Let's expand on what the PA Democratic Party actually did. The Party itself filed a law suit leading up to the election against the State Secretary running the elections and the election board of every county in the state, including those run by Democrats. The Secretary was appointed by Democrats, working in a Democratic administration. They were on both sides of the suit.
You say this as if it's suspicious, but it's not. Both American parties are absolutely enormous organisations with control of innumerate local and state institutions. It's almost inevitable that oftentimes two such institutions will have conflicting priorities. Republicans have challenged other Republicans over voting rules and laws too; it's not a stitch up.
And there is no claim that they started setting up election sites believing it was already legal based on the 2019 laws, they were suing the state for injunctive relief because the 2019 laws didn't allow for what they wanted to do.
They started the suit that allowed for satellite election offices in July. The Secretary, a member of the political party that was suing, appealed to the State Supreme Court for a ruling a month later in August. Philadelphia was already advertising satellite offices before that appeal to the Supreme Court was made. They moved forward with a policy that they filed suit over because it wasn't legal well before even officially asking the Court to make a determination. And it went to September until the Democrats that had the majority on the court decided to give the go-ahead.
The only thing here that's untoward is distributing advertising stuff that pre-empted any decision by the court. I can understand producing it, because such materials will always need to be produced ahead of time, and are frequently ordered even if their ultimate need isn't a sure thing. But distributing it was certainly wrong.
Hardly as important as having adequate infrastructure to accommodate those who wish to vote. And also, this simply isn't a reason to deny the request itself.
Once again: Trump also requested extended hours in PA, at late notice, elsewhere. Which was granted.
Then perhaps consult them about this. You will not have difficulty finding a number of studies and books that refute idea that turnout benefits the Democrats.
Which is why Republicans are absolutely desperate to restrict voting wherever and however they can, because they know with how Trump has screwed everyone over, their party will get absolutely annihilated in the midterms if every American citizen is allowed to keep their rights.
It was entirely within their power though. Taking actions in which you have the power and reaching a result which advantages you in the acquisition of that power looks strongly of corruption, at minimum. They used their statewide powers in ways that offered advantage to the places that give them that power. Even if that happened by coincidence, it's still bad. So perhaps, stop treating it like it's good.
Indeed, I have: the general consensus among political data analysts is that higher turnout broadly benefits Democrats more than Republicans.
The first listed author is a political science professor in California with degrees in mathematics and political science who has been called on to help redraw away racial gerrymanders. I'm sure you're not upset with that sourcing, despite the implication of "what I have in mind", though I will of course take issue with your habit of just declaring a consensus supports you without having ever seen a single piece of evidence...
We now wish to briefly review what is known empirically about the answers to these three questions.
Even the first and easiest of these questions to answer, whether nonvoters are more Democratic in their voting preferences than are voters, is more complicated than it might first appear because the answer appears to have changed with recent elections and may not be the same across different types of elections. Nonetheless, as we review the previous research evidence, we have concluded that the answer to the first question, “Are low turnout voters more likely to vote Democratic than high turnout voters?”, has largely been settled, at least at the level of presidential voting. That answer has historically been YES (Miller and Wattenberg, 1982) but, in recent presidential elections, a bandwagon effect sweeping the less politically aware voters toward the winning candidate, combined with weakening Democratic party loyalties among lower income and education white voters, has made nonvoters marginally more Republican in candidate preference than voters (Texeira, 1992: 87; Gant and Lyons, 1993). Thus, at the presidential level, the present answer to that first question now appears to be NO.
With respect to the second question, although the previously published studies (e.g., DeNardo, 1980, 1986; Tucker and Vedlitz, 1986; Texeira, 1992; Radcliff, 1994) do not agree on the answer, in our view the answer to the second question: “Are elections with higher turnout higher in Democratic vote share?” is, in fact, generally NO, at least for congressional and legislative elections. For example, Grofman, Collet, and Griffin (1995) generated the correlations between change in turnout (among registrants) and change in Democratic vote share for each pair of California Assembly elections at time t and time t + 4 that take place within the same redistricting cycle.18 Only one of the six pairs has an adjusted r2 above .02, and that value is only .06. Thus, as expected, the results do not show Democratic gains when turnout goes up.19
With respect to the third question, “If turnout were to have increased in some given election, would Democrats have done better?”, we believe, quite simply, that this question remains unanswered. In our view, none of the previous research on the link between turnout and Democratic success used a research design that would legitimately allow this question to be resolved. Instead, earlier empirical studied did not really address this critical third question, but answered questions one or two instead.20
The idea that Democrats benefit from increased turnout largely justifies itself from the idea that non-voters lean Democrat. That is already a contentious claim in itself, as that is more or less true in different times and different places. That source above indicates that the party preference of non-voters measured at some times are different than those at other times, and they wrote that in 1999, long before Trump got involved. They mentioning weakening loyalties of the lower income voters to the Democratic Party in 1999, imagine the impact of that now, in a world where self-identified independents are nearing the combined total of the two major parties and a ton of previous non-voters sided with Trump out of nowhere in 2016.
But even if we accept that to be true, if we pretend the population of non-voters always favor Democrats, that does not therefore mean that increased turnout increases voter share for Democrats. They are, as those analysts above suggest, logically independent claims. The former does not necessitate the latter. And when looking at the data of whether than trend exists in practice, they say unambiguously that "the results do not show Democratic gains when turnout goes up."
And, being experts in the field, given their research and the research of many others to work with, they decline to even give a stance on the question "if turnout were to have increased in an election, would Democrats have done better." If experts doing research cannot find a basis from which to make that claim, from where do you get yours? I welcome your answer independent of my opinions, but Wolf has characteristically given me a perfect prompt for my own conjecture below.
Which is why Republicans are absolutely desperate to restrict voting wherever and however they can, because they know with how Trump has screwed everyone over, their party will get absolutely annihilated in the midterms if every American citizen is allowed to keep their rights.
This is the whole thing. This is why people say turnout helps Democrats. In a rational world, the world you both want to exist, the flow of information would go like this:
- Democrats perform better with higher turnout
- People observe this phenomenon
- Democrats try to increase turnout and Republicans try to suppress turnout based on the above facts
- You all condemn Republicans for voter suppression
And if all those things were true, I'd be right there on number 4 with you, but there isn't any evidence that the top part is true. Rather, people believe it is true because they logic in the opposite direction:
- People view Republicans as opposition they want to condemn
- The desire to condemn leads to accusations of voter suppression
- The accusation requires a motive, which leads people to believe voter suppression benefits them
- The belief that voter suppression benefits Republicans leads to the belief that higher voter turnout benefits Democrats.
And now the world is awash with all sorts of hit pieces about how Republicans are suppressing voters for their own gains. The silliest of these accusations is about voter purges. Google voter purges, you will find hundreds of articles about how voter purges are an unfair an undemocratic attempt to suppress voting rights, with either the implication or the explicit claim that it's a tactic to benefit Republicans. But the current federal laws around voter registrations were established by the National Voter Registration Act of 1993 (https://www.congress.gov/bill/103rd-congress/house-bill/2), which explicitly establishes the removal of ineligible voters from the rolls, including on the basis of neither voting nor confirming their address across two election cycles. There are whole organizations dedicated to condemning this practice by Republicans, but the practice was signed into law by Bill Clinton, on a bill written by Democrats, sponsored by a Democrat, and co-sponsored by 72 other Democrats. Every state purges its voter rolls, they are obligated to by law that was written by Democrats as recently as 1993, but god forbid there's a Republican Governor in Georgia taking part, cause that's just blatant disenfranchisement to benefit the Republicans.
Seems the disgusting Elonjuchend showed no regret for all the lives they ruined by having Musk loot the government and cripple its ability to provide for the US. Almost as if they're complete sociopaths who a real government should jail for treason once the ability presents itself.
Deposition videos are providing a look at two people responsible for the largest termination of federal grants in the National Endowment for the Humanities' history.
Yes, states are constitutionally empowered to determine their own election procedures, a fact which seemed to be lost on the President, and which you seem to consider suspicious.
Taking actions in which you have the power and reaching a result which advantages you in the acquisition of that power looks strongly of corruption, at minimum. They used their statewide powers in ways that offered advantage to the places that give them that power. Even if that happened by coincidence, it's still bad. So perhaps, stop treating it like it's good.
Whether it constitutes corruption obviously depends on what those actions were, and whether they were inherently unfairly preferential to one candidate. In this case the action is providing easier voting options, in areas in which the existing infrastructure was plainly insufficient. Options that are open to everyone regardless of affiliation.
If easier access to voting "advantages" Democrats more than Republicans, perhaps you should consider why that is, rather than insinuate unfairness you cannot substantiate.
Now, if you want to substantiate the unfairness, you should be able to provide an example of somewhere where expanded access was just as sorely needed, would have benefitted the Republicans, and was refused. Of course, I've pointed out several times that when Republicans did request expanded access elsewhere, it was granted.
OK, so the article there is from 1997, with the studies on which they base these conclusions are from 1992 and 1993. Bit more recent: in 2024, non-voters identified Democratic over Republican by ~24%, and preferred Harris to Trump by ~20%, according to the CES. https://data4democracy.substack.com/p/did-non-voters-really-flip-republican
OK, so the article there is from 1997, with the studies on which they base these conclusions are from 1992 and 1993. Bit more recent: in 2024, non-voters identified Democratic over Republican by ~24%, and preferred Harris to Trump by ~20%, according to the CES. https://data4democracy.substack.com/p/did-non-voters-really-flip-republican
I have. I can see that you assumed, without data or any more recent analysis, that the factors they identified back then would mean their conclusions were still perfectly applicable today.
So, you added some speculation about continuing relevance to their findings from >34 years ago.
I have. I can see that you assumed, without data or any more recent analysis, that the factors they identified back then would mean their conclusions were still perfectly applicable today.
So, you added some speculation about continuing relevance to their findings from >34 years ago.
Data analysts, specifically ones who are experts in elections, wrote a paper explaining that the question of whether nonvoters lean in support of a certain party is logically independent of whether higher turnout would benefit that party. Their assertion is not dependent on the age of the data. Your response was that you found more recent data showing non-voters prefer Democrats, which by their assertion, does not offer a basis to conclude that higher turnout would benefit Democrats.
It's not a "their data vs yours" scenario, you are making a logical jump that experts just explained to you is invalid.
Data analysts, specifically ones who are experts in elections, wrote a paper explaining that the question of whether nonvoters lean in support of a certain party is logically independent of whether higher turnout would benefit that party. Their assertion is not dependent on the age of the data. Your response was that you found more recent data showing non-voters prefer Democrats, which by their assertion, does not offer a basis to conclude that higher turnout would benefit Democrats.
It's not a "their data vs yours" scenario, you are making a logical jump that experts just explained to you is invalid.
I provided that newer information for a specific reason: the section you quoted began with a paragraph about how the non-voter affiliation with the Democrats had weakened or disappeared. I reasonably assumed that was therefore part of the argument you were making, so it was relevant to point out that the affiliation has reasserted itself more recently (and quite strongly).
Onto the logical independence. As you say, the authors state that the basic questions "Do nonvoters prefer the Democrats" and "Would higher turnout benefit the Democrats" are logically independent.
The simplest explanation for this they present is:
"Thus we see that even if the answer to question 1 is yes, whether the answer to question 3 is yes depends exactly how the turnout increase comes about. Even under the simplest possible assumptions given above– ones involving linear relationships, and a uniform distribution of voter characteristics– the way in which changes in h(v) are linked to changes in m and b is quite complex".
Which essentially means that it depends on distribution and motivation, as I've already acknowledged. Under the "simplest possible assumption" of a uniform distribution, it would indeed benefit the Democrats, and the strength of such a benefit would be affected greatly by the partisan bias among nonvoters. Indeed the strength of that bias would also affect the likelihood of higher turnout benefitting Democrats in more complex situations, so it remains relevant.
Which essentially means that it depends on distribution and motivation, as I've already acknowledged. Under the "simplest possible assumption" of a uniform distribution, it would indeed benefit the Democrats, and the strength of such a benefit would be affected greatly by the partisan bias among nonvoters. Indeed the strength of that bias would also affect the likelihood of higher turnout benefitting Democrats in more complex situations, so it remains relevant.
To get to the conclusion that bias among nonvoters would impact the likelihood of turnout benefitting Democrats in more complex situations, you are making at least two implicit assumptions:
1) That the nonvoters who can be enabled/persuaded to vote are similar in leaning to the whole of nonvoters. Imagine the scenario where there were millions of borderline anarchist libertarians who chose not to vote on principle. Between the two, they would likely report preference for Republicans, but that preference is irrelevant if they would never agree to vote, and thus the population of people who could be gained as voters would support the Democrats more. Consider the general sentiment of users on this forum, you have plenty of examples of people who would prefer the Democrats every time but still despise them and are not going to vote for them in almost any scenario, and that changes who exactly we are talking about. Polling all non-voters is not the same as polling non-voters who could be both interested in voting and willing to vote for the major parties, which is the group you'd actually want to consider as far as impact on electoral success.
2) That the act of voting itself is not connected to changing opinions. You are assuming that non-voters turning into voters would just be a transference of their existing opinions into the voting pool. But when they answer their question 2, they conclude that there's no real evidence of that, that gained voters in high turnout elections tend to be reflective of the existing voting population, not the non-voting population. That could be just that the subset of people I discuss in my point #1, those who could be interested in voting, reflect the voting population, and the non-voting population leans further left because of a bunch of lefties who will never vote. It could also be that the action of voting, of considering the parties and their positions with the intention of choosing between them, leads people to the opinions that the voting population holds, rather than the opinions of the non-voting population.
Regardless of the underlying reason, something we just may not be able to know, the point is that polling the opinions of nonvoters does not tell you what would happen if they had voted, and the results of elections at different turnout rates do not support the assumption that increased turnout changes the results based on of the pre-existing preferences of non-voters.
No, if I'm discussing likelihoods in the absense of any additional information about distribution, i need no such assumption.
Say we have a pool of 100 non-voters. As in the 2024 sample, 60 are Democrat-identified, and 40 are Republican.
We take 10 from the 100 to represent the subset that can be 'activated' or persuaded to vote. With no data on those 10, it is essentially a random sample of the 100. Each is likelier to fall within the 60 than the 40.
2) That the act of voting itself is not connected to changing opinions.
Again, in the absense of any meaningful information on whether it hugely diverges, an assumption is not required. Its just another "maybe, maybe not, and also maybe the other way" possible confounding variable.
So shuffle the deck again before you draw your 10 from the 100! The 100 still contains 20 more blues than reds, and the pick is still random as far as we can judge.
Except that we have many, many elections of varying levels of turnout and no established discernable pattern demonstrating the outcome that you're predicting. It's not illogical to imagine that the effect of increased turnout would depend on the opinions of the non-voters, but something sounding reasonable is not a substitute for evidence, and there isn't direct evidence tying turnout to Democratic Party success. It's just truthiness.
Except that we have many, many elections of varying levels of turnout and no established discernable pattern demonstrating the outcome that you're predicting.
Which is something you'd only expect to see if that one factor was enough to override every other factor. We can see no discernable pattern if we plot election results solely against evangelical turnout, either, but its a fact that they vote strongly Republican. A thousand other factors are also present.
That's not true. If you were only considering wins and losses as data, sure, but we have access to the voting numbers, we can calculate the parties' performances to fractions of a percent, if there were a significant enough effect on the results from increased turnout, we could see the trend over the course of enough elections. It wouldn't be a nice linear graph with Democrats getting vote share perfectly proportionally to turnout, but there would be a at least a tendency for the high turnout end of the graph to favor Democrats more.
Regardless, you are at best making the argument that we can't know the effect of increased turnout, you are not justifying the claim that it seems to benefit Democrats.
Why? The trend might be submerged beneath 50 others.
We see no significant correlation between evangelical turnout and Republican success. Not because evangelicals don't favour Republicans. But because 50 other factors occlude that one.
Regardless, you are at best making the argument that we can't know the effect of increased turnout, you are not justifying the claim that it seems to benefit Democrats.
I am arguing that if you have a pool of 60 blue and 40 red, and pull at random, it is more likely to be blue. There might be other factors affecting that pull-- in either direction-- but their impact is purely speculative, so we're back to random.
I am arguing that if you have a pool of 60 blue and 40 red, and pull at random, it is more likely to be blue. There might be other factors affecting that pull-- in either direction-- but their impact is purely speculative, so we're back to random.
To give an analogy, imagine 10 people apply for a job and 8 of them had no related experience, the other 2 did. If you pick randomly, chances are high that the final hire would have no related experience. In actuality, it is not random, and the people with experience are much more likely to get the job. Treating it as random when it isn't leads to the wrong conclusion, it's a fallacy.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
