Still think online learning is inherently worse than traditional teaching? That may have once been true, but the digital classroom has come a long way. Last May, Ithaka Strategic Consulting and Research released a study titled “Interactive Learning Online at Public Universities: Evidence from Randomized Trials.” Much to traditionalists’ surprise, it seems that hybrid learning environments can produce identical outcomes to regular classes.
Though there have already been hundreds of studies on the efficacy of online learning, Ithaka’s experiment is the first of its kind. 605 statistics students were studied at six campus from three universities – State University of New York, City University of New York, and the University of Maryland. More importantly, these students were randomly placed into “control” (traditional) and “treatment” (hybrid) classes at the same school after they agreed to participate. Their identical course content, competency tests, and final exams allowed for a direct comparison between teaching methods.
Ithaka conducted this study to answer three questions, the most important of which was “Can sophisticated, online courses be used to maintain or improve basic learning outcomes in introductory courses?” Tuition costs are rising at an unsustainable rate, and the researchers believe online learning may finally allow schools to lower their per-student costs without compromising quality. Here’s what they found.
Nearly Identical Results
Students in the control group attended traditional lectures and TA-led discussion sections three to four times per week. Those in the experimental group relied mainly on an online statistics course originally developed through Carnegie Melon’s Open University program. They supplemented these machine-based “lectures” with once-per-week, hour long question sessions with their professors.
The outcome? The online students actually outperformed their peers on the final exam, 56.7% to 54.7%. Results were similar for the CAOS standardized stats test: a 47.5% average for the hybrid classes, 46.9% for the in-person lectures. When all was said and done, 79.7% of online students and 76.4% of the control group passed the class.
These differences aren’t significant, but that fact certainly is. Ithaka notes that teachers’ and deans’ most common reason for resisting online education is that they “worry that basic student learning outcomes (pass rates and mastery of content) will be hurt, and they won’t expose their students to this risk.” If these results are any indication, those fears are unfounded.
In addition to neck-and-neck class averages, online and in-person students exhibited almost identical grade distributions. Even when researchers controlled for race, gender, socioeconomic status, parental education, and prior GPA, “there were no groups of students that benefited from or were harmed by the hybrid format.” This is a crucial consideration for cost-cutting schools, since students at the lower end of the bell curve are the most likely to leave. As it turns out, that’s no more a problem for digital classes than it is for normal ones.
Differences in Preparedness
Can impersonal, online instruction be just as helpful when students are completely unfamiliar with the material? That was another of Ithaka’s main concerns, and the answer turned out to be yes. Participants also took the CAOS standardized test at the beginning of the term, and low scorers performed just as well as high scorers by the end. Controlling for pre-term scores, there was no significant difference in the distributions for the final exam or the post-term test. While these results may not hold true for advanced material, they do indicate that intro classes – which typically have the highest enrollments and largest waitlists – can be efficiently and effectively taught online.
Despite their scores, students who took the hybrid stats class just didn’t like it as much. On a scale of 0 to 2.5, the traditional format got an average rating of 2.3, while the online version got a 2.0. Online students also claim to have learned slightly less, and to have struggled slightly more with the material. However, participants in both classes reported the same level of raised interest in statistics – 1.7 out of 2.5.
An admin at one of the participant schools believes the lower ratings are due to online learning’s lack of “addictive” or “Disney-like” appeal. Students are bound to miss out on instructors’ anecdotes, jokes, and other endearing “extras” that can add enjoyment to an otherwise-dull class. Still, the test scores don’t lie. In a time when education is becoming increasingly costly, results are what matter most to struggling students and cash-strapped universities.