| dc.description.abstract | As online learning platforms become increasingly integrated in educa
tion, there is a growing need for scalable methods to evaluate their effective
ness in fostering learning. Traditional evaluation methods, such as pre-post
tests in classroom settings, are time consuming and difficult to scale. This re
search explores the use of platform-embedded experiments as a cost-effective,
continuous method for evaluating learning platform effectiveness, using Squla
and StudyGo as case studies. In Squla, an experiment was conducted to as
sess the impact of practicing relevant topics through the platform on quiz
performance. The results showed that for math and language, answers on
a new quiz were 3% more likely to be correct after practicing the relevant
topic on the platform, compared to practicing off-topic content. However,
no significant improvement was observed for spelling and grammar or read
ing comprehension. The StudyGo experiment focused on learning within a
single attempt at a set of practice questions. The order of questions was
manipulated and the results showed that questions were 3% more likely to
be answered correctly when placed at the end of a set compared to the be
ginning, suggesting that students learned from previous questions. These
results demonstrate the potential of platform-embedded methods for scal
able and efficient measurement of learning outcomes. Future research should
address the limitations of these methods, such as their limited generalizabil
ity, and explore their applicability across a broader range of content and
different learning platforms. | |
