There have been a number of surveys that sampled Grandview resident’s opinions on the school facilities. Not all were done by the school. This is what I know about the surveys, and some advice for some parents who ran their own survey. I added a new section at the end to show how easy the school survey could have been hacked.
Surveys should always be viewed skeptically, both because of the small sample size, and the information they might be pushing (Push polls are a well understood way to inject opinions into the public mind). They are useful when they show an overwhelming percentage – like the 75% that said the school board should not be moving kids out of Stevenson, or replacing the high school.
I will end this post with some discussion on the integrity of online polls. Short version – don’t believe that polls on the internet are worth much, no matter what the company selling them tells you.
The school polls
The school board has run two public polls so far (August 2017), and one focus group meeting for “empty nest parents”. There was a third separate poll done for High school students (although there was nothing stopping those students from posting in the other online surveys). These were administered by a company called Triad Research.
As of June the school has paid Triad at least $17,000 for the online surveys and the focus groups. Triad’s summary of the surveys and the focus group is on the school website (the Pdf at the bottom).
The online surveys were poorly designed, identifying the owner of the poll is only done with one line at the start. The body of the survey contains nothing but a series of questions, with no tracking of the progress. You can know that you are on a school owned survey by looking at the domain name up in the address bar, they used “sawtoothsoftware.com”, subcontracting the online polling service.
The First sawtoothsoftware survey was posted online in the first week of May, it was located at (this now closed URL).
There were 597 responses, the questions were mostly about the original 7 options for school facilities, as presented in the April 26, 2017 meeting. The survey only asked about those seven original plans, there was no “fill in your own idea” for the school facilities. The $35 million renovation plan was the least expensive option given.
The board implied with a question in this survey that there may be a deal in the works to turn Stevenson into a “community center”, but no council member had knowledge of any plans for the use of the building by the city. The plan to vacate Stevenson is not part of any current school plans, but the school board still has the option to ignore the recommendations.
Second survey was located at this address (now closed).
Once again, the survey used push polling to try to force parents into choosing from the three facility options the school, and pushed the idea that $44 million was a base number for renovations, implying that the $35 million renovation option was inadequate.
The results of the second school survey are going to be posted on the school website after the Aug. 3 meeting. (Update Aug 10 the second survey results have still not been posted on the web page where they said it would be, instead you need to go to the community planning homepage, and find it at the bottom of a long page).
An important fact – the data from the surveys was only summarized in the posted PDF files, there has been no release of the raw data. Because the company that conduced the survey is a private business, they have no reason to release that data. FOIA requests don’t work on private businesses. Maybe this is why the school chose to farm out work that could have been done internally?
The Focus group – and are 90 year old buildings obsolete?
The school paid 11 older “empty nest” community members to attend a focus group in May of 2017. This was done because they know that older people are least likely to respond to the online surveys. The small size of the group made it unreliable for any true view of the general group of voters in Grandview.The group had the expected confidence in the quality of the school, and fear of raising taxes. Maybe the most surprising finding was that none of the group ever went to the school website, so all of the school’s attempts to push for building new schools online will do nothing for this group. (I also assume this group will not be reading my blog).
One item from the focus group jumped out at me. The school has been pushing hard on the the idea that 1. most people don’t know the age of the schools, and 2. they would be willing to replace them if they know the age of the buildings.
I think this quote from a member of the focus group, composed of older community members who have no children in the school, is the answer the average Grandview resident will give about the age of the schools.
“90-years-old — you’ve got to tear it down? Well, is somebody going to buy my house that’s almost 100-years-old and tear it down? No. They’re going to fix it, they’re going to renovate it, and they’re going to make it look beautiful.”
The Parent Poll that used Clout Research
There was a very short survey summary of a facility phone poll that was released by somebody named Cindy Morgan back in May 2017. It was released to a list of emails addresses that had been gathered from parents in the schools. I have no idea who did this poll, but given the timing I assume that this was done by a parent group that looked at the facility options released in the first community meeting, and didn’t trust the school to have an unbiased survey.
Culp said the Clout poll was not done by the school. He said that the last question on the survey, which said that “complete renovation or rebuilding school facilities could cost homeowners up to almost $600 per $100,000 of home value” was not an accurate number.
Culp said nothing about the survey results which showed more than 80% of the respondents would like a “tax neutral” option to be available from the school in future meetings. I’m not sure what that means, but I’m guessing they wanted an option that could fix the schools up with no increased taxes. That would be possible, as current PI levies expire, there could be money freed up by replacing them at the same total mills level. There is also some new money from inside mills created by the last tax assessment. This would be way less than the school board wants, but the school could scrape by on it.
A message to the parents group
The parent group that conducted the phone survey has kept their identifying information off of the survey. I am not a current parent, so I don’t know how well the members of this group are known in the schools (since this is a small town, I’m guessing the names of the group members are circulating in private conversations). I understand why they don’t want to go public, there is a small but nasty group of school board supporters who will go on the attack when they see anyone who is expressing opposition to the board.
The school board and its supporters will be pushing back against any school group that expresses opposition to the expensive plans to come. They will argue that the board has the best consultants, the best parents on the facility task force, and therefore the parents who oppose them are uninformed cheapskates.
Don’t waste a lot of time trying to argue school building logistics with the board, they will have numbers that will support whatever they are pushing (like a new building at the middle school location). They made the numbers, and can make them do whatever they want.
There is no expert on the question of “How high should taxes be in Grandview?”. We are all experts on our own household expenses, and what we can afford to pay in taxes. We can look at the total property tax rate, and see that Grandview has the highest rate out of all the school districts in the county.
The question that has to be answered by each of us is, “how much tax is enough? We are the experts. It is up to us to answer that question, not some high priced consultant the school board has hired.
Don’t trust the online survey
There are lots of questions about the push polling that was done in the school’s online survey, but let’s look at the question of “can the survey be fooled with false data?”. The video from the June meeting showed that the administration has no idea of how the security holes make the survey unreliable.
At 1:16:00 in the meeting, Culp said the survey used by the school (the Sawtooth survey) was not protected from multiple votes from the same computer. Culp said that the “company that did the survey eliminated responses from the same “URL” , but this doesn’t make any sense – URLs are addresses on the Internet, they don’t correspond to users. I think the word Culp was searching for was “IP addresses” , which are mostly unique to a household (but might be the same for an entire business with hundreds of employees). Culp said that two or three surveys from the same “URL” would not trigger deletion of the result, but some number higher would cause some kind of reaction – he didn’t say if it would delete all results from that house, or only allow some.
When asked why a password system could not have been used to insure that only one response to the survey could be made per Grandview resident, Culp said that it wasn’t possible. This is where he showed his tech ineptitude, a password system could have been used, it would have been harder to set up, but it was well within the capability of any professional survey company. The reason it was not done probably was because the school wanted “cheap”, and they thought that “security through obscurity” would protect the survey results from anyone who was casually trying to game the system. Any tech person with a small amount of skill would know that using multiple IP addresses would allow the survey to be filled with as many fake votes as you wanted.
Culp expressed confidence that the results of the school survey were accurate and not gamed, but if his confidence was based on his lack of tech knowledge and some BS from the survey company, it was not something the residents of the city should believe.
Surveys are cherished by the number-focused bureaucrat. They love to argue that a 5% difference in a poll makes a significant difference. This is folly. In the end, the vote for the tax levy is the only number that matters.
How the survey could have been skewed
Here is an example of how the results of the survey could have been skewed by one determined person. The second survey had 61 survey responses that asked for “Other”, non-school approved, something that was not the three option. That was said to be 14% of the survey respondents. Remember, a survey respondent is not one person, it is one completion of the online survey.
The survey was supposed to block multiple respondents from one IP address if it was higher than some undisclosed number. Let’s supposed that number is six.
One person could have gone to 10 coffee shops and fast food places, anything with free wifi. At each of them they could have filled out the survey six times, say they vote for the number one option. That would have added 14%, bringing the number of votes for the first option up to 29%.
Maybe that one person was determined to make a difference, and went to 30 coffee shops. That adds 42% to the first option, bringing it up to 57%. It would then beat the 54% who voted for option three, and make the number one option the winner.
The school board wanted to use the unsecured online polls, even though it is obvious they can be faked with minor effort. Was there a good reason they chose the online poll?