When I was a middle school science teacher in the inner city of Los Angeles, I often wondered whether my students “made it” in high school, in college and in a job. We had our annual articulation day when we met with our feeder high school, but sadly, it was just a lunch, with our faculty sitting on one side of the cafeteria and the high school teachers on the other. That’s it. Articulation accomplished. Speeches were made, but not a word was spoken about my students’ success or their particular and personal impediments or accelerators to learning. Where the student ended up was a mystery, but we all believed that the good students landed in good places. There was nothing to back up that belief but anecdotes, usually delivered by students who tracked me down many years later to thank me for believing in them.
Where the student ended up was a mystery, but we all believed that the good students landed in good places. There was nothing to back up that belief but anecdotes, usually delivered by students who tracked me down many years later to thank me for believing in them.
Well, I am happy to write that things have changed since then! In our work at ERP, we now use data science to follow a student’s pathway across all segments, from early childhood education to elementary school to middle school and through high school, and finally into college and the workforce. Now, there is no guesswork on what happened to the student, no mythology, wishful thinking or anecdotal success stories. Even better, we can actually reverse engineer success from those who are landing living-wage jobs with healthy predictable wage gain over time, and see what happened in college (or no college), in high school (or no high school), middle school, elementary school and in early childhood education environments that worked for them. We are now in a historical position to use the analysis of those successful pathways to predict and accelerate success for other students.
In our work at ERP, we now use data science to follow a student’s pathway across all segments, from early childhood education to elementary school to middle school and through high school, and finally into college and the workforce.
When we first began to link the data sets among different educational segments, we noticed the majority of the students disappeared from one segment to the next. Where did they go? Why did so many disappear? More important, why did many students who succeeded in elementary school mathematics never take algebra in middle school? Why did those who succeeded in middle school math never take Algebra 2 in high school? Why did 7 out of 10 ethnic minority, socio-economically disadvantaged students who succeeded in high school drop out of community college? Is this a problem with articulation between the segments or is it a massive teaching problem?
We learned very quickly from these analyses that many students have the capacity to succeed yet were sorted out of the systems using proxy measures for success, usually some sort of placement or readiness test or jury of experts. Another way of saying this is that students who have done very well and were prepared to succeed after seven years of elementary school math may not get algebra in the 8th grade, based purely on inexact judgement or measurement. Actually, this happens to most ethnic minority and socio-economically disadvantaged students. Likewise, students in school for 12 years of elementary and high school may do poorly on a single placement test for college. This proxy measure sorts out many students who would succeed given the chance.
We now know many students have the capacity to succeed, but the status quo policies often focus on additional learner preparation and re-teaching – even when they will do just fine if left alone and given an opportunity to succeed.
We now know many students have the capacity to succeed, but the status quo policies often focus on additional learner preparation and re-teaching – even when they will do just fine if left alone and given an opportunity to succeed.
The California Community Colleges Chancellor’s office made a huge leap forward when it officially recommended a multiple measures approach to placement in for-credit college coursework. In other words, use data analytics to understand the students’ capacity to succeed as opposed to their score on a single test. The thinking is that if they succeeded in the past, then they should succeed in the future, given the right supports. This has major implications for transitions from segment to segment in every aspect of the pre-K to Job pipeline.
The California State University System has also taken a huge step forward in moving from remedial education to placement in for-credit courses by adding extra support. Our work at ERP is to better understand the efficacy of that additional support. Maybe that extra level of support is not even needed.
All of this brings us back to the barriers to success. Are they preparation barriers (more teaching) or are they institutional barriers (our best guess on what students know) and what’s the difference between the two?
It’s time to bring data science forward. Thinking back to my middle school/high school articulation: What a fine lunch that would be to bring multiple measures of success for each student forward as a handoff to the high school teachers – and then from the high school to the community college faculty, and from those folks to the university professors, and everybody bringing it forward to the employers.
It’s time to rethink our educational pathways and our huge discriminatory sorting machine.
James Lanich, Ph.D., is the founder, president and CEO of Educational Results Partnership (ERP), a nonprofit organization committed to improving equity and educational outcomes through the application of data analytics. Our work focuses on identifying successful systems, practices, programs, and pathways in public education that improve college and career readiness. Additionally, we foster collaboration across academia and business to align educational curriculum with workforce needs.
コメント