Tag Archives: Emerson College Polling Society

The Roy Moore Polling

By Jim Ellis

Left: Former state Supreme Court Chief Justice Roy Moore (R) | Right: Ex-US Attorney Doug Jones (D)

Nov. 16, 2017 — Last evening’s political news featured heavy coverage of a new National Republican Senatorial Committee poll that produced bad news for Alabama GOP special Senate nominee Roy Moore, but not enough information was released to determine the reliability aspect. In fact, the NRSC didn’t even release the name of the polling company with which they contracted to conduct the survey.

This is significant because their finding that Democrat Doug Jones is leading Judge Moore, 51-39 percent, is clearly out of line with every other poll taken in the corresponding time frame. It is also interesting that they would even release such a poll considering the Republican candidate is doing so badly. It does, however, justify their previous position of cutting the Moore Campaign from additional funding because of the sexual impropriety allegations against the candidate that has dominated recent news coverage.

Below are the post-scandal publicly reported surveys as complied by the Ballotpedia website:

• National Republican Senatorial Committee (Nov. 12-13)
No pollster credited
Jones 51%; Moore 39%
500 Sample

• Fox10 (Mobile) (Nov. 13)
News Strategy Research
Moore 49%; Jones 43%
3,000 Sample (Automated)

• Emerson College Polling Society (Nov. 9-11)
Moore 55%; Jones 45%
600 Polling Sample

• JMC Analytics & Polling (Nov. 9-11)
Jones 46%; Moore 42%
575 Sample

• Decision Desk HQ (Nov. 9)
Moore 46%; Jones 46%
515 Sample

As we can see, the NRSC poll returns the most inconsistent results in comparison to the other available data during the same time frame; the period just after the Moore sexual scandal broke.

Additionally, because the NRSC did not release the name of their pollster or the survey methodology, not enough information exists to determine if their data are skewed in any particular manner.

The lack of available information does not necessarily mean that the Senatorial Committee’s results are inaccurate. It is curious, however, that the other results — and, all have larger sample sizes than the reported NRSC calling universe — finds much different ballot test margins.

Interestingly, the Fox10 poll from a local Mobile television affiliate, which is the latest released survey prior to the NRSC study, and the Emerson College Polling Society find the complete opposite result and their methodologies utilize much larger sample sizes within the studied polling grouping. The Fox10 3,000-person sampling universe clearly suggests that the questionnaire responses were obtained through an automated device, but such does not necessarily mean this poll is less accurate than the live operator polls.

The Alabama race continues to deteriorate, and it is becoming more evident that Jones is now in a strong position to win. But, despite all the negative news coverage, this phantom NRSC poll is the only one that shows him trailing badly.

The other survey to find him dropping behind, from JMC Analytics, featured a sampling universe where 56 percent of the respondents are female, a potential skew in Jones’ favor since this subset broke his way, 46-40 percent. In comparison, men favored the Democrat only 46-45 percent in the JMC crosstabs. Therefore, with a sample where the Democratic-leaning female sector was over-sampled by approximately five percentage points, correcting this skew likely brings the Jones’ 46-42 percent ballot test result back into a tied range.

The NRSC results and partial poll release is intriguing to say the least. Hopefully, we will see more substantiation of their data later today.

More Troubling Signs in Alabama

By Jim Ellis

Sept. 14, 2017 — Now just 13 days away from the Sept. 26 special Alabama Republican Senate run-off election two new polls have entered the public domain, likewise bringing bad news for appointed Sen. Luther Strange.

The senator has been languishing in surveys to varying degrees since the Aug. 15 primary election. All show him trailing, but the ones with sampling groups comprised primarily of evangelical voters find him down by very large margins, well into double-digits. Unfortunately for the Strange camp, these new polls project former state Supreme Court Chief Justice Roy Moore to be holding equally large margins. They, however, do not feature strong evangelical-based sampling universes.

Strategic National (Sept. 6-7; 800 registered Alabama voters) polled the Alabama electorate and surveyed a respondent group that is 84 percent Republican with all saying there is minimally a 50/50 chance that they will vote in the run-off election, and the overwhelming majority claiming a much larger participation preponderance (81 percent certain to vote; 13 percent very likely; 6 percent, 50/50). Alabama does not register voters by political party; hence, the sample correctly included a fraction of self-identified Democrats and Independents who are eligible to cast ballots in a partisan run-off election.

Continue reading →

Finally, a Montana Poll

By Jim Ellis

April 27, 2017 — While the Montana at-large special election has been heating up with both major party candidates approaching the $2 million mark in campaign receipts and each heavily spending on media, we had surprisingly not seen any polling data … until yesterday.

The lack of polling led some analysts to believe the race was trending toward Republican Greg Gianforte. Democrats, normally quick to release survey numbers that favor their candidate, had been unusually quiet about nominee Rob Quist’s ballot test status. GOP strategists typically tend to play their polling cards closer to the vest, but often publicize survey data in response to what they see as embellished numbers for the opponent.

It’s possible that such an argument scenario has some validity in this instance. The Emerson College Polling Society made public their recent survey totals that post Gianforte to a major advantage over country rock singer Rob Quist (D).

Continue reading →

Senate Still in Limbo

By Jim Ellis

Nov. 3, 2016 — Entering the last week of campaigning, the Democrats are on the cusp of re-claiming the Senate majority they lost in 2014, but virtually no competitive outcome is yet secure.

The latest Hillary Clinton email revelations may cause irregular Republican turnout to increase, which should help the GOP Senate candidates. A demoralized Republican voter base, thinking that Donald Trump would have no chance to prevail against Clinton, is about the only way Democrats could have gained a wave effect, but that is no longer expected.

It appears that nine of 10 Democratic in-cycle states will remain in party control. Only Nevada is competitive on their side of the ledger. Republicans look to have 15 safe seats of their own, with another five: Arizona (Sen. John McCain), Iowa (Sen. Chuck Grassley), Georgia (Sen. Johnny Isakson), Florida (Sen. Marco Rubio) and Ohio (Sen. Rob Portman) all trending either strongly or nominally their way.

Democrats are in favorable position to convert incumbent Republican states in Illinois (Rep. Tammy Duckworth-D, unseating Sen. Mark Kirk-R) and Wisconsin (former Sen. Russ Feingold-D, re-claiming the seat he lost to Sen. Ron Johnson-R in 2010), in addition to being favored in the open Indiana seat (former Sen. Evan Bayh-D ahead of Rep. Todd Young-R).

Continue reading →

Differing Data

By Jim Ellis

Oct. 17, 2016 — The presidential map has swung significantly toward Hillary Clinton in the past week, which is of little surprise considering the revelations surrounding Donald Trump. If the election were today, our count projects Clinton to receive 338 electoral votes as compared to only 200 for Trump.

As has been the case since the beginning of this campaign, in order to win the national election Trump must carry the states of Florida, Ohio and North Carolina in order to develop a base that melds into a winning coalition. Before the videotape flap, Trump held the advantage in his three staple states. This week, however, he has fallen behind in each place, albeit by small, single-digit margins.

While it is mandatory for Trump to carry Florida, Ohio and North Carolina, Clinton can win the national election without any of the three. But, should Trump rebound fully in the aforementioned, he is still not done. In addition to carrying the 23 states that have voted Republican in every presidential election in this century – all are unanimous with the one exception of Indiana, which voted for President Obama in 2008 by one percentage point — Trump needs an additional 17 electoral votes in order to actually win the election.

Continue reading →

A Centennial Swing

By Jim Ellis

Sept. 28, 2016 — Even before the first presidential debate was complete, we began seeing some political movement particularly in one critical battleground state.

In the 21st Century, the states of Florida, Ohio, North Carolina, Virginia, New Hampshire, Iowa, Colorado, and Nevada have been traditionally regarded as the swing battleground pool in the presidential race. In the last two elections, all but North Carolina voted Democratic. Such a pattern was continuing to take hold in Virginia, Colorado, and Nevada, though the 2013-14 elections did show Republican gain. Most of this particular shift, however, was attributable to voter turnout patterns instead of any ideological shift toward the GOP.

Now in the presidential general election, the political tide is beginning to turn in several of these states. Colorado, a place that had clearly been trending Democratic in the previous few elections and appeared poised to easily vote for Hillary Clinton earlier in the cycle is now exhibiting signs that Donald Trump is at least in position to contend for the Centennial State’s nine electoral votes.

Continue reading →

Bloomberg’s Polling Report

By Jim Ellis

July 5, 2016 — Often, people want to know which are the most accurate pollsters within the huge volume of data being released into the political marketplace. Bloomberg Politics reviewed the 2016 presidential pollsters who routinely release their numbers into the public domain and publicized their ratings late last week. They isolated a dozen polling firms and corresponding media organizations that picked the presidential primary winner in at least 70 percent of their studies. Polling frequency and reliability stats were also recorded.

The firm projecting the primary winner most often — 95 percent of the time — was Gravis Marketing, which conducted 19 surveys in 12 different states. The students who comprise the Emerson College Polling Society in Massachusetts, who have fared very well in previous performance review studies despite being amateurs, placed a close second in win predictability, proving correct 94 percent of the time. They polled 16 times in eight states. Completing the top five are Opinion Savvy (91 percent winner accuracy; 11 polls; eight states), NBC/Wall Street Journal/Marist College (90 percent winner accuracy; 29 polls; 15 states), and the Survey Monkey (90 percent winner accuracy; 10 polls; eight states).

But capturing the correct victory margin proved to be another story for all 12 pollsters. None of them came within the average polling error margin pertaining to victory spread projection in relation to their cumulative polling average.

Continue reading →