Survey of suburban Detroit district raises questions of accuracy
If you find articles like this valuable, sign up for our daily email newsletter and support us with a donation.
In a new survey for the Deadline Detroit news website, polling group Target Insyght (TI) shows Michigan's 10th District Congressman John James (R-Farmington Hills) trailing Carl Marlinga — his 2022 opponent, and a 2024 Democratic hopeful — by a substantial margin.
Conducted July 14-15, the poll of 400 District 10 "people" — with a plus or minus 5% error factor — reveals some head-scratching numbers that spark questions about the survey's reliability.
The first point to discuss is the sample itself. While the 400 respondent number is certainly an adequate sample size for a congressional district, using the term "people" — as opposed to "registered voters" or "likely voters" — suggests that the sample is comprised of both voters and non-voters. The term "people" might even call into question whether all of the respondents were even adults. Additionally, there is no disclosure as to how this poll was conducted.
Another red flag is the poll's error factor — at plus-or-minus 5%, the figure is unusually high.
Together, the sample of "people" and the +/- 5% reported error factor indicate the survey was conducted online without personal interviews.
The FiveThirtyEight data organization rates the 10th District, a northeastern Detroit suburban division, as R+6. The Dave's Redistricting App statisticians find a different trend, calculating a partisan lean of 49.5D–47.9R. And yet, the TI survey finds former Judge and Macomb County prosecutor Carl Marlinga (D) leading Rep. James 49-43% — a surprisingly large margin.
Based upon actual voting behavior, neither FiveThirtyEight's rating, nor that of Dave's Redistricting App, would suggest a challenger could develop a lead of this size against an incumbent House member without an associated scandal.
In another example, the survey finds Rep. James' favorability rating at 29:52%, favorable-to-unfavorable — a poor showing. It is hard to believe that even a group of swing district respondents would rate the incumbent Congressman so badly. Conversely, Marlinga, who lost District 10 in a close race in 2022, has a favorable-to-unfavorable rating of 42:18%. If he is so well-regarded, and James so poorly thought of, why didn't Marlinga win — handily — in 2022?
Furthermore, despite having to wade through a four-person Democratic primary scheduled for August 6, Marlinga raised only $141,000 in the second quarter, with receipts of just over $600,000 for the entire election cycle. Compare these financials to those of Rep. James, who raised $1.55 million in the second quarter, and more than $6.1 million for the election cycle. If James' favorability is as low as this poll suggests, he would not have the credibility to raise such sums.
Another difficult-to-understand (and believe) pair of data points are the racial preferences. According to the Target Insyght survey — and it is not delineated how many respondents are of which race — Rep. James leads among the black participants, 35-29%. This, within the group that constitutes the bedrock of the Democratic Party. Among whites, a segment that now trends much more heavily toward Republicans, the TI poll finds the segment breaking 54-40% for Marlinga. This type of racial voting ratio is exactly the opposite of what is found in most districts and states.
Another topsy-turvy finding is the gender breakdown. Women, who traditionally favor Democrats in large margins, lean here toward Rep. James, 47-46%. Among men, who traditionally favor Republicans, Marlinga holds an abnormally huge lead — 53-38%.
Furthermore, there is an additional inconsistency when comparing these results with another firm's poll of Michigan's open 7th District; a seat anchored in Lansing, but with similar partisan ratings as CD-10.
District 7 is now open because Rep. Elissa Slotkin's (D-Lansing) district is running for the Senate.
In a poll of the 7th District taken just days before the Target Insyght 10th District survey, Noble Predictive Insights found former state Sen. Tom Barrett (R) — the 2022 nominee who lost to Rep. Slotkin, 52-46% — leading former state Sen. Curtis Hertel (D), 48-41%. This result better reflects the District 7 partisan voting characteristics than does the TI showing for District 10.
Most of the information found in the Target Insyght poll simply doesn't make sense and should be considered unreliable. Expect the James campaign to soon counter this data with a poll of its own in order to set the record straight.
Jim Ellis is a 35-year veteran of politics at the state and national levels. He has served ss executive director for two national political action committees, as well as a consultant to the three national Republican Party organizations in DC, the National Federation of Independent Business, and various national conservative groups.
Born and raised in Sacramento, California, he earned a B. A. in Political Science from the University of California at Davis in 1979. Jim raised his daughter, Jacqueline, alone after his wife died following a tragic car accident. He helped establish the Joan Ellis Victims Assistance Network in Rochester, NH. Jim also is a member of the Northern Virginia Football Officials Association, which officiates high school games throughout the region.
I distrust most polls. Who is taking the polls, who are in charge of the polls - too questionable for me to trust any of them.