Enclave Data FAQ
Last Updated: February 16, 2016 at 5:57 pm EST
What is Crisis Text Line?
A not-for-profit organization that provides 24/7 support to people in crisis by text. Top issues include depression, anxiety, suicidal ideation, family issues, and romantic relationships. We also cover issues like substance abuse, sexual health, sexual abuse, and eating disorders.
Why is Crisis Text Line opening up for research?
We believe that the data itself can save lives.
We’re hoping that more research will lead to more funding for effective solutions, better police and school board policies on these issues, smarter journalism on mental health, substance abuse, violence, etc.
From day one, system change was one of our goals. Nancy explained Crisis Text Line as: (1) to help people one-on-one move from a hot moment to a cool calm and (2) leverage the data to improve care for people in crisis everywhere.
Why does data matter?
Currently, the largest open dataset on mental health and crisis is a survey run every other year by the Center for Disease Control (CDC). Researchers just don’t have a lot to work with, so research on mental health and crisis is slow and expensive. (Meanwhile, people are suffering!)
Over 13 million text messages have been exchanged via Crisis Text Line since its launch in August 2013, qualifying this as the nation’s largest open set of crisis data. This volume of messages, velocity in 2.5 years, and variety of content is incredibly valuable to researchers. With it, we’re hoping to inspire smarter school board policies, legislation, and journalism.
Will my personal information be shared? Will researchers be able to figure out who I am?
No. No. NOOOO!
All data is auto-scrubbed for personally identifiable information, including names, addresses, dates of birth, phone numbers, and email addresses, prior to sharing with researchers.
While no auto-scrubbing tool can reach 100% accuracy, we are using practices to reach that goal. For example, if we scrub the name Tiffany, we replace it with Sam. Thus, all data becomes effectively anonymous, in that it is not possible to see what personally identifiable information has been replaced or not.
As a texter, can I opt out of this completely? Can I request my data be deleted?
What will researchers do with all this data?
Researchers will apply for specific data pulls. They will NOT have access to the entire thing. Nope. No way.
Approved research projects will have access to a limited subset of data for analysis within our secure environment; for example, a research project might request the dates and timestamps of messages that reference “dark cloud,” “just so sad all the time,” “can’t get out of bed” and match this against weather patterns, which would allow them to investigate the relationship between weather and depression!
How are you protecting the privacy of texters?
We love our texters. We are intensely serious about protecting the health and safety of individuals. To quote the original Tron movie, we “fight for the user.” (Fun fact: one of our developers was even in the Snowden movie.)
We have spent nearly a year (1) looking at best practices from the NIH, CDC, and the University of Michigan, (2) establishing an Ethics Committee, (3) preparing a stringent application procedure based on best practices, (4) creating a custom research review process comparable to an IRB, and (5) undergoing multiple security and privacy reviews by 3rd parties.
Simply, we have been quite cautious and collaborated with the best possible people.
Who are these “researchers”?
Only university or research institution-affiliated researchers (who must be faculty or post-doctoral researchers and have IRB-approval) may apply for access to Enclave Data, and only for non-commercial use. All applications will be scrutinized for four criteria: privacy, security, ethical considerations, and value to people in crisis.
Who is blocked from access to the data?
Like, almost everyone on the planet! Only a handful of researchers are going to be approved.
All researchers must pass through a stringent application. In addition, we will NEVER share data…
- For commercial use.
- With individuals not associated with a university or research institution.
- “Just because.” Yep. We reserve the right to reject any application, for any reason.
Are you selling this?
Nope. Heck no. Not gonna happen. Yuck. Gross.
(Read: no commercial use. Never ever ever.)
Are you giving “back door” access to the government?
Government researchers, such as those at the National Institutes of Health, may apply, but must go through the same process as everyone else. All applications must meet stringent requirements for the same four criteria: privacy, security, ethical considerations, and value to people in crisis.
How will researchers get approved for access to the data?
Researchers must pass three tall hurdles:
- Submit an Enclave Data Application to Crisis Text Line
- Receive approval from a university or independent IRB
- Receive approval from the CTL Ethics Committee, which meets quarterly behind closed doors.
If approved, what will researchers have access to?
Crisis Text Line is opening three levels of data to researchers – conversation, actor and message level – with each level increasing in detail and barriers to access.
- Conversation level, will allow researchers to explore questions such as, “What crisis issues occur most on holidays, such as Christmas and Valentine’s Day?” and “How are bullying and depression related?”
- Actor level, will allow researchers to answer questions like “For a texter experiencing depression, how do issues fluctuate over time?”
- Message level, will allow researchers to dive deep into interactions, and uncover patterns such as how texters struggling with self-harm describe their experience.
I’m a Crisis Counselor, will my personal data be shared with researchers?
Researchers will not be able to identify individual Crisis Counselors. Our Crisis Counselors are not part of this research. We will not be sharing our Crisis Counselor list with anyone. Enclave Data is focused on learning about mental health in general–not any particular person.
Could a researcher keep all the data for other projects?
To get a little wonky and technical, we will not actually share data with researchers. We allow researchers to access a limited, pre-approved slice of data, for a limited time. We can turn off access, just like turning off a light. Researchers can’t export, copy, or otherwise remove data from the system.
What’s to stop people from running a bot to read and save all the data?
Crisis Text Line restricts access. First, we only approve applications that meet our stringent criteria. Second, we protect our data using best practices in security and privacy. Third, we allow researchers to access to a pre-approved slice of data, for a limited time. Fourth, we constantly monitor network traffic and server activity to ensure there are no anomalous behaviors that indicate circumvention of our security policies. Fifth, to get a bit wonky, the environment is strictly controlled and no outside users have “sudo” or superuser access allowing installation of custom programs.
Will this data be available on data.gov and other data warehousing sites?
Hell no. We do not allow Enclave Data to be hosted on data.gov or any other data warehousing site. Enclave Data is only hosted on Crisis Text Line’s private, secure servers.
How long does Crisis Text Line store all those messages?
How will open data benefit Crisis Text Line’s texters?
All research must aim to benefit people in crisis. We’ll create a better experience for our texters in three ways: (1) we’ll make all research papers available on our website (2) we’ll incorporate findings into our training for Crisis Counselors (3) this research should lead to more funding for good things, better journalism, less stigma, and other stuff that should help people. That’s kinda the whole point.
Who is Crisis Text Line funded by?
Crisis Text Line’s open data work is funded by the Robert Wood Johnson Foundation.