I've made online teaching and learning a regular theme of this blog (see the long list of links at the end of this post). I'm still of the opinion that online teaching creates a learning deficit among less-motivated or low-ability students, although on average the outcomes are similar for different teaching and learning modes. I desperately want someone to convince me that there is a solution that will at least lead to similar outcomes as we observe in face-to-face teaching, across the entire ability distribution. I'm yet to find it.
Most of the literature I have read relates to teaching and learning in the university context. That makes sense - until recently, online teaching and learning has been much more prevalent in tertiary education. However, the pandemic forced online teaching at all levels of schooling, providing a broad natural experiment on the impacts of online learning. So, I was interested to read this new working paper by Rebecca Jack (University of Nebraska-Lincoln), Clare Halloran (Brown University), James Okun (MIT), and Emily Oster (Brown University), which focuses on online teaching of US students in Grades 3 to 8. Specifically, Jack et al. combine exhaustively-collected data on school learning mode by school district across 11 states, with data from standardised tests conducted in 2016 to 2019, and in 2021 (with no testing being conducted in 2020 due to the pandemic). They distinguish three types of schooling mode:
...1) "in-person" (all or most students had access to traditional, 5-day-per-week, in-person instruction); 2) "virtual" (all or most students received instruction online, five days a week); and 3) "hybrid" (schooling modes that did not fall into one of these approaches).
Jack et al. then measure the percentage of school days in each schooling mode for each school district, and test the relationship between schooling mode (as a continuous variable) and pass rates (at the school district level) in English language arts (ELA) and maths. They find that:
The coefficients are quite stable across specifications and highly significant in all of them. In terms of magnitude, these regressions suggest that moving a district from fully virtual to 100% access to in-person learning would have reduced pass rate losses in Spring 2021 by 13 to 14 percentage points in math and about 8 percentage points in ELA. Moving from fully virtual to fully hybrid would have reduced pass rate losses by about 7 percentage points in math and 5 to 6 percentage points in ELA. Focusing on within-state, within-commuting zone variation in schooling mode, we estimate districts with full in-person learning had an average decline of 13.4 percentage points less in math and 8.3 percentage points less in ELA.
Those declines are quite substantial, and Jack et al. also find (what I characterise as) suggestive evidence that the effects are larger in school districts with a greater share of African American students, although they find no difference in impact between school districts with a greater or lesser share of low-income students (as measured by eligibility for free or reduced price school lunches). Also of interest, when looking at each grade individually, they find that:
In general, we see higher impacts in younger grades. This is especially true for in-person learning. This may reflect a greater benefit of consistent in-person time for younger students, although based on these data alone it is difficult to fully elucidate mechanisms.
One possible mechanism is that older students may be more self-directed in their learning than younger students are, and therefore better able to negotiate the online learning environment. That would accord with the limited difference in the literature in learning outcomes (on average) between online and face-to-face learning at tertiary level. However, it would be interesting to know whether there was heterogeneity in the effect of online learning between students at different levels in the ability distribution at this level (as observed in tertiary education). However, without longitudinal data on individual students (rather than aggregate data at the school district level), it would be difficult to investigate that heterogeneity.
Taking this study alongside others (particularly at the tertiary level), it seems to me that we need to understand more about how academic self-efficacy (or how self-directed students are) interacts with online teaching and learning. It would also be good to understand more about how academic self-efficacy can be promoted in the online environment, particularly among students in the bottom half of the ability distribution.
[HT: Marginal Revolution]
Read more:
- Online vs. blended vs. traditional classes
- Flipped classrooms work well for top students in economics
- Flipped classrooms are still best only for the top students
- Online classes lower student grades and completion
- Online classes may also make better students worse off
- Meta-analytic results support some positive effects of using videos on student learning
- New review evidence on recorded lectures in mathematics and student achievement
- Meta-analytic results may provide some support for flipping the classroom
- The value of an in-person university education
- The pandemic may have revealed all we need to know about online learning
- Live streamed video lectures and student achievement
- Low-performing students, online teaching, and self-selection
- A slightly more optimistic take on online learning during the pandemic
- Research-based instructional strategies and flipped classrooms in economics
- Using microcommitments and social accountability to improve online and hybrid learning
No comments:
Post a Comment