Free Ed Tech Resources eBook

  • Over 100 Pages of Free Instructional Technology Resources
  • Tools for collaboration, gamification, active learning, screencasting, tablets, smartphones, and much more!
  • YOURS FREE just for signing up for blog posts!

Sign Up Now


Gathering Evidence that Flipping the Classroom can Enhance Learning Outcomes

by Kelly Walsh on March 10, 2013


Three universities provide empirical evidence supporting the potential for ‘the flip’ to make a measurable difference in engagement and learning.

As an advocate of the potential of the flipped classroom, it’s rewarding and encouraging when student and teacher feedback supports the benefits of this approach, and this happens quite often. However, a wealth of measurable evidence that the technique can improve learning outcomes would go a long way towards convincing educators everywhere that this is an important technique to consider leveraging further in our schools.

Not long ago I stumbled across an article about San Jose State University that discusses measurable improvements in test scores in a course in which some students used a flipped model. This weekend I went in search of more such examples, and share these findings here. (If you know of any other examples like these, please comment and tell us about them.)

“San Jose State U. Says Replacing Live Lectures With Videos Increased Test Scores”

The Chronicle of Higher Education, October 17, 2012

“In an effort to raise student performance in a difficult course, San Jose State University has turned to a “flipped classroom” format, requiring students to watch lecture videos produced by the Massachusetts Institute of Technology and using class time for discussion. And initial data show the method is leading to higher test scores, university officials announced this week.”

Khosrow Ghadiri, an adjunct professor, explains that concerns about the course “Engineering Electronics and Circuits,” and its historically low passing rate (40% of students in the class received a C or lower last semester), led San Jose State professors to MIT. There they worked with the edX team, a partnership of MIT, Harvard, Berkeley, and the University of Texas at Austin focused primarily on developing MOOCs. They developed an approach wherein the 85 students in the flipped course watched edX lecture videos (from professors at leading universities) at home and attended class twice a week to practice what they had learned and ask questions while two other sections of students took a traditional version of the course.

The result: Midterm exam scores for students in the flipped section were higher than those in the traditional sections. Even though the midterm questions were more difficult for the flipped students, their median score was still 10 to 11 points higher.

“Flipping the Classroom”

Vanderbilt University ‘Center for Teaching’ Web Site

Carl Wieman and colleagues at Vanderbilt have published evidence that flipping the classroom can produce significant learning gains (Deslauriers et al., 2011). Two sections of a large-enrollment physics class were compared. For the first 11 weeks of the semester, the classes were both taught via interactive lecture methods for the majority of the semester and showed no significant differences prior to the experiment.

The flip came during the twelfth week of the semester, with one section being exposed to new material prior to class via reading assignments and quizzes, so that class time could be devoted to small group discussion and questions delivered through clickers and written responses. “The control section was encouraged to read the same assignments prior to class and answered most of the same clicker questions for summative assessment but were not intentionally engaged in active learning exercises during class.” These impressive results were evidenced:

  • At the end of the experimental week, students completed a multiple choice test, resulting in an average score of 41 +/- 1% in the control classroom and 74 +/- 1% in the “flipped” classroom, with an effect size of 2.5 standard deviations.
  • During the experiment, student engagement increased in the experimental section (from 45 +/- 5% to 85 +/- 5% as assessed by four trained observers) but did not change in the control section.

While this exercise was conducted with a very limited time frame, the dramatic increase in engagement and learning speaks well to the use of the flipped classroom model.


Cara A. Marlowe, MONTANA STATE UNIVERSITY, June 2012

This thesis documents research conducted by a student in pursuit of a M.S. in Science Education at Montana State. The document abstract explains, “In this investigation, the effect of the flipped classroom and associated differentiation was studied to measure the impact on student achievement and student stress levels. For the second semester of their senior year, students watched video lectures outside of class and completed assignments during class time.”

The effect of the flipped classroom on student achievement and stress levels was tested on 19 students in an Environmental Systems and Societies course. Students were taught using traditional lecture methods for an extended period of time and then using the flipped classroom approach for a period of time. In the flipped classroom method, lecture videos were recorded and published to YouTube. Students were responsible for watching videos and submitting questions they had about concepts after watching the videos (or a summary if they understood the lecture and had no questions). These questions and summaries were used to stimulate classroom discussions. Remaining classroom time was devoted to working on projects, lab activities, readings, research and other assignments that might otherwise have been assigned for homework. Grades, including formative and summative assessments were compared between semesters and in some instances, across the student’s high school careers. Students were interviewed and asked to reflect on their learning and stress levels throughout the process.

Jumping to the published conclusions, “Students reported lower stress levels in this type of classroom environment compared to other classes. Results of the study indicated that use of differentiation and independent study, through the implementation of the flipped classroom model, was successful. The grades from semester one and semester two were significantly different, with the majority of students seeing an average increase of three points in semester grades.”

I have little doubt that these are only the first of many such studies and reports proving out the benefits of the flip when it’s delivered well and used judiciously. If any readers know of other examples like this, please do comment and tell us about them (and if you know where we might find information online about these example, that would be great too!). Thanks.

Related Posts (if the above topic is of interest, you might want to check these out):
8 Great Reasons to Flip Your Classroom (and 4 of the Wrong Reasons), from Bergmann and Sams
Collecting studies focused on the impact of Education Technologies
Flipped Classroom Successes in Higher Education



Kelly Walsh is Chief Information Officer, and an adjunct faculty member, at The College of Westchester in White Plains, NY and is the founder and author of As an education technology advocate, he frequently delivers presentations on a variety of related topics at schools and conferences across the U.S. Walsh is also an author, and online educator, periodically running Flipped Class Workshops online. His latest eBook, the Flipped Class Workshop in a Book was published in September, 2013 and is available here. In his spare time Walsh also writes, records, and performs original (and cover) songs (look for "K. Walsh" on iTunes or or check out his original song videos on here on YouTube ).

Print This Post Print This Post

{ 2 comments… read them below or add one }

Kelly Walsh June 5, 2014 at 5:50 pm

Thanks for the in depth and well stated feedback Liza. These studies were relatively limited in scope and, yes, there was certainly some room for improvement in the methodologies. Fortunately, the body and validity of studies about the use of flipped instruction techniques, with positive results, continues to grow. Hopefully some of these will help to deliver the more conclusive and less refutable evidence of effective techniques that you speak of.

Liza Loop June 5, 2014 at 1:11 pm

I have one comment and three reservations about the research mentioned in this post. The comment is to note that the flipped classroom approach was already being explored in 1963 when I was in college at Cornell University. Our chemistry professors, Sienko and Plane, had created a ‘programmed workbook’ version of their text book. The students who used the workbook but did not attend lectures performed just as well as those of us who dutifully arose on Saturday morning to sit through the professor reading his notes. At neighboring Ithaca College, all lectures were recorded and available via closed circuit TV to every student. Many chose to watch the lectures from their dorm rooms. This leads into my concerns about research design.

Problem 1: confounding of variables. In the case of Sienko and Plane, we can be reasonably assured that the treatment and control groups were receiving the same academic material. However, many flipped classroom studies provide different stimuli (local classroom teacher live presentation as compared to nationally-recognized, recorded lecture) to each group. These studies do not allow us to distinguish between the effect of high quality presentation and the medium of presentation. We need to compare apples to apples.

Problem 2: Few studies track student time on task. It is pretty well accepted that more time spent studying results in higher academic performance. Do students using prerecorded lectures at home spend more time with them than students attending the same lectures live at school? You can’t replay the live professor if you happen to miss a point and many students would not dream of interrupting to ask a question even if the instructor requested that they do so. I was amazed to find that my own preschoolers in the early 1970s rewound the BetaMax tapes of Sesame Street repeatedly until they had mastered the lesson. Is this going on with older flipped classroom students? We need to control for time-on-task if we are to understand the outcomes of flipped classrooms as measured by improved test scores.

Problem 3: Not all students are alike. Some may thrive in the complex social settings of face-to-face classrooms. Others may find classrooms distracting when compared to quietly watching a video at home. Still others, those with more chaotic homes, may find the classroom a calm oasis. When study sample sizes are small the effects of individual differences and social settings may be much larger than the effects of flipping the classroom. This could account for mixed results and relatively small effect sizes in research to date.

I have spent my career advocating for educational technology and believe, at an intuitive level, that it can enhance learning when properly used. My disappointment is that poor research designs are interfering with our ability to demonstrate that edtech really works and to guide us in refining the design of even more effective learning aids.

Leave a Comment

{ 39 trackbacks }

Previous post:

Next post: