[EL] State by State party ID breakdown

Thu Dec 3 07:48:36 PST 2015

David

Well, there is an hour of my life I’ll never get back!  The things you do for an editorial board member ...

This is a great question, and I wish I could find a better answer for you. I’m honestly surprised that that data providers that I list below have not made easier to produce.  

I am assuming for the moment that you are a non-technical user, I apologize if that is the wrong assumption, but the point is that data providers should, in my view, service the non-technical as well as the technically sophisticated audience.   In fact, I’d say it is more important to serve the non-technical audience.  Charles Stewart and I can always find the answer to this kind of question, but in some respects it’s a lot more important for journalists, advocates, and heck regular old citizens to be able to easily find some reliable data that speaks to some sort of social and political concern.

Sorry if I am editorializing …

The reason you cannot easily find the answer you want, as Charles sort of implied in his response, is that most national surveys are not designed to produce reliable samples at the state level.  So you will see below that if you had asked “What is the balance of partisan identification over the past 40 years,” there is a very easily accessible source to answer your question.  But you asked about STATE level estimates, and that’s where it gets more difficult.

In each listing below, I provide a link to an online data analysis tool and tell you why you cannot produce what you want.

1) Cooperative Congressional Election Study: is a large (N=35,000-50,000) survey designed to produce state level estimates.  The 2012,, 2010. and 2008 versions are publicly available.  

I went here first because this SHOULD be a place to relatively easily obtain the information you want.   BUT the problem is that the online analysis tool (“Two Ravens”) is nearly impossible to use, at least I could not figure out how to use it.  My memory is that it used to be easier and have more exploratory tools, now there are lots of advanced tools—every multivariate analysis tool you’d want—but producing a simple crosstab of party ID x state seems to be impossible.  Someone will surely correct me if I’m wrong, but I’m at a loss.  

	URL of the 2012 dataset: https://dataverse.harvard.edu/dataset.xhtml?persistentId=hdl:1902.1/21447 <https://dataverse.harvard.edu/dataset.xhtml?persistentId=hdl:1902.1/21447>
	URL of the series: https://dataverse.harvard.edu/dataverse/cces <https://dataverse.harvard.edu/dataverse/cces>	 <https://dataverse.harvard.edu/dataverse/cces>
	URL of the data exploration tool: https://rserve.dataverse.harvard.edu/dataexplore/gui.html?dfId=2500154& <https://rserve.dataverse.harvard.edu/dataexplore/gui.html?dfId=2500154&>

—> This by the way is what I would recommend, download data subsets of the 2012, 2010, and 2008 CCES, but you’re still going to have to process these with a statistical program.  If go this route, the variables you want are named weight, inputstate, pid7, and pid3. 

2) National Exit Polls: These are housed at the Roper Center (https://ropercenter.cornell.edu/ <https://ropercenter.cornell.edu/>) and can also be used to obtain state level estimates of partisanship among those who turned out to vote.  Roper Center has created an online data analysis tool called the Roper Explorer that provides a nice interface to do various kinds of analyses, although not all surveys are integrated into the Explorer system.  Your academic institution needs to be a member to use IPoll and the Explorer system. 

Problem?  The NEP have been restricted as of January 1, 2014.  Perhaps a list member may know why this has become the case; I don’t think is used to be the case.  If you gain access, I’m not sure they are accessible via the Explorer system.  Its very likely that you could find state level polls to construct state level party ID but this would be a very tedious exercise. 

	URL of the 2008 NEP: https://ropercenter.cornell.edu/CFIDE/cf/action/catalog/abstract.cfm?label=&keyword=USMI2008-NATELEC&fromDate=&toDate=&organization=Any&type=&keywordOptions=1&start=1&id=&exclude=&excludeOptions=1&topic=Any&sortBy=DESC&archno=USMI2008-NATELEC&abstract=abstract&x=32&y=9 <https://ropercenter.cornell.edu/CFIDE/cf/action/catalog/abstract.cfm?label=&keyword=USMI2008-NATELEC&fromDate=&toDate=&organization=Any&type=&keywordOptions=1&start=1&id=&exclude=&excludeOptions=1&topic=Any&sortBy=DESC&archno=USMI2008-NATELEC&abstract=abstract&x=32&y=9>

3) The National Election Study and the General Social Survey: both are very high quality national polls conducted in presidential election years and off years, though the NES off year coverage stopped after 2008 (if memory serves) due to funding.  Both are NOT intended to provide state level estimates, sample sizes are between 1500-2500.

These surveys are some of the best to get quick and very reliable answers to a lot of questions about public opinion on a wide variety of political, social, and economic issues.  The have been conducted for decades, so if you wanted to know how the public’s attidues have changed about abortion, or gun control, or the state of the economy, all of this is readily accessible.

And best of all, in my view, both have been integrated into a powerful and I think relatively easy to use online data analysis system (“SDA”).  It’s not point and click, but requires no specialized software.  

	URL for the GSS and NES at Berkeley: http://sda.berkeley.edu/archive.htm <http://sda.berkeley.edu/archive.htm>

Below is a graphic I produced in 30 seconds demonstrating that the level of Independent affiliation has not continued to grow, contrary to what many in the media continue to assert (it grew in the 1970s but has remained relatively stable or declining since then).  

4) Reuters Election Polling: I mention this only because it is a very slick graphical interface that lets you quickly plot Reuters polling data on a variety of subjects and using a variety of controls and filters.  No state level data.

	URL: http://polling.reuters.com/#poll/TR130/ <http://polling.reuters.com/#poll/TR130/>

Long and long?  I can’t help you!  

tl;dr

Happy holidays,
Paul g. 

---
Paul Gronke	

Professor, Reed College and
Daniel B. German Endowed Visiting Professor, Appalachian State University

Director, Early Voting Information Center
3203 SE Woodstock Blvd
Portland OR 97202

EVIC: http://earlyvoting.net

 <http://people.reed.edu/~gronkep/36E051EA.asc>
> On Dec 3, 2015, at 8:42 AM, Schultz, David A. <dschultz at hamline.edu> wrote:
> 
> Hi all:
> 
> Does any one know of a database that has already organized state by state  population party identification  or affiliation over the last few election cycles?   (I.e. of those surveyed, what percentage identify as Democrat, Republican, or independent).   I would prefer to be able to look at presidential election cycles over the last few elections but also including congressional (midterm) is fine too.  What I have in mind is the exit poll data that does the state by state breakdown of partisan affiliation.  I know I can go back and go to CNN or other sites and create the data myself but before I do that I want to see if anyone has already done that.
> 
> Thank you.
> 
> -- 
> David Schultz, Professor
> Editor, Journal of Public Affairs Education (JPAE)
> Hamline University
> Department of Political Science
> 1536 Hewitt Ave
> MS B 1805
> St. Paul, Minnesota 55104
> 651.523.2858 (voice)
> 651.523.3170 (fax)
> http://davidschultz.efoliomn.com/ <http://davidschultz.efoliomn.com/>
> http://works.bepress.com/david_schultz/ <http://works.bepress.com/david_schultz/>
> http://schultzstake.blogspot.com/ <http://schultzstake.blogspot.com/>
> Twitter:  @ProfDSchultz
> My latest book:  Presidential Swing States:  Why Only Ten Matter
> https://rowman.com/ISBN/9780739195246/Presidential-Swing-States-Why-Only-Ten-Matter <https://rowman.com/ISBN/9780739195246/Presidential-Swing-States-Why-Only-Ten-Matter>
> FacultyRow SuperProfessor, 2012, 2013, 2014
> _______________________________________________
> Law-election mailing list
> Law-election at department-lists.uci.edu
> http://department-lists.uci.edu/mailman/listinfo/law-election

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20151203/e69aa469/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-1.tiff
Type: image/tiff
Size: 60550 bytes
Desc: not available
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20151203/e69aa469/attachment.tiff>