=0,6)
- If you want to include mobilization as well, then estimate turnout by the party strength groups, then control for mobilization (i.e. run a diff of means first for party str groups, then use mobilz as a control var).
Low info + strong party prefs. Now you are really getting into specifics, but you could make an interaction term between educ or pol info x party strength (as defined above).
GM,
You look fine. First, look at turnout by frequency of religious attendance (I think this has no more than seven categories). Then do the MV model using other "standard" controls (educ, income) to see if the religiosity var is a spurious relationship.
Evan,
Good question but unfortunately hard to answer with this survey.
We do have some media usage variables, and you can use these "unpacked" to discriminate btween TV viewers, newspapers readers, etc. I can help you code a special variable if you like.
I think there may be some items asking about talk radio and what news programs they watch (e.g. Fox TV viewers) but you may have such a small sample at that point that you cannot make many inferences.
Kyle,
You seem in good shape.
Biv analysis=satisfaction x turnout.
Biv analysis 2 = satisfaction x turnout, control variable is levels of expressed patriotism.
Then do this all in the regression context. Make sure you include other "standard" controls (income, education, etc)
Meriam,
This is fine. You have to run a set of parallel tables:
military service x turnout, use education as a control variable. Probably collapse education into a smaller number of categories [e.g. v3254(r:1-2;3;4-5;6-7)].
Coding military service is not hard--there are two candidate variables. Don't just look at the R military service; you might try "family member" variable also. One other student has used these variables with some success. Make sure these are coded as dummy (0,1) variables.
Post a Comment
<< Home