Skip to main content

View Diary: Renters make good Democrats, and other demographic observations (129 comments)

Comment Preferences

  •  I live in NM-03, Ben Ray Lujan's district. (7+ / 0-)

    Wit the exception of Santa Fe, Rio Rancho and Farmington (Rio Rancho and Farmington are Republican) the district is extremely rural and heavily Democratic. The only area where many people rent would be Santa Fe. My County, which is over 70% Democratic is made up of ranchers, 4-7 persons per square mile.

    I guess we don't fit your model.

    I didn't understand your very first table. Can you explain it for us non-math speakers?

    Thank you for the analysis!

    And even though it all went wrong I'll stand before the Lord of Song with nothing on my tongue but Hallelujah! -Leonard Cohen .................@laurenreichelt

    by TheFatLadySings on Sun Mar 17, 2013 at 02:30:44 PM PDT

    •  Correlations (6+ / 0-)

      are measured from 1 to -1.  1 implies the strongest relationship.  For example, being Black and voting Democrat would have a correlation pretty close to 1.  On the other hand, being evangelical would be pretty close to -1.

      Hope that helps.

      20, CA-18 (home), CA-13 (school)
      politicohen.com
      Socially libertarian, moderate on foreign policy, immigration, and crime, liberal on everything else.
      UC Berkeley; I think I'm in the conservative half of this city. -.4.12, -4.92

      by jncca on Sun Mar 17, 2013 at 03:01:41 PM PDT

      [ Parent ]

    •  non-math explanation of first table (5+ / 0-)

      We are trying to determine what numbers are predictive.  Imagine a variable that means nothing at all, say electric vs gas stoves (no idea if it is actually true).  I can tell you the number in all districts and then the math has no correlation.  It would give you a zero.

      Now, imagine a variable that has everything to do with it and is perfectly predictive.  Lets call it percent of voters for the Democratic congressional contestant in two way vote.  This variable would accurately predict which districts are D and R everytime, the greater the percentage the more likely the D wins until you fall below 50%, then Republicans win.  This variable gives you a 1.  The Republican opposite (voters for R cong ect) would yield a -1.  This number perfectly predicts voting because, surprise, it is voting.

      Statisticians look for the variables where the math yields as close to 1 or -1 as possible.  I'm not 100% remembering but I think that .1 to -.1 are effectively seen as zero.  So that chart is the five strongest variables eachway rated from strongest to weakest.  So renters is the strongest variable from the census data, beating out even white voters, despite being imperfect.

      •  Look at linear combinations not single variates (0+ / 0-)

        Variables may be relatively more highly correlated than others, but not necessarily predictive.  Yes, it seem counter-intuitive, but keep in mind that all such variables are likely to be proxies for the latent, underlying variable that you seek.  

        From your data, it is clear that you really have NO highly correlated variables (an r value of .8 or higher would be generally regarded as strong in most studies dealing with organisms).

        You would do much better to understand the associations among combinations of your variables and then target those populations that best fit that particular combination of co-variates.  

        What you are saying you want to do is to perform discriminant functions analysis or alternatively principal components analysis to better understand the latent variables of most interest, which are in this case far more relevant to the prediction of democrats than any of your variables alone in a series of multiple uni-variate analyses.  This will be especially true if you fail to correct for experiment wise error (the more tests you perform the greater the probability that you will reject a true null hypothesis by chance alone; if you set alpha at 0.05 then 5 out of 100 such tests should be expected to reject the null hypothesis of no difference by chance alone).  

        Use Dunn-Bonferroni or Sheffe tests when multiple comparisons are being made, unless you can a priori identify a smaller number of tests, in which case Fisher's Protected Least Significant Difference Method or Bryant Paulson's modification of Tukey's Honestly Significant Difference Method, all of which can be used for multivariate designs.

        Discriminant functions analysis provides you with at linear combination of variables that provides the greatest prediction.  Principal components analysis would provide you that linear combination of variables that explains the most variance in the sample.

    •  I'll give it a try. (6+ / 0-)

      "Correlation" means that, after many different measurements, it becomes apparent that there is a relationship between two different types of measurement.  The mathematical "correlation" is a measure of how strongly these two types of measurements are related to each other.  Its value can range from +1.0 to -1.0.  If correlation is positive, then as one measurement increases, so does the other.  If the correlation is negative, then as one measurement increases, the other decreases.  The larger the (absolute) value, the stronger the correlation.  If the correlation is zero, then there really isn't any kind of relationship between the two measurements.

      In the very first table, one of the measurements here is always the percentage of votes for the Democratic Congressional candidate in the 2012 election.  The other measurements are the percentages of people who are renters;  who are white;  who are 25-34, and so on.  We see that the largest positive correlation (0.59) with people voting Democratic is for people who rent.  So, districts that have a lot of renters tend to send Democrats to Congress.  On the other hand, the largest negative correlation with voting Democratic (-0.58) is with the percentage of white people in the district.  So, the whiter the district, the more likely the Congressional Representative is a Republican.

      I hope this helps.

      -5.13,-5.64; If you gave [Jerry Falwell] an enema, you could bury him in a matchbox. -- Christopher Hitchens

      by gizmo59 on Sun Mar 17, 2013 at 03:12:37 PM PDT

      [ Parent ]

    •  Don't be too sure (6+ / 0-)


      Look at Española, Chama, Pecos, most of rural Rio Arriba county, Ojo Calliente, Truchas, Taos, and Questa for examples. Most of those people are renters, not in apartments but in all those adobe casas and trailers.   Santa Fe is too damm expensive for renters anymore.  

      If I remember correctly that district is mostly Democratic until you get east of the Pecos, then it turns into the Plains farmland.  I will say, having grown up there, Northern New Mexico is different, everyone talks politics, and pays attention to what is going on.  

      ... the watchword of true patriotism: "Our country - when right to be kept right; when wrong to be put right." - Carl Schurz; Oct. 17, 1899

      by NevDem on Sun Mar 17, 2013 at 03:30:01 PM PDT

      [ Parent ]

    •  I vote in NM-02, which is not too different. Lots (2+ / 0-)
      Recommended by:
      Eric Nelson, TheFatLadySings

      of sparsely populated areas, but more Republican than northern NM. You guys have done a great job of electing progressive Dems and meanwhile we're stuck with that jackass Steve Pearce. I'll jump for joy if/when he's deposed.

      The world is not interested in the storms you encountered, but did you bring in the ship.

      by Hanging Up My Tusks on Sun Mar 17, 2013 at 03:36:27 PM PDT

      [ Parent ]

      •  difference (3+ / 0-)

        seems to be higher citizenship rates among Hispanics, and lots of Native Americans too.  NM-02 Has lots of Hispanics, but not many vote.

        ...better the occasional faults of a government that lives in a spirit of charity, than the consistent omissions of a government frozen in the ice of its own indifference. -FDR, 1936

        by James Allen on Sun Mar 17, 2013 at 05:52:51 PM PDT

        [ Parent ]

        •  In northern NM the Hispanics tend to be (6+ / 0-)

          descendants of the old Spanish land grant families who settled the area long ago. In southern NM, the population tends to be Mexican and more recently arrived in the state. You're right about Native Americans; they're predominantly in the northern half of the state. My observation of sameness was essentially that the entire state has huge areas with little population.

          I remember visiting OH relatives as a child and thinking that every 10 miles there would be another small town, whereas in NM you might go 50+ miles before reaching another town. Lots of wide open spaces, punctuated by spectacular skies.

          The world is not interested in the storms you encountered, but did you bring in the ship.

          by Hanging Up My Tusks on Sun Mar 17, 2013 at 07:10:33 PM PDT

          [ Parent ]

    •  The previous commenters (3+ / 0-)

      have already, collectively, done a good job of explaining it, but, yes, the left column is the five variables that are most strongly associated with voting Democratic (at the House level), and the right column is the five variables most strongly associated with not voting Democratic. 1 or -1 indicates the strongest possible relationship; 0 indicates no relationship at all.

      So the fact that nothing here goes far beyond 0.5 or -0.5 shows that none of these are especially strong relationships in terms of actually predicting voter behavior (as some other stats-minded commenters have correctly pointed out, elsewhere in this thread), but they do at least point us in the right direction and get the conversation started.

      Editor, Daily Kos Elections.

      by David Jarman on Sun Mar 17, 2013 at 04:35:25 PM PDT

      [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site