Close Menu
    Daten

    [OC] Bedeuten strengere Wahlregeln weniger Wähler? Vergleich aller 50 Staaten (2024)


    [OC] Bedeuten strengere Wahlregeln weniger Wähler? Vergleich aller 50 Staaten (2024)

    Von ptrdo

    View 34 Comments

    34 Kommentare

    1. ptrdo on

      [OC] Do Tougher Voting Rules Mean Fewer Voters? Comparing All 50 States (2024)

      **Gist**

      – Each point is a state.
      – X-axis = cost of voting (COVI)
      – Y-axis = turnout (VEP)
      – Larger dots = smaller margin of victory
      – Color of dots = partisan lean (±5% = swing)

      There’s a modest pattern: states where voting is more difficult tend to have lower turnout—but with lots of exceptions.

      Competitive states (larger dots) often have high turnout regardless of voting rules, suggesting engagement may matter as much as access.

      **Description**

      This chart compares the 2024 Cost of Voting Index (COVI) with 2024 VEP turnout across U.S. states. Each point represents a state, colored by presidential margin (±5% = competitive). Larger points indicate more competitive races.

      Quadrants are split at median values. While there is a modest negative relationship—states with higher voting costs tend to have lower turnout—the pattern varies, and competitiveness appears to play a major role.

      Notably, several high-turnout states remain highly competitive despite higher voting costs, while some low-cost states still exhibit lower participation.

      Interpretation is left open: voting rules may matter at the margins, but they are only one factor among many influencing turnout.

      **Data sources**

      COVI (2024): https://costofvotingindex.com/

      VEP turnout (2024): https://election.lab.ufl.edu/dataset/national-vep-turnout-rates-1789-present/

      Election results (2024): https://www.presidency.ucsb.edu/statistics/elections/2024

      **Method**

      – Merged state-level datasets (DC excluded).
      – Plotted COVI vs VEP turnout.
      – Margin% = 100 * Vote_Margin / Votes_Total
      – Dot size = competitiveness (inverse margin).
      – Color = partisan lean (±5% = competitive).
      – Quadrants split at medians.

      **Tools**

      R (tidyverse, ggplot2, ggrepel), Adobe Illustrator

      **Notes**

      Single-year, correlational. Other factors (competitiveness, demographics, mobilization) also influence turnout.

      **Data**

      | STATE | Votes_Total | Votes_Trump | Votes_Harris | margin_pct | VEP | COVI |
      |—————-|————-|————-|————–|————|——-|——-|
      | Alabama | 2265090 | 1462616 | 772412 | -30.47 | 58.7 | 1.46 |
      | Alaska | 338177 | 184458 | 140026 | -13.14 | 63.55 | -0.17 |
      | Arizona | 3390161 | 1770242 | 1582860 | -5.53 | 63.88 | 0.31 |
      | Arkansas | 1182676 | 759241 | 396905 | -30.64 | 54.06 | 1.35 |
      | California | 15865475 | 6081697 | 9276179 | 20.13 | 62.26 | -2.34 |
      | Colorado | 3192745 | 1377441 | 1728159 | 10.98 | 73.4 | -2.14 |
      | Connecticut | 1759010 | 736918 | 992053 | 14.50 | 67.0 | -0.05 |
      | Delaware | 511697 | 214351 | 289758 | 14.74 | 66.78 | 0.36 |
      | Florida | 10893752 | 6110125 | 4683038 | -13.10 | 67.4 | 0.73 |
      | Georgia | 5250047 | 2663117 | 2548017 | -2.19 | 67.97 | 0.95 |
      | Hawaii | 516701 | 193661 | 313044 | 23.10 | 50.26 | -2.42 |
      | Idaho | 904812 | 605246 | 274972 | -36.50 | 64.18 | 0.32 |
      | Illinois | 5633310 | 2449079 | 3062863 | 10.90 | 63.51 | -1.56 |
      | Indiana | 2936677 | 1720347 | 1163603 | -18.96 | 59.05 | 0.67 |
      | Iowa | 1663506 | 927019 | 707278 | -13.21 | 70.95 | 0.17 |
      | Kansas | 1327591 | 758802 | 544853 | -16.12 | 63.32 | 1.07 |
      | Kentucky | 2074530 | 1337494 | 704043 | -30.53 | 62.9 | 1.06 |
      | Louisiana | 2006975 | 1208505 | 766870 | -22.01 | 61.15 | 0.88 |
      | Maine | 830794 | 377977 | 435652 | 6.94 | 74.55 | -0.48 |
      | Maryland | 3038334 | 1035550 | 1902577 | 28.54 | 69.43 | -0.49 |
      | Massachusetts | 3473668 | 1251303 | 2126518 | 25.20 | 68.44 | -0.79 |
      | Michigan | 5664186 | 2816636 | 2736533 | -1.41 | 74.85 | -0.81 |
      | Minnesota | 3253920 | 1519032 | 1656979 | 4.24 | 76.5 | -0.68 |
      | Mississippi | 1228008 | 747744 | 466668 | -22.89 | 57.75 | 1.94 |
      | Missouri | 2995327 | 1751986 | 1200599 | -18.41 | 64.62 | 1.29 |
      | Montana | 602963 | 352079 | 231906 | -19.93 | 68.35 | 0.71 |
      | Nebraska | 947159 | 564816 | 369995 | -20.57 | 68.61 | 0.31 |
      | Nevada | 1484840 | 751205 | 705197 | -3.10 | 66.15 | -2.32 |
      | New Hampshire | 826189 | 395523 | 418488 | 2.78 | 73.74 | 1.57 |
      | New Jersey | 4272725 | 1968215 | 2220713 | 5.91 | 67.51 | -0.46 |
      | New Mexico | 923403 | 423391 | 478802 | 6.00 | 59.71 | 0.01 |
      | New York | 8262495 | 3578899 | 4619195 | 12.59 | 60.5 | -0.87 |
      | North Carolina | 5699141 | 2898423 | 2715375 | -3.21 | 70.65 | 0.66 |
      | North Dakota | 368155 | 246505 | 112327 | -36.45 | 63.89 | -0.5 |
      | Ohio | 5767788 | 3180116 | 2533699 | -11.21 | 65.4 | 1.35 |
      | Oklahoma | 1566173 | 1036213 | 499599 | -34.26 | 53.46 | 0.93 |
      | Oregon | 2244493 | 919480 | 1240600 | 14.31 | 71.87 | -2.46 |
      | Pennsylvania | 7034206 | 3543308 | 3423042 | -1.71 | 71.44 | 0.49 |
      | Rhode Island | 513386 | 214406 | 285156 | 13.78 | 63.32 | 0.01 |
      | South Carolina | 2548140 | 1483747 | 1028452 | -17.87 | 62.26 | 1.33 |
      | South Dakota | 428922 | 272081 | 146859 | -29.19 | 64.51 | 1.14 |
      | Tennessee | 3063942 | 1966865 | 1056265 | -29.72 | 57.92 | 1.1 |
      | Texas | 11388674 | 6393597 | 4835250 | -13.68 | 56.83 | 1.5 |
      | Utah | 1488494 | 883818 | 562566 | -21.58 | 64.71 | -2.33 |
      | Vermont | 369422 | 119395 | 235791 | 31.51 | 70.75 | -2.32 |
      | Virginia | 4505941 | 2075085 | 2335395 | 5.78 | 71.46 | -0.68 |
      | Washington | 3924243 | 1530923 | 2245849 | 18.22 | 70.23 | -2.43 |
      | West Virginia | 762390 | 533556 | 214309 | -41.87 | 55.56 | 0.04 |
      | Wisconsin | 3415787 | 1697626 | 1668229 | -0.86 | 76.66 | 1.37 |
      | Wyoming | 269048 | 192633 | 69527 | -45.76 | 61.23 | 1.2 |

    2. danSwraps on

      It’s not the correlation between friction and participation that jumps out to me, looks like a super low R squared. However, it’s interesting that the dem and gop points are almost perfectly split by the median cost of voting

    4. 1FellSloop on

      I strongly dislike the shading of the quadrants here. That combined with the inset labels about the dot size and race closeness in different quadrants make it seem like you’re labeling quadrants, which you’re not. Repeating the axis labels on all four sides, bolding the **MEDIAN** labels as well as the axis labels, putting the tick labels inside the graph, all make this much harder to read.

    5. Yongtre100 on

      Cool data. What I’d want to see is if changing to a more or less friction system changes is associated with a change in participation. I feel like turnout has a lot of other influences, so makes sense it’s a bit of a mess even if there is a vague/general trend.

    6. sxyvirgo on

      The *larger* the dots the *smaller* the margin? Help the reader – this is counterintuitive.

    7. Top_Willow_9953 on

      Maybe I’m thicker than most, but I really have a hard time extracting useful information and drawing conclusions from this graphic – even after reading the lengthy explanation in the comments.

    8. Lung_doc on

      Very cool – so it looks like margin may be a stronger influencer of voting. And either Republicans make it hard to vote or when it is hard to vote, those who do vote make things tilt Republican?

      My son moved from Colorado to Texas and was very surprised – there a ballot shows up at your house if you registered. In Texas mail in ballots are reserved for the disabled.

    9. NotAnotherEmpire on

      Wisconsin is an outlier because they have had same-day voter registration forever. So while the legislature put in a strict Photo ID law, the turnout on Election Day always is an incredibly high %. 

    10. Proof-Zookeepergame8 on

      I know what your trying to show, but the biggest thing for me is turnout vs result, and how much lower turnout favors the right. Always heard it, but to see the data is something else.

    11. mfive_ on

      EITHER republican states just have more restricting voting rules OR more friction means more republican results. you be the judge.

      regardless, if we ignore the close races (which clearly increases turnout), then more friction does seem to reduce turnout. or maybe nobody goes since those republican states appear to have big margins of victory anyway.

      looks like a real pickle.

    12. the_answer_is_c on

      I would guess education levels, median age, and close federal elections are the real cause here.

    13. data_daria55 on

      this screams correlation ≠ causation – turnout is way more about how competitive the race is than how hard it is to vote

    14. TahaEng on

      The line in this graph shows a small negative relationship between friction and turnout. But there are three relationships plotted here in a 2-D plot, hard to pull that apart with a single best fit line. The competitiveness of the races appears to be more of a driver of turnout than the friction – the more competitive, the higher the participation, somewhat independent of your friction variable. If you chart only the competitive races, you end up with a fill line that goes the other way – higher participation with more friction, although that is almost a flat line excluding Nevada.

      But I would just ditch the line. There is no meaningful best fit line for a three variable plot like this, and it is pretty noisy anyway. It isn’t doing any meaningful work here, and I think it is somewhat misleading as shown.

    15. FaPaDa on

      Interestingly this implies states where there is very little participation are pretty red while when the majority of the populus votes its usually a blue state.

    16. Lord_Puding on

      As I European from normal functioning democracy and normal election system, the USA definition of „tougher voting rules“ is extremely dumb for me and it suggests its something bad while in fact most of these things are perfectly normal.

      In some USA states you dont even need to have some kind of identification to vote. Like what in the flying 21 century fuck you’re doing. How can you vote without identifying yourself.

      And when someone wants to make that obligatory you call that „tough voting rules“.

      Also, dont get me started on that mail voting thing where you can vote days after the elections even though 99% of population have an access to internet and you can introduce electronical voting.

      Your whole voting system is stuck in 18th century and there does not even seem a will to change that.

    17. acEightyThrees on

      I’m in Canada, and we have to show ID to vote. I’ve been told that the issue in the US is that it’s harder to get ID than it is in Canada, so it’s discriminatory.

      I’d compromise down in the US if I was in government. Pass give the Republicans voter ID laws, and in the same bill, introduce an ID that is free to get for citizens. And increase the number of DMVs (or wherever else you’d get ID) to have a minimum number of locations per population, and have them open a couple nights each week as well as on weekends, so people don’t have to take time off from work to get an ID. Then have the law say that voters can only be ID’d in counties that have hit that minimum number of ID locations.

      Problem solved.

    18. Silver_Smurfer on

      This is a highly correlational chart, and everyone knows that correlation doesn’t prove causation. In order to provide a solid argument for your claim, you’d have to look at voter turnout by state before and after „friction“ was increased/decreased (closest thing to an A/B test given the circumstances). Given that your data includes a friction calculation for the past 30 years, you could calculate a ratio of friction to turnout by year and plot that to try and identify if your hypothesis holds true. However, there are a ton of confounding factors that influence turnout rates and trying to control for them would be nearly impossible. Expanding the time-frame helps, but doesn’t eliminate, that issue.

    19. the_mellojoe on

      As a quick side note: whenever conversations come up surrounding USA voting rules, ask yourself if there is a cost to the voter. Poll Taxes are illegal specifically because they are used to shift voting away from the poorest and most vulnerable individuals and give more power to those with money. If there is a cost to the voter, that must be examined extremely closely to ensure it isn’t a loophole to asses a Poll Tax.

    22. rejeremiad on

      What stands out to me is that swing states have lots of participation regardless of friction and „shoe-in“ states have little participation becuase there is no point for a lot of people to vote if they are just going to be drowned out by the established party.

    23. Monster_Dumps_2026 on

      I was very curious what the cost of voting index could mean and what goes in it. Their website sucks but i found the inputs on the website. I wish they would show which of these aspects caused the biggest difference. Like was it the time off to vote? Polling times? Absentee voting? Automatic Voting? Etc.

      [https://en.wikipedia.org/wiki/Cost_of_Voting_Index](https://en.wikipedia.org/wiki/Cost_of_Voting_Index)

      When i first saw the title i thought it meant actual $$$ of voting. I was like WTH it shouldnt cost more to vote in one state vs another

    24. Prestigious_Carry619 on

      Why make the impact linear? The hypothesis of books like Nudge is making choices slightly easier or harder has an outsized impact on choices. If voting becomes a little harder, a large number of people might just give up.

    25. wrenwood2018 on

      I mean should I care? I can’t tell because the „cost of voting“ index has no meaning. I also can’t see the scale on the y axis. Is it a 20 point difference? 2? One of thoe ranges would matter, one wouldn’t. Also . . . I don’t think the plot is picking up on the most salient property. They are trying to make it a liberal vs. conservative effect, but to me the participation effect is almost entirely how competitive the election was. the „safely GOP“ tend to have larger margins.

    26. RhinoBro33 on

      Ain’t no state that is hitting 70% voting participation… not a single one… the vast majority of people don’t believe that their vote matters and they don’t waste their time voting…. Any state over 50% should be investigated, which is all of them

    27. ZeekLTK on

      I’m just curious what actions were decided to be „friction“ and how values were assigned.

      For example I was expecting to see Maine near the left side („little to no friction“) because you can just show up and register to vote seconds before receiving your ballot. If that is considered middle of the road, what is Oregon and Colorado doing to make it even easier than that? You don’t have to register at all or what?

    28. CupBeEmpty on

      I am just really curious about how they measure “friction.”

      NH are OH are the last two states I voted in. Both were extremely easy to get registered and vote.

      I am curious whey they rank high in “friction?”

    29. Traditional-Storm-62 on

      so what you’re saying is
      harder to vote = more republican

      that sounds like voter manipulation / voter repression

    30. Adept_Duck on

      There appears to be a sizable relationship between closeness of race and participation. This heuristically makes sense. I think if you control for this factor you will see a stronger effect from voting friction.

    33. Chuck_the_Elf on

      To me it looks like the closeness of the race was the greatest contribution. That said I have questions about how the eligible population to vote was determined for each state opposed to the “friction”. I feel like if tighter requirements to register are counted as generation of friction than the two variables are not independent outside of the proposed correlation. If the Voter eligibility was not registration based then was it on adult population? If so you have to consider immigrant and felon populations could skew that.

    Leave A Reply