https://www.tpr.org/post/how-can-we-change-minds-about-mental-health
Mental health is something that has become more of a foci within Student Life and on college campuses nationwide. Student well-being has comprised a lot of time and planning with higher education administrators to develop evidence-based responses as well as proactive, prevention programming. These discussions and planning activities have centered on the importance of ensuring our students are getting the appropriate support they need to accomplish both their interpersonal and academic goals while enrolled in higher education, as well as best supporting their growth.
So is your institution gauging student’s well-being through assessment?
Survey says...
Higher Education Today, a blog by the American Council on Education, published, “The mental and physical well-being of incoming freshmen: Three decades of research.” The post highlighted results from the CIRP Freshmen Survey, which provides data on incoming freshmen regarding demographics and overall expectations, behaviors, and attitudes they have prior to enrollment, specifically a subset of questions focusing on mental health and well-being were evaluated with 20 years of data.
The longitudinal trends indicated that students today are reporting poorer physical health than in previous decades. The author, Ellen Bara Stolzenberg, makes a note to point out that women were less likely to rate their physical health as above average compared to their male counterparts. It was especially striking to see that students who identify as Latino/a dropped 15.4% points in self-reported physical health since the CIRP first asked the question in 1985.
Further noted, White and Black students were more likely to indicate their physical health was above average, while Asian students were more likely to rate their physical health as ‘at least’ above average compared to their peers.
“College student mental health and well-being”, a 2019 blog by the American Council on Education focused on mental health of young adults, is also of interest. Data has shown 3 in 10 students have experienced depression in the last two weeks, while 1 in 4 are experiencing anxiety.
With a host of articles to reflect, guide, and provide practices for campus to engage in more proactive discussion of mental health support, the goal of this blog is to reiterate mental health with college students needs to be a foci and to leverage resources to help us formulate responses on our respective campuses.
Because there is a lot of evidence that mental health is a concern for college students, what are we going to do about it? Thinking sequentially about the student experience, how are we gauging the well-being of students when they arrive on our campuses?
https://towardsdatascience.com/machine-learning-and-mental-health-7981a6001bd5
Head Start
With the information referenced above, it is evident that mental health is a concern for college students today. It is critical that we not only continue to understand and measure student well-being, but that we come up with an action plan on how best to support students as we welcome them to our campus. Within Student Affairs, our profession has the foci of supporting the whole student. This can range from engagement to intervention initiatives. Equipped with knowledge of a students’ well-being, data-informed practice can enhance how we best support our students’ needs.
The timely response on utilizing pre-admission or early experience data from students can be a critical component to institutional success in taking action for students’ well-being. Consider that information in light of this article on best using data to positively impact our students’ transition. A data scientist, Nabil Abbas wrote an articulated response that may guide higher education on how we can more appropriately respond in a timely fashion, “Machine learning and mental health: Can big data help the mental health field?”
The crux of Abbas’ work is focused on how we can rely on computers and algorithms to support our higher education colleagues in developing predictions that can aid us in our timely response regarding mental health. This is built on the concept that every student should have a responsive and tailored intervention based on their unique needs and well-being profiles based on measurement of well-being constructs. The advancement of technology allows such machine learning to utilize students’ self-reported data along with other student records and passive information (e.g., engagement with campus activities) to see if there are patterns of behavior with the ability to understand any triggers that might predict certain behaviors. It is essentially using person-level information to predict future behaviors as well as customizing support plans that best meet their individual needs.
Preparation Check
Are you currently measuring students’ well-being as they enroll on your campuses? If so, how are you using this information?
Machine learning as described above would not replace current practice per se, but would automate and offset some of the challenges higher education professionals have in terms of timing and the type of proactive support for students with mental health concerns. With higher education enrollment becoming more diversified according to the CIRP data, and we know that students from diverse backgrounds have a difficult time accessing mental health services, this may be a way to fill a gap in how we appropriately respond to students’ well-being needs in a timely fashion. As Student Affairs professionals, we encourage student engagement in programming and experiences so they can connect with a community. This method would be a facilitator of these types of connections and interaction.
Given capacity in our field, most institutions may not be able to support machine learning or may not currently have one in place. Knowing that, what systems of data collection and measurements do you have in place that are capturing student well-being data? Do you use national benchmarking surveys to help support this need? Systems of data don’t often communicate with one another on our campuses. It is often the responsibility of the assessment professional to lead these conversations and ensure data is being collected and utilized in the most ethical ways.
https://www.sfn.org/sitecore/content/Home/BrainFacts2/Diseases-and-Disorders/Mental-Health?page=4
Do you see a gap in access to mental health support between your traditionally underrepresented students and their peers?
Gauging Students’ Well-being in Action
On our campus, we have been measuring students’ mental health and well-being through our Student Well-being and Success Initiative (SWaSI), which is a tailored, multi-cohort, longitudinal assessment measuring various aspects of belonging and several psychosocial factors that may contribute to student success, specifically well-being. It has the capacity to assess various institutional inputs from advisement, high-impact practices and other programmatic experiences, and prevention/intervention efforts in terms of well-being and success outcomes. This stemmed out of our Assistant Director’s dissertation, Dr. Brian Clark, as a way to best understand our students’ needs early on with the goal for timely intervention and connection to resources as students enroll on our campus. Our long-term goal is to best connect this information using trends to understand how best to support and respond appropriately to our students during their first few weeks on campus.
Capacity in our office is always an issue, but machine learning has been a concept that would be beneficial in helping us move the needle in terms of timely response for our students regarding mental health. Using machine learning can offset the capacity of our office in terms of timely and advanced data analytics by allowing employees to focus on other pressing issues such as helping colleagues understand how to use the information and create action plans.
We have a lot to do to get to implementing a real machine learning model that will supplement and support our current services. We can be working in parallel to continue our existing data-oriented and reactive interventions, while also continuing to model and pilot predictive efforts via learning models. Wherever you may be on the spectrum of progress in this arena, there is likely room for you to grow, too. It is worth the effort, though, as students are the most important asset on our campus. We need to maintain our ability to be data-informed, while focusing on efficiency and effectiveness of our actions to ensure our students well-being is supported.
Has your campus been discussing machine learning in the context of student success or student well-being? How is your campus discussing how to appropriately respond to mental health concerns of your students? How are you connecting early on with your students to encourage them to find a community?
Renee Delgado-Riley, University of Oregon