What should we do in response to the several emails we received today about Enos?
I think we have no other choice but to act. We need to respond to these emails and say we are investigating.
AJPS editorial board insiders
-
What should we do in response to the several emails we received today about Enos?
I think we have no other choice but to act. We need to respond to these emails and say we are investigating.Exactly. This is out in the open now and can no longer be swept under the rug. Even those who don't understand R code can look at Enos' data file and official data in parallel, and understand that hundreds of precincts were either deleted or had turnout and/or vote values changed.
-
What should we do in response to the several emails we received today about Enos?
I think we have no other choice but to act. We need to respond to these emails and say we are investigating.Exactly. This is out in the open now and can no longer be swept under the rug. Even those who don't understand R code can look at Enos' data file and official data in parallel, and understand that hundreds of precincts were either deleted or had turnout and/or vote values changed.
This shows there are data problems, it doesnt show fraud. I dont really have a dog in this fight but the "smoking gun" would be showing results differ in a replication using the full dataset.
-
What should we do in response to the several emails we received today about Enos?
I think we have no other choice but to act. We need to respond to these emails and say we are investigating.
Exactly. This is out in the open now and can no longer be swept under the rug. Even those who don't understand R code can look at Enos' data file and official data in parallel, and understand that hundreds of precincts were either deleted or had turnout and/or vote values changed.This shows there are data problems, it doesnt show fraud. I dont really have a dog in this fight but the "smoking gun" would be showing results differ in a replication using the full dataset.
There are data problems that Enos is aware of and hasn't addressed. He provided no R code for how he produced the datasets containing so many problems, including values that don't exist in official data.
The results would differ since the deleted precincts just "happen" to be high turnout and/or high Republican vote in close proximity to former housing projects. That's what the maps in the report show. In other words, he deleted data points that were unfavorable to his hypotheses.
And that's just one dataset with problems. Another is the individual-level voter data. The effect sizes and confidence intervals changed across paper versions, and all changes are in the direction of Enos' pet theories.
Too many implausible coincidences, and Enos has remained silent for nearly 3 months. Remember, the onus is on him to show these were honest errors.
-
Now THAT really is something!
A successful professor at Harvard caring more about his own career than those of others!
OH MY GAWD!!! SO HEINOUS!!!
AAAAAAAAAAAHHHHHHHH!!!!!!!!!! AAAAAAAAAAAAHHHHHHHH!!!!!!!!!
There is also a serious misreading of Ryan in these allegations. Ryan doesn't care about social justice or racism- Ryan cares about Ryan and advancing his own career. Ask any grad student at Harvard while he was DGS how much Ryan cares about other people.
-
I'm trying to help here because I think there is a real chance Ryan fabricated something, just knowing the type of person he is. But the evidence I have seen people discuss involve data problems that could reasonably just give the same results. You can't prove Ryan made anything up, most likely, but if you are able to show his conclusions are wrong it doesn't matter what his intent is; the article should be retracted.
That said, whatever Karl is doing with his WN crusade about racism not being real is hurting the cause a lot. He isn't very smart and has a clear agenda. People do not want to be associated with him for good reason.
-
Now THAT really is something!
A successful professor at Harvard caring more about his own career than those of others!
OH MY GAWD!!! SO HEINOUS!!!
AAAAAAAAAAAHHHHHHHH!!!!!!!!!! AAAAAAAAAAAAHHHHHHHH!!!!!!!!!
There is also a serious misreading of Ryan in these allegations. Ryan doesn't care about social justice or racism- Ryan cares about Ryan and advancing his own career. Ask any grad student at Harvard while he was DGS how much Ryan cares about other people.My point is he is a nasty person. The idea that Enos fabricated data to advance some social justice cause is laughable to anyone who knows him. The idea that he fabricated data to advance his own career makes sense.
-
Remember, the onus is on him to show these were honest errors.
Remember, only a completely deIusional nuttter would think the onus is on any academic to answer a shjtflood juveniIe accusations of random internet troIls like KarI or his anonymous gaggle of high school dropouts.
-
Remember, the onus is on him to show these were honest errors.
Remember, only a completely deIusional nuttter would think the onus is on any academic to answer a shjtflood juveniIe accusations of random internet troIls like KarI or his anonymous gaggle of high school dropouts.
Nah, that's not what happened.
-
^^
Enos fabricated data to appear to advance social justice because anything woke gets published these daysApparently, you're having a pretty tough time finding anyone else on planet Earth who comes to this same conclusion aside from KarI and his handful screeching Chymps.
-
The latter.
Apparently, you're having a pretty tough time finding anyone else on planet Earth who comes to this same conclusion aside from KarI and his handful screeching Chymps.It's in the report, sweetie.
If someone sent a basic report showing the data problems, then a replication of Enos models and how the results changed with stripping the WN nonsense from it, I am sure there are some reputable people who would put their name to it.
-
Remember, the onus is on him to show these were honest errors.
Remember, only a completely deIusional nuttter would think the onus is on any academic to answer a shjtflood juveniIe accusations of random internet troIls like KarI or his anonymous gaggle of high school dropouts.Nah, that's not what happened.
Nah? Then who, exactly, is demanding responses to Karl's shjtflood of unsupported accusations, other than Karl and a handful of anonymous Chymps on the internet?
-
I haven't read the report. Is the problem that his effect sizes are bigger, or are that there are no/trivial effects when all precincts are included? The former is basically nothing, but the latter is a problem worthy of retraction.
Why do you write lengthy comments without having read the report?
There are two separate sets of issues. One is that on the individual voter data effect sizes increased both through the addition of observations coming out of nowhere, and the deletion of others for no apparent reason. The second is that on the precinct-level data one third of data points are missing or altered, and those data points contradict his pet theories. Adding back the missing precincts and fixing the incorrect data values would mean >1,000 observations that contradict Enos' hypotheses would be added to his dataset, thus invalidating all his results (including the EI and matching analyses).