WHAT TO MAKE OF EDUCATION POLLING RESULTS
Seems there have been an unprecedented number of education polls this year - 2016 began with the release of an Albertson Foundation poll, which was closely followed by a survey by BSU's Public Policy Department, A poll from Dan Jones and Associates for Idaho Politics Weekly, and another by BSU. Along the way, we asked for the help of Patinkin Research Strategies with our own "Customer Satisfaction" poll.
In each of these polls, questions were asked about similar educational issues, with somewhat different phrasing. And...they yielded very different results!
Quality of Idaho Schools
As indicated, the year began with the Albertson Foundation "Landmark" Survey, a statewide poll of 1000 citizens with an error margin of +/- 3%.
Here's question 2 from the Albertson Foundation survey (pink numbers represent percentages):
So, not really encouraging. Only 7% were totally satisfied with their public schools. But at least 72% thought the schools were "okay" or "in very good shape".
Both BSU surveys asked a somewhat similar question. Here are the results of the question on the first survey.
On the first BSU survey (1000 respondents, +/- 3.1%), about 28% of the respondents rated the quality of Idaho's k-12 education as "Good" or "Excellent". When we add in "Fair", the percentage jumps to 65%, seven percentage points lower than the Albertson survey.
Dan Jones and Associates conducted education polling (601 respondents, +/- 4%) for Idaho Politics Weekly in August, and found decidedly different results. Fully 66% of respondents rated public schools a "4" or "5" on a five-point scale, a "favorable" rating. Another 19% rated public schools a "3"(neutral), meaning that 85% were either neutral or responded favorably.
Further, Jones and Associates found that only 55% expressed faith in charter schools, in almost direct opposition to the Albertson Foundation poll, which found that 80% "strongly favored" or "somewhat favored" charter schools.
The results of the second BSU poll came out in September. The statewide results for the exact same question were different than those in BSU's first poll.
In the second BSU poll (1000 respondents, +/- 2.3%) 39.2% rated the quality of Idaho's K-12 schools as "Good" Or Excellent" compared with 27.6% in the first poll. And the percentage rating K-12 schools as "Fair", "Good" or "Excellent" was 71%, compared with 65% in the first poll.
Quality of Local Schools
The Albertson Foundation poll asked a question about local schools, then used all of the responses to aggregate a statewide "local public schools" grade from A to F.
This poll found that 42% of respondents would give local public schools a grade of "A" or "B", and 78% would give a grade of "C" or above.
The second BSU poll asked a similar question, but identified the responses related to particular school districts and in the Treasure Valley, and used the same "Excellent/Good/Fair/Poor" response matrix used for their overall education quality question.
If one considers "Fair" and above as similar to "C and above" the BSU survey found a similar percentage of responses in those categories to the Albertson survey. In fact, 76% of respondents answered "Fair", "Good" and "Excellent" as to the quality of Treasure Valley public schools on the BSU survey, and 78% responded "A", "B" or "C" on the Albertson survey.
However, the BSU survey found that 50% of Treasure Valley respondents rated their schools "Excellent" or "Good", while the Albertson Foundation survey found that 42% of respondents statewide rated local schools "A" or "B".
In May, the Boise District conducted its "customer satisfaction" survey with Patinkin and Associates. The poll was conducted with 600 patrons, and had an error margin of +/- 4.9%. Here are survey responses to a question that attempted to get at the same kind of quality assessment among Boise patrons.
On this question, 83% of respondents respondents chose "Very well" or "Pretty well" as to how the statement "Does a good job with limited resources? describes the school district. The percentage rose to 92% when we subtracted "I don't know" from the totals and recalculated.
There have been some interesting articles lately about factors which can influence polling results. Since the articles are about election polling, some of the references don't apply to the polls done this year on education. But some do.
From the New York Times' Upshot research blog, and article by David Rothschild and Sharad Goel entitled "When You Hear the Margin of Error Is Plus or Minus 3 Percent, Think 7 Instead" explores assumptions built into polls and weighting effects. From CNBC, David Schoen writes about the presidential elections polls in "Here's how to make sense of the latest Clinton, Trump poll results".
Schoen writes,"Why do the margins vary so widely from one poll to another? The short answer: Opinion polling includes a lot of guesswork and assumptions, and pollsters make different choices when setting up their surveys."
And, for some fun reading, here's another Upshot article, by Nate Cohn, that describes the undue influence that weighting might give some groups or individuals in "How One 19-Year-Old Illinois Man Is Distorting National Polling Averages".
And, finally, here's an interesting article published in Boise Weekly (originally in Propublica) about faulty ballot design. You may recall that we had some issues with ballots in the 2014 School Board elections. These examples are amazing.