Tried to Tell You - CheckYourself Survey is Wrong to Use in Seattle Public Schools - Part 1

Part One of this blog post will be about the article but in Part Two, I hope you jump in with a discussion about student data privacy, what's the line on with data collected to help students with mental health issues (and who gets to see it) as well as other issues with this kind of data collection happening in schools. 

I do want to note from the outset that it was the Seattle Times who first filed for this information, not parents.  (All bold mine.)

 

This report is from The 74 and was picked up by the Seattle Times. Here's the headline:

Seattle-Area Schools Say Deeply Personal Survey Saved Lives. Then They Released Student Data

Since 2018, more than 36,000 students across the Seattle region have shared their hopes, fears and family secrets in an online questionnaire called Check Yourself. 

Supported by a voter-approved tax levy, King County, which encompasses Seattle, has spent more than $21 million on the program since 2018. The funds help pay for mental health counseling for students and to track trends across the 13 districts that participate. Seven schools in Spokane County, in eastern Washington, and a few districts in Oregon also use Check Yourself.

Districts promise students their answers to over 50 personal questions will be kept confidential. But a group of parents has been able to obtain reams of sensitive survey data from multiple districts through the state’s Public Records Act.

Stephanie Hager, whom I have worked with on this story for years, is a human bulldog. Her nose for the multiple issues raised by the use of the CheckYourself survey and her tenacity in seeking answers may, in the end, save the privacy rights of Puget Sound students and their families.

One of them, Stephanie Hager, is on a six-year crusade to expose what she considers to be the program’s lack of privacy safeguards. To prove her point, the former Microsoft program manager said she correctly identified six students based on nothing more than details they provided in the survey and a simple Google or social media search.

“We know their school, gender, age on a certain date, grade level, language they speak, their dogs’ names, friends’ names, race, their unique interests, what sports they play, if they are religious, and anything else they feel like writing in — plus their whole mental health record,” said the Snoqualmie Valley mother of four, whose son took the survey in 2019.

 “I can’t imagine any parent saying OK to that.”

We now live in a time and space where data is king. And, for entities like schools, the more the better on the premise that it will help schools know kids. But does it?

Online mental health surveys also face backlash from activists and lawmakers, who find them frequently intrusive, inappropriate and removed from school’s main purpose. 

“Schools are really under a huge amount of pressure to address student mental health,” said Isabelle Barbour, a consultant who developed a school-based mental health program for the state of Oregon. “But when they try to adopt something that can work in their setting, it brings up all of these other pressure points around privacy.”

What DO parents get to know about this screener?

Parents get two chances to opt their children out of the screener, and students can also decline to complete it on the day of the survey. But districts reveal nothing that would alert anyone to its potential risks. Quite the contrary. One district promotes it as a “successful, proactive approach to providing support to students.” Another promises “personalized feedback and strategies for staying healthy.”

Promises are made that the data is "secure" and that "potentially identifying" student data will be removed as well as saying that county officials and researchers only see anonymous answers. 

But an investigation by county ombudsman Jon Stier, triggered by parents’ concerns, suggests this hasn’t always been the case. A report released last summer revealed that in the program’s early years, county officials were able to connect student names to their responses, although Stier said that practice has ended.

A handful of districts concealed some personal details. But several redacted little, if anything.

Privacy experts say that wiping information such as race, home language and favorite activities from a document in order to make it truly anonymous is no easy task. But without such measures, a combination of answers could identify a student, in the language of the law, “with reasonable certainty.”

Hager, Check Yourself’s most outspoken parent critic, obtained an email thread through an open records request that shows officials were well aware of the survey’s potential privacy pitfalls. In one email, a former Tickit Health executive warns county officials that if a student “were to enter identifiable information in the free-text sections, theoretically this would be accessible.”

Use of 13 open-ended questions may lead students to writing incredibly revealing details about themselves. Why that is still in the survey is a mystery to me.

For example, during one recent school year in the Kent School District, there was just one student from a certain tribe in Washington State. That student stated that fact in an open-ended question which then allowed someone to know who that student is.

Now King County is HAPPY to pass off these privacy problems embedded within the survey to the districts. 

Marc Seligson, a King County spokesman, insisted that “student data security is paramount,” but that responsibility for interpreting privacy laws falls to the districts.

She  (Margaret Soukup, King County) said she was shocked districts released records to parents. “I was very upset because I didn’t even think that that was a possibility.” 

The 74 also contacted Cari McCarty, a University of Washington researcher who helped develop the survey and now evaluates the King County program. She said districts are obligated to protect “the confidentiality of student information,” but directed further questions to the county.

These statements are a little frightening coming from people who created the survey and who oversee the program.  

One section of the story that I do not love relates an uptick in public disclosure requests to right-wing groups like Moms for Liberty. While that uptick may have come more from the Right in recent years, you can't lump all these requests into one category of people. 

Then there's this which I consider shocking naive:

Officials in King County, and from six districts that responded to a request from The 74, said they’ve received no reports of cyberthieves or child predators gaining access to Check Yourself and using results to target students.

You know, most child predators don't tell you when they are doing their stalking. 

Then there's this from districts as well:

They point to internal data showing that students feel more connected to school when they’re referred to an “intervention” after taking the survey. In focus groups, students expressed “favorable opinions” about the screener. In a survey of almost 400 students referred to a staff member after completing Check Yourself, the percentage saying that an adult at school listens, cares and tells them they do a good job increased. 

And in those focus groups there were no "unfavorable opinions?" C'mon!

But there have been district officials trying to sound the alarm:

In 2019, an official in the Tukwila district, south of Seattle, wrote in a report to the county that the survey was “causing considerable angst” and that with many “vulnerable” and “traditionally marginalized” families, educators didn’t want to “create unnecessary harm.”

That same year, a Seattle school counselor called it a “super personal survey,” according to an email The 74 obtained through a public records request. She questioned why the district needed the information and whether it would be able to keep it confidential.

 How is the survey taken?

Ali, the former student who found Check Yourself beneficial to her well-being, had a distinctly nuanced take on her experience.

While praising the personal attention she received from a counselor,  Ali described a “rowdy” atmosphere in the sixth-period history classroom where she took the survey, with classmates buried in their phones and chatting with friends. It made it difficult to express some of the conflicts she was experiencing at the time. 

“It was a bunch of juniors just goofing off. I was sitting next to my friend, and she would just ask me, ‘Oh, what did you answer?’” she said. The atmosphere, she added, “felt like it wasn’t as serious as it should have been.”

Do kids dare each other to lie? Look over a classmate's shoulder? 

For some King County districts, however, Check Yourself simply proved to be too much.

The Lake Washington district pulled out of the program three years ago and instead contracts with full-time mental health specialists to respond to students’ needs.

The intensely personal questions — and the resulting risk of privacy violations — also helped push the Bellevue school system to drop it in 2019. 

Bellevue SD is using something else and I'm going to read up about it. 

Officials opted for a different survey, and because of their sensitive nature, results are “considered some of the most privileged data the district has,” said Naomi Calvo, who served as Bellevue’s director of research, evaluation and assessment until 2023. “I didn’t even have access to it.”

Calvo understands why districts jumped to implement Check Yourself and most continue to use it. “Students have needs that were going unaddressed and there is a dearth of options available,” she said. 

But as a mental health professional with a young son at the time, she felt skeptical. 

“As a researcher, I believe in surveys,” she said. “But I would not have let my child take that survey.”