Thanks for doing this work.

Welcome to the party, Hillary supporters.

Thank you so much for this information. What can we do? Everyone deluge email/call/protest Obama to look into it? Change.org petition to Obama? Getting it in main stream media? I think we actually have some republicans, media, etc. to help, no one who’s using their brain wants reactive Trump to be president and have his hand near the nuclear buttons. I believe this info could change things. (From moment one 11/8 after Trump “chosen” I wondered why no one talking about election fraud/rigging).

Thank you!!!

im curious as to what can be done with this and do you forward it to any election officials in those states so it can be investigated ? I mean whats the point of sharing this with us if you arent going to do anything with the clear truth of fraud ?
This is a pretty big deal , we are looking at a leader who’s practices runs parallel with Hitler
Millions of marginalized communities are at risk right now ,, and if there is a clear path that can show that these things need to be investigated then it should also be your responsibility to forward it to the proper authorities in the GOP and DEM parties ,,

Mr. Soares, I have a couple of questions for you. I’m a business journalist in New York, but I’m not reaching out to you on behalf of any news organization. I’m just asking as a concerned citizen.

First, let me say I’m with you in spirit. I’m asking these questions only to get a better understanding of your work and to get an idea of how compelling this story might be to a broader audience.

Re: the exit poll data — I went to the Edison Research website. They don’t publish a state-by-state list of the winner. Instead, they publish data on voter characteristics — gender, ethnicity, age, etc.

You say in your “about” section that you used the “gender” results to arrive at an overall winner. I assume that means you used the percentages of men and women who voted for HRC and Trump to get an overall figure of how many votes each one got.

I understand using the gender category, since there can only be two genders, and everyone has to be one the other — theoretically, the gender number should be all-inclusive.

But how do you know that the gender number wasn’t adjusted?

Edison says that they routinely adjust all their exit poll data — in other words, they “conform” it to the actual results — in order to come up with results they think is more truly representative of how people actually voted.

An article published on Raw Story in April quoted an Edison staffer saying two things about their process that make me wonder if their data can be used to create an assumed overall result, as you’ve done.

First, he said the “voter characteristics” exit poll that they do for US elections is not designed to catch voter fraud — it has 20 questions on it, and the sheer length dissuades some people from taking it, which skews the results.

In contrast, an exit poll that’s designed to catch voter fraud, which is used in some foreign elections, typically only has one question on it — “Who did you vote for?” — and gets a much higher response rate.

Second, he said all their data is adjusted, which suggests that the overall results numbers you’ve assumed might not be right.

Here’s a clip from that story:

http://www.rawstory.com/2016/04/on-tim-robbins-election-fraud-and-how-nonsense-spreads-around-the-internet/comments/

I asked Joe Lenski, executive vice president of Edison Media Research, about all of this. Edison has conducted all of the exit polls for major U.S. media organizations since 2003, and Lenski has also done exit polling in Azerbaijan, Georgia and Venezuela.

As for using his results to suss out fraud, he says that American exit polls are “just not designed for that type of precision. They’re surveys, and like any other survey, they have a margin of error. The precision that a lot of these people are talking about just doesn’t exist with our polls.”

In emerging democracies, says Lenski, “the exit polls are designed specifically to catch any manipulations of the vote count, and also to bring some transparency so voters can trust the vote count. They have a lot more locations in the sample, they do a lot more interviews and they use a much, much smaller questionnaire. In some cases, they just ask, ‘who did you vote for?’” A brief questionnaire, he explains, increases the response rate. “The more interviews you do, the more locations you cover and the shorter the questionnaire, the higher response rate you’ll get, and that all leads to a much smaller margin of error.”

The news organizations that sponsor our exit polls are just looking for a sense of who voted, and what motivated them to vote. They use longer questionnaires – typically with about 20 questions – and it takes a little longer for voters to fill them out. Lenski says that while the methodology hasn’t changed much since exit polling was first introduced in 1967, the fact that news organizations post preliminary data in real-time leads to “a lot of commentary on social media.”

When people claim that early exit poll data are “unadjusted,” they’re wrong. Edison adjusts its data throughout the day to compensate for “non-response rates and other sampling issues that come up when we conduct the survey.” When someone is approached to take a survey and says ‘no thanks,’ Edison’s pollster notes that individual’s gender, race and approximate age. They then adjust the raw data to match the demographics of the people they saw voting at that location during the day.

The early data news organizations post on their websites has already been adjusted. In states with large numbers of absentee voting, the data are also merged with phone interviews conducted before Election Day. If the number of absentees is significantly higher or lower than expected, they adjust for that difference as well. And then, finally, the data are adjusted again to match not the results of the election statewide, but the vote tallies at the specific polling places they surveyed.

Lenski stresses that pre-election polls are also adjusted to conform their samples to what pollsters know about the populations they’re trying to measure. The irony of all of this is that the adjusted data are far more accurate than the raw data.

“Remember the process here,” says Lenski. “In most of these states, we’re talking about having interviewers at several dozen locations. The vote returns at those locations are used in the model, and the vote counts at the county level are used in the model, so if there were some sort of fix involved, there would have to be hundreds or thousands of people manipulating the data at all of those locations to make it work.”

So, Mr. Soares, I wonder how would you respond to these points? Do you think it’s statistically valid to use an assumed number for your state-by-state results that’s based on adjusted data for gender? And do you think the fact that these Edison exit polls were long and time-consuming could have skewed the results in way would make your assumed number invalid?

Thank you in advance for your response.