Flipped Classroom Field Guide
The Flipped Classroom Field Guide is a compilation of best-practices and community resources centered around the flipped classroom and blended learning initiatives of the Coursera-partner community. In addition to the resources outlined in the table of contents (including our Three Golden Rules of Flipping which have been shown to significantly improve student outcomes), be sure to click on the reports below to learn more about the instructors and their flipped classrooms that motivated and informed the creation of this guide:
Class: Child health and nutrition
Size: 85 students
Size: 150 students
Class: Machine Learning, Databases
Size: 10 students, 30 students
Name: Scott Klemmer
Class: Human-Computer Interaction
Size: 250 students
Class: Organizational Theory
Size: 44 students
Class: Genetics and Evolution
Size: 450 students
Class: Introduction to Python
Size: 70 students
Class: Writing in the Sciences
Size: 12 students
Class: Introductory Biology
Size: 75 students
Class: Global History
Size: 76 students
Class: Intro to Electronic Music
Size: 15 students
Class: Democratic Development
University: University of Wisconsin
Class: Analytical Methods for Electromagnetic Engineering
Size: 80+ students
Class: Introduction to Programming
Size: 500 students
Jeff Himpele, Princeton
University: University of Wisconsin
Class: Introduction to Programming
Size: 40 students
Active learning activities
The flipped classroom is a teaching strategy that allows instructors to more actively engage with students in the classroom. In the flipped classroom, instructors typically assign recorded video lectures as homework, and use class time for active learning exercises and direct engagement with students. As Mung Chiang from Princeton said, “class time is for two-way interactions.” Flipped classrooms help make these two-way interactions possible.
Derek Bruff from Vanderbilt contributed his own flipped overview to the Coursera Partners’ Portal forums along with the following diagrams highlighting the differences between the traditional and flipped classroom model:
Flipped classrooms are a form of of blended learning, a term that refers to any form of education that combines face-to-face instruction with computer-mediated activities.
In most flipped classrooms, students watch online lectures outside class and participate in engaged learning activities inside class. This approach allows instructors to more deeply engage their students with evidence-based learning practices that can significantly improve student outcomes 1. Despite the many ways to implement this model, we’ve found that the most successful flipped classrooms have found creative ways to increase time on task and student engagement in the following ways:
Results and examples
In addition to quizzing, other effective active learning activities include applications, extensions, asking a sequence of questions, student-generated content, experiential learning, discussion activities, giving a packet of problems, Classroom Assessment Techniques (CATs) and collaborative learning activities, all of which are described in detail in our Active Learning Activities section.
1: Deslauriers, L., Schelew, E., & Wieman, C. (2011) Improved Learning in a Large-Enrollment Physics Class, Science, Vol. 332 no. 6031 pp. 862-864
In flipped classrooms, instructors are able to more deeply engage with students in class by assigning lectures and other passive learning activities as homework and using in-class time for more active learning activities. This strategy yields a number of benefits to both students and instructors, including:
Improved educational outcomes: Studies have shown that flipped classrooms and blended learning environments can significantly improve educational outcomes when compared to traditional classrooms. The efficacy of blended learning environments and active learning strategies is well-documented in the literature: a 2010 meta-analysis from the Department of Education based on 45 studies showed that online learning is as effective as face-to-face learning, and that blended learning is considerably more effective than either. Additionally, Deslauriers, Schelew and Wieman (2011) compared two large sections of an introductory undergraduate physics course: one section was taught as a traditional lecture by an experienced, highly-rated instructor; the other was taught by an inexperienced instructor using active learning strategies. In the active-learning group, student engagement nearly doubled, attendance increased by 20%, and average scores on assessments increased from 41% to 74%. Similar results by Wieman (2011), Mazur (2009), and others, were obtained across multiple disciplines and diverse institutions. After Jeremy Adelman of Princeton flipped his class “A History of the World since 1300,” he reported that he had “...never seen such good final papers and take-home exams in 20 years of Princeton teaching.”
Efficiency: Lecturing and assignment grading are time-intensive activities. The long-run time savings from automating some of the repetitive lecture and grading activities allow faculty members to spend more time on active learning activities, teaching problem solving, and giving students personal attention and individual help. Similarly, autograded assignments allow instructors to quickly and efficiently evaluate each student’s understanding of learned concepts.
Interactive lectures: On the Coursera platform, instructors are able to intersperse in-video quizzes throughout their lectures, making the student lecture experience more interactive, dynamic, and personalized than traditional lectures. Instructors are also able to implement elements that would be difficult to incorporate into traditional lectures, such as animations, simulations, interviews with distinguished individuals in the field, etc. These features have value not only in maintaining student focus and engagement, but also in improving student performance: research shows that even simple retrieval exercises (e.g. in-video quizzes) are more effective than traditional study strategies and can significantly improve student outcomes (Karpicke & Roediger III, 2008; Karpicke & Blunt, 2011). Students also have the ability to rewind, pause, and speed up lectures in addition to watching them with subtitles, giving students more control in the way they navigate course content.
Data and analytics: Imagine starting your class knowing that 65% of your students answered the same wrong answer on a quiz they took the night before. In blended learning environments, educators are able to collect such data on student performance, giving opportunities to target their lessons and address gaps in student comprehension while refining their teaching strategies for subsequent offerings of the course. This is particularly useful for those employing Just-in-time teaching.
Student-driven lectures: Despite the great variance of student learning styles and ability that exist in classrooms, traditional lectures are often one-size fits all. Video lectures, on the other hand, put students in the drivers seat, and allow them to engage with the material at their own pace, review confusing concepts, or break the lectures into easily-digestible chunks. More generally, flipped classrooms allow instructors to more easily implement a variety of differentiated instruction techniques.
Mastery learning: In mastery learning, students are encouraged to master each concept before proceeding to the next. This practice has been shown to increase student performance by about one standard deviation over more traditional forms of instruction (Bloom, 1984). To put things into perspective, this means that if 50% of a class passed a course in a traditional educational environment, approximately 84% of the same class would pass in a mastery-learning environment. For example, Martin Odersky from École Polytechnique Fédérale de Lausanne allowed students to take his quizzes as many times as they wanted in his Functional Programming Principles in Scala course. Many students kept taking the quiz until they received 100%, which unsurprisingly led to higher scores for that particular quiz. But students who chose to improve their scores by taking the quiz multiple times did better on future quizzes as well. For students of similar current performance, mastery-based score improvements correlate with future performance:
Multiple submissions encourage grade improvement. The data shown correspond to total cumulative scores across all assignments for all students who submitted every assignment in Ecole Polytechnique Federale de Lausanne’s “Functional Programming Principles in Scala” taught by Martin Odersky.
Integrating Coursera’s assessment tools (e.g. randomized assignments, immediate feedback, etc) into a flipped classroom allows for the utilization of this powerful pedagogical strategy.
Flipped classrooms (and blended learning environments more broadly) can vary greatly depending on the attributes of the class, student needs, and the level of involvement required by the instructor. Below, we’ve outlined some possible blended learning models. These models are placed on a spectrum based on the level of instructor involvement in the course; we then further group these student-instructor interactions as synchronous and nonsynchronous:
In this section, we’ll describe the two broad categories of blended learning environments: instructor- and student-led.
In an instructor-led flipped classroom, the instructor is typically involved in every aspect of the course and has ample opportunities to engage with students. Outside of class, students engage with the more passive elements of the course; inside class, students participate in active learning activities coordinated by the instructor. Such activities might include group discussions, role-playing games, case studies, group problem solving, team projects, etc.
One common instructor-led model involves Just-in-Time Teaching (JiTT). In most JiTT learning environments, students respond electronically to autograded assignments due shortly before class. After the specified deadline, the instructor reviews student submissions "just-in-time," and uses these data to adjust the lecture to better suit the students' needs. Mohamed Noor of Duke used this strategy with considerable success. The night before every lecture, Noor would review his students’ answers to autograded assessments; he would then spend the first 5-10 minutes of lecture overviewing the concepts that students found confusing.
In a student-led blended learning environment, the instructor’s involvement in the day-to-day running of the course is relatively low. Students have more flexibility and freedom in navigating the course content, and instructors have the option of using a variety of lower-touch strategies to interact with their students. In this model, the instructor acts more acts as an organizer or guide in the learning process, providing the resources and support necessary to allow learners to develop and shape their own learning. Examples of low-touch strategies that can be implemented by the instructor or other course facilitators include:
Doug Fisher from Vanderbilt utilized a student-led flipped classroom model in his Fall 2012 Machine Learning course. Fisher required that his students take Andrew Ng’s Machine Learning in its entirety and present a Statement of Accomplishment at the conclusion of the course. Because the course’s assessments, lectures, and assignments were automated on the Coursera platform, Fisher only had to meet with his students once a week for 90 minutes. During these meetings, Fisher and his students discussed assigned journal readings, sometimes corresponding closely to topics in Ng’s lectures and sometimes very different topics that Fisher thought it important to cover.
The following lecture series, created by Courser co-founder Andrew Ng, provide a high-level overview of the flipped classroom:
Learn why instructors should consider flipping their classrooms.
Learn the basics of the flipped classroom, first-timer tips, and potential pitfalls.
Learn about engaged learning activities that can be used with students during class time.
Learn strategies for engaging your students in collaborative learning activities.
Learn about Just-in-Time Teaching, a strategy which allows instructors to use class time to target the concepts that their students need help with most.
Learn how different tools on the Coursera platform can be used in a flipped classroom environment.
Watch a review of the major elements of the Flipped Classroom Lecture Series.
Ready to flip your class now? While every flipped classroom will vary depending on a number of factors (e.g. class size, content, professor, school resources, etc.), many follow a process similar to the following:
Flipping your class for the first time can be a difficult process. To help you put the best foot forward, we’ve compiled a list of potential pitfalls collected from our partners:
Pitfall: Professors don’t “sell” the flipped classroom
Notes: Given the pervasiveness of the standard lecture, some students may resist the implementation of a flipped classroom. Be mindful of how you frame their new learning environment. For example, some professors avoid using words such as “experimenting” or “flipping,” as these terms are loaded with preconceived notions which might not be true or applicable to your class. However, it’s also important to be as transparent as possible about the motivations behind the change as well as the benefits of flipped classrooms and blended learning environments more broadly. Similarly, make sure that your students understand the logistics of the new system, from the grading of participation and assessments to how to use the technology. As Adrienne Williams of UCI reports, its important to “sell the class”: “Many students are concerned that the flipped format will force them to "learn on their own" and increase their study load. Take time to explain the research supporting it. Here is a sample of our "Introduction to the class" video students watched before the first day of class in order to understand the format, and a copy of the first day's in-class presentation slides that explain why we are flipping.”
Pitfall: Classroom space is not conducive to flipping
Notes: The physical space of a classroom is one of the most important (and often, underestimated) factors of a successful flip. Scott Klemmer reported that the classroom for his Human-Computer Interaction class was “terrible” for flipping, and that this was the biggest obstacle he encountered during the flip. Because Stanford only has seven classrooms that hold more than 239 students and they’re all “too cavernous,” Klemmer plans to cap enrollments for his next offering at 239 people, which will enable him to use more flip-friendly spaces on campus. Klemmer stated that his “number one recommendation for Stanford is to physically to redesign the classrooms” to make them more conducive to flipping.
Pitfall: Students don’t show up for lecture
Notes: Flipping your class can have mixed effects on attendance: Duke observed improved attendance for each of their flipped classes because students felt that class time was more engaging; other partners have experienced a dip in attendance after flipping a class, possibly due to students feeling there’s no need to come to class when the lectures are readily accessible online. These anecdotes highlight the importance of properly pitching the flipped classroom when implementing it for the first time, and also incentivizing students to come to lecture. Mohamad Noor from Duke noticed a significant drop in attendance for his course: of the 450 students enrolled, an average of only 250 or so showed up for his in-class activities. The next time he runs his course, Noor will administer in-class quizzes to address the drop in attendance. To accommodate his students’ schedules, he’s considering allowing them to drop their seven lowest quizzes. He hopes this will nudge them to attend the in-class lectures, while still providing enough flexibility for those with legitimate excuses for missing class.
Pitfall: In-class activities are not relevant to out-of-class lectures
Notes: Many professors have reported difficulty in designing in-class activities that cohere with the out-of-class lectures. Scott Klemmer stated that one of the greatest challenges that arises when putting your lectures outside of class is trying to design in-class activities that overlap with those lectures. Similarly, Doug Fisher (Vanderbilt) reported that the most difficult part of flipping his classroom was finding moments of overlap and connectedness between the journal articles his students discussed in class and the pre-recorded lecture videos on Coursera: “students had more of a sense of schism or disconnect..the degree of coupling is something that instructors should really focus on when planning their flipped classrooms.”
Notes: Many educators are accustomed to giving assignments outside of class that complement their in-class lectures. When flipping, it’s important to be sensitive to how changes to the classroom structure affect your students’ workload. In particular, by not adjusting the amount of homework students are expected to complete outside class in addition to the added video lectures, professors risk doubling the amount of out-of-class work. To avoid overworking your students, try reducing their homework by focusing your learning goals and trimming unnecessary content. This tip was summed up well by Adrienne Williams of UCI: “Remove unnecessary content. We were able to easily convert our lectures to video because we had already written our learning goals, and had previously trimmed extraneous content from our lectures to make space for clicker questions and group activities. If you are converting a dense, traditional lecture to a flipped class, take this opportunity to remove content that only tests student recall and does not further your learning goals. This essay by G. Bergrom at Univ. Wisconsin is an excellent example of how to modify a course.” Instructors could also try reducing class time to compensate for the increased amount of out-of-class work.
Pitfall: Not all students have equal access to online materials
Notes: It’s important to remember that not all students have access to basic technologies such as computers or high-speed internet. With this in mind, it’s important that every student is aware of public, easily-accessible computer resources (e.g. school libraries, public libraries, internet cafes, etc) that will allow them to access the course materials. Publicly sharing such information helps ensure that those who need extra help will have it available to them, even if they’re too embarrassed to ask. Also, make sure students understand the technology before asking them to use it. Stanford’s Dan McFarland ensured that his students were up to speed by having representatives from the IT office in class when students tried to access his course content for the first time.
Pitfall: Instructors don’t realize the amount of preparation necessary for in-class activities
Notes: Some professors have reported being surprised at the amount of work that flipping a classroom requires. After flipping his Machine Learning Course, Doug Fisher from Vanderbilt says that one of the main things he learned was that the instructor “has to do a lot more work in advance when flipping the classroom.” Don’t fall into the trap of assuming that technology will reduce amount of the time you invest in your course; many professors report investing more time in their flipped courses (at least initially- reusable lectures can increase efficiency in the long run) due to the extra time involved in filming their lectures and creating in-class activities. Remember, it’s not the technology, but the pedagogy that drives the success of blended learning environments. Flipping your classroom for the first time will take work, innovation, and perseverance to realize the long-term gains.
Pitfall: Students aren’t completing work assigned for outside of class
Notes: Some professors have reported difficulty in getting their students to complete the pre-class activities (e.g. watching video lectures). This can be problematic in a flipped classroom environment, in which the pre-class activities are often a necessary prerequisite for the in-class activities. As such, students who don’t prepare are unable to realize the full benefit of the flipped classroom. Using assignments to test student comprehension of the lecture videos is one way to ensure that students do the work outside of class and are therefore able to be more productive in the subsequent lectures. Another strategy is to encourage students to watch lectures together in groups. This will help students keep one another on track while providing additional opportunities for discussion around the presented topics. Scott Rixnor and Joe Warren of Rice addressed this problem by dividing their students into two groups: those who had done the homework and those who hadn’t. Those who hadn't spent class time working on the homework; those who had were presented with challenging problems that extended the homework in interesting ways. Rixner reports: “By splitting up class...I was able to give individualized attention to students lagging behind; for the students running ahead, I could give individualized attention...and push them further and have interesting debates and discussions about the material in the class” (more info here).
Pitfall: Instructors have difficulty accommodating the varying ability levels of their students during in-class activities.
Notes: Accommodating for the variation in student ability in a flipped classroom can be difficult. Scott Klemmer from Stanford experienced this when trying to coordinate programming labs for his Human-Computer Interaction course: “the challenge is that different students program at different rates. In any project that I assign, there will always be one student who is done in 40 seconds, and another who won’t be able to open the file by the end of the class.” Klemmer said that designing in-class activities that can accommodate both of these populations is very tricky.
Pitfall: Instructors are unable to successfully flip large lectures
Notes: Flipping a lecture hall of 300 is a daunting task; doing so with limited or no teaching assistants may seem impossible. Some instructors have found it necessary to break up their classes into smaller groups. Philip Zelikow of UVA decided to break his 120-person class into two back-to-back sessions because he finds it “prohibitively hard” to run discussion if there are more than 50 or 60 students in a class. Similarly, Maya Adam reported that she prefers not to teach a class of more than 50 (“I like students to feel that I know who they are”), so she divided the students into two smaller groups.
For those who wish to flip large lectures with limited resources, try training a group of student TAs to facilitate your blended learning activities. For example, a professor wishing to flip their large lecture could take the following steps:
While this model will work for some, there are still a number of obstacles when flipping large classrooms. One of these is dealing with the physical constraints of the classroom: sometimes, no matter how many TAs or how well-planned the activities, a classroom simply will not be conducive to small group problem solving activities. In these situations, try relying on partner work, individual work, or engaged learning activities that can be completed with one or two students.
Pitfall: Professors don’t put in place mechanisms to collect metrics for feedback
Notes: There’s no flipped formula that will work for every classroom, so its important to use metrics to evaluate the efficacy of your newly-flipped class. Give your students ample opportunities to give feedback on the activities, assessments, and their overall course experience. For example, you could compare different active learning techniques across sections by comparing assessment scores. This feedback paired with performance metrics can be used to recalibrate and refine your flipped classroom experience. Duke professor Mohamad Noor used qualitative feedback to adjust his lessons by including the question “what did you find interesting or confusing?” at the end of each of his autograded assessments.
Pitfall: Professors don’t engage fellow instructors about their flipped experiences
Notes: Often times your fellow educators will be your best resource for designing and implementing an effective flipped classroom; use them as much as possible by sharing your ideas and soliciting their input! Some of our instructors have had great success in setting up regular lunch chats with their colleagues, during which they discuss their flipped classroom experiences and brainstorm innovative ways to more deeply engage students with the material. Also, be sure to visit the Partners’ Portal forums to discuss flipped classrooms and blended learning with others in the Coursera community.
Pitfall: Professors still act as the “sage on the stage” rather than the “guide on the side”
Notes: Perhaps one of the most common and deeply-ingrained conventions of the traditional lecture is that in-class time revolves around the professor. In flipped learning environments, class time revolves around the students, a switch Jeremy Adelman of Princeton experienced; Adelman described his new role in his flipped classroom as that of a “coach.” Getting used to this shift can be difficult, especially for professors who have grown accustomed to being the “sage on the stage” after many years of lecturing. There are ways to proactively address these old habits: Andrew Ng from Stanford would consciously turn his back to students and do his own work during in-class activities, encouraging them to not rely on him for the answers and communicating that that they were facilitating the activity, and therefore, their own learning.
Filming your lectures is just the beginning of flipping your class. The next step, often the most important and challenging, is deciding what to do with the newly-available in-class time. In this section, we’ll review some effective learning activities for a flipped classroom and tips for their implementation.
Application activities give students the opportunity to use what they’ve learned to solve problems and grapple with learned concepts in novel ways. The variations of application activities are endless. Some examples from different courses include:
Extension activities require students to derive properties or theoretical extensions of what they’ve learned. Some professors characterize extensions as the concepts or skills that they would’ve taught had there been more time. These activities deepen students’ understanding of the material by encouraging them to apply learned concepts to a novel domain. Some examples of extension activities include:
Scott Rixner and Joe Warren used extension activities in their Introduction to Python course. Once in class, students were divided into two groups: those who had done the homework and those who hadn’t. Those who hadn't spent class time working on the homework; those who had were presented with challenging problems that extended the homework in interesting ways. Rixner reports: “By splitting up class...I was able to give individualized attention to students lagging behind; for the students running ahead, I could give individualized attention...and push them further and have interesting debates and discussions about the material in the class.”
In a sequence of questions activity, complex problems are broken into smaller parts and then solved systematically with the students. This process of breaking complex problems into smaller bits has been shown to improve learning outcomes (Marzano, Pickering, & Pollock, 2001). Some best practices for this technique include:
Examples of the sequence of questions technique include:
In many traditional classrooms, students are viewed as content consumers whose role is to absorb and reproduce information for various assessments. In blended learning environments, students are able to create and share “student-generated content” with their peers. There are a variety of ways to solicit, curate, and use student-generated content that vary by course and the available technologies. Some common ways to have students interact with student-generated content that can be applied across any number of courses include:
There are many advantages to integrating student-generated content into your flipped classroom. For one, students feel as if they have more ownership over their learning and the trajectory of the course. In addition, the use of student-generated content in a course has been correlated with improved student outcomes. For example, in a study conducted at the University of Edinburgh, researchers demonstrated a significant correlation between participation in student-generated content activities and course performance in an introductory Physics course. The following graph illustrates that students who generated more content performed significantly better than their peers on in-class exams:
A number of Coursera tools can be used to facilitate the creation and sharing of student generated content. These include:
Keep in mind the following best practices as you attempt activities involving student-generated content:
In experiential learning activities, students learn through immersive, hands-on learning experiences. These may differ from application activities in that students are learning new concepts from experiences, rather than reinforcing previously learned activities through novel applications.
The pedagogical benefits of these learning experiences have been well-documented in the literature and demonstrate the efficacy of simulation environments and modeling in enhancing learning (Gordin & Pea, 1995).
Experiential learning activities can take a number of forms, including role-playing, experimentation demonstrations, trips, labs, computer simulations, competitions, debates, and trips. When evaluating your planned experiential learning activity, try asking yourself the following questions:
Modified lectures can be used to introduce new topics or individuals into the classroom. Example modified lectures could include guest lectures, advanced lectures, or lectures on current events. When possible, try to engage learners in these guest lectures as much as possible and avoid the passive-lecture paradigm, possibly by incorporating complementary engaged learning activities.
Organized discussion activities are frequently used in flipped learning environments to facilitate peer-to-peer exchanges and a deeper engagement with the course content. Many effective discussion activities:
Discussion activities have the potential to yield great benefits to your flipped class. These include:
There are a number of ways to structure your in-class discussion activities. Check out some ideas for ideas on engaging students in classroom discussions from Brookfield and Preskill’s Discussion as a way of Teaching (1999) below:
The Circle of Voices Activity
Circular Response Activity
There are six ground rules in this activity:
A very simple engaged learning activity (especially for those teaching large lectures) is to give students a packet of problems to solve in class. In Mohamed Noor’s Introduction to Genetics and Evolution lecture, the first 5-10 minutes were spent overviewing the concepts that students identified as confusing. After this review exercise, students would break into groups and solve a packet of questions distributed by Noor and his TAs. During this time, Noor and his TA’s would walk around the room and offer help or challenge students as needed (e.g. “How did you come up with this answer?” “Why do you think that?” etc.). Twenty minutes before the end of the class, Mohammed would review the solutions to the problems in the packet. The last 5-10 minutes of the course were spent discussing what students indicated as “interesting” on the pre-class quiz (e.g. “We talked about topics like 23 and me, what services they offer, the ethical responses to the service, etc.”)
Classroom Assessment Techniques (CATs) are ongoing formative assessments that facilitate learning and provide both student and professors with feedback about their learning and teaching, respectively. For example, used a quick CAT Mohamed Noor (Duke) by asking his students "What did you find interesting or confusing?" after every quiz. He would then review their progress the night before class, use their answers to guide the slides he'd make for the class period. At the beginning of class, Mohamed would address what students found confusing; at the end, he would elicit conversations around what they found interesting.
Effective CATs are typically learner-centered, beneficial to both the instructor and student, formative, context-specific and ongoing.
Some commonly used CATs that have been used successfully in higher learning environments include:
The minute paper: The minute paper is a short informal writing assignment that typically occurs at the end of a lecture. Minute papers are straightforward, require relatively little time, and can be assessed easily, which is likely why they are one of the most commonly used CAT methods used in large lectures. Steps for using the minute paper technique are outlined below:
Concept maps: concept maps are a common CAT that involve asking students to produce drawings and diagrams that illustrate the connections between learned concepts. Research has shown that presenting information graphically as well as symbolically reinforces vocabulary learning and supports reading skills (Brookbank Grover, Kullberg, & Strawser, 1999; Moore & Readence 1984). With the aid of concept maps, students can accomplish the following:
Additional CAT activities can be found here. For more information about CATs, check out this overview of Classroom Assessment Techniques: A Handbook for College Teachers, by Thomas A. Angelo and K. Patricia Cross.
Most engaged learning activities can be used in collaborative learning environments. The engaged-learning activities below are unique in that they rely on collaboration amongst peers, and lend themselves well to collaborative/group learning opportunities.
Regardless of the flipped model you plan to implement, the introduction of collaborative learning activities are encouraged; studies have shown that students who meet in learning groups at least once a week show significant improvements in learning (Marzano, Pickering, & Pollock, 2001).
The foundation of most collaborative learning activities in flipped classrooms is the formation of effective student groups. There are a number of factors to consider when grouping your students, some of which include:
Some best practices for putting your students into effective groups include:
Small-group problem solving activities are some of the most popular uses of in-class time. After breaking up the class into small groups, the instructor can present any number of problem-solving activities (many of which could be modified versions of the aforementioned engaged learning activities) for the groups to tackle. While students are working, the professor and his/her TAs can visit each of the groups, guide the conversation, help the groups that are struggling, and further challenge those that are not. Many studies have demonstrated the efficacy of small-group problem solving, such as those that suggest improved learning in STEM classes, particularly amongst science minorities such as women and ethnic minorities.
“Peers make the best instructors because they remember what it is not to understand.”
Mazur’s peer instruction model is a systematic process for encouraging collaborative learning amongst peers. This model is easy to implement, cost-effective, and applicable to nearly any subject or class. The model in it’s most basic form typically includes the following steps:
Beth Simon, Professor in the Department Of Computer Science and Engineering at UC San Diego, is an expert and strong advocate of peer instruction methodologies. Check out some of her awesome resources below to learn the ins and outs of effective peer instruction techniques:
In peer feedback activities, students participate in the process of evaluating their peers. This turns every stage of the assessment process into a learning opportunity, from completing the assignment to evaluating the work of others. A study conducted by Atay and Kurt (2007) revealed that peer feedback activities can increase student self-confidence, build social affective strategies, and allow for deeper engagement with the course material.
This strategy can be applied to nearly any course or subject matter. One common peer feedback process is comprised of three stages:
Despite many instructor’s emphasis on the creation of the assignment, well-written rubrics have been shown to be particularly important in improving learning outcomes: research has shown that criterion-referenced feedback can significantly improve student understanding (Crooks, 1988; Wilburn & Felps, 1983). Check out some effective rubrics created by other instructors on the Peer Assessment Gallery in the Partners’ Portal, and a detailed overview of creating effective rubrics and assignments in our support docs. Consider the following suggestions when creating your rubric:
Consider the following suggestions when creating your peer assessment assignment:
Just-in-Time Teaching is a pedagogical strategy in which instructors use student-performance data to understand which concepts students are struggling with and pinpoint particular students who are more at-risk, in order to give these students more personalized attention.
In most JiTT learning environments, students respond electronically to autograded assignments which are due shortly before class. After the specified deadline, the instructor reads student submissions "just-in-time," and uses these data adjust the classroom lesson to address the students' needs. Thus, the heart of JiTT is a "feedback loop" formed by the students' performance on out-of-class activities and the subsequent lectures tailored to their performance:
Effective JiTT relies on well-designed assessments administered before lectures. These assessments will likely be short, autograded, and designed to give professors relevant feedback on the direction of lectures that follow. Ideally, the pre-class assessments are administered as close as possible (e.g. the night before) to the subsequent lecture, ensuring that the information covered is as relevant and timely as possible:
The different strategies for using these analytics are constantly evolving. For example, some have found great success in grouping students in different sessions based on their assessment performance: if a group of students are struggling with a particular concept, one could hold a session just for those students, or an advanced session for students who displayed a greater mastery of the material.
Interested in trying JiTT? Read over the following best practices generated from the experiences of other JiTT instructors:
Maximize student participation: Make an effort to maximize student participation in the pre-class assessments. The more students who participate, the better understanding you’ll have of your class’ relative strengths and weaknesses.
Ask the right questions: Be meticulous when structuring your pre-class assessments. Ensure that each of the desired concepts is covered, and that each question will yield relevant and actionable data. One way to test each question is to make a list of the possible explanations of why students might choose each incorrect response; the fewer explanations you’re able to generate, the more confident you’ll be in addressing the source of their confusion in lecture.
Be realistic: After looking over the pre-class assessments, select a reasonable number of concepts to cover in the following lecture. Remember that the pre-class data will likely reveal more gaps in student comprehension that you can cover in a single lecture. Instead of superficially addressing all of these gaps, select a few that you can thoroughly cover within the lecture before moving onto the next concept, a strategy that helps engender the benefits of mastery learning.
Autograding: Whenever possible, rely on auto-graded assessments to inform the direction of your lectures.
Give timely feedback: Whenever possible, try to address student comprehension gaps as soon as they are identified: research has shown that delay in providing students with feedback on misunderstood concepts diminishes its value for learning (Banger-Drowns, Kulik, Kulik, & Morgan, 1991).
Allow sufficient time to calibrate lectures: While it’s important to minimize the amount of time between assessment and the subsequent lecture, it’s also important to allow sufficient time to calibrate the lectures based on student performance. Each professor will differ in this regard depending on the content and specific features of the class: some professors find they need a day or more to review student data; others wait until just before the start of class.
The Coursera platform collects a tremendous amount of data about the way students learn and navigate a course. These data are valuable in tracking student progress and identifying gaps in their comprehension, a critical element of Just-in-Time teaching strategies.
There are four ways of accessing analytics for your course on the Coursera platform:
The Coursera platform also gives instructors access to a number of quiz exports, including:
You can access more information about data and analytics on the Coursera platform on our support docs.
Online lectures are a critical element of flipped classroom environments. Instead of giving the same lecture year after year, instructors are able to craft modular, high-quality, well-structured video lectures that can be used multiple times across different classes. Jeremy Adelman of Princeton reported that his benefited from “being able to replay lectures and work off comprehensive notes instead of the hasty scrawl derived from watching live lectures” in his class “A History of the World since 1300.”
The Coursera lecture interface gives students an unprecedented level of personalization as they navigate through different elements of the course. Key features include:
Consequently, interactive video lectures yield a number of benefits, including:
Adrienne Williams of UCI put together some useful best practices on creating effective flipped-classroom lectures: “Keep it short. If you require more than 12 minutes of brisk speaking in order to cover the content, consider reducing content in order to free up time for connections and applications. Plan your class activities in conjunction with developing the pre-class video. The class activities should reinforce the video material, not repeat it or present new information to memorize. Students like to see at least some "face time" during the video. Incorporate video of you talking if you are comfortable doing so.”
In-video quizzes are a type of informal assessment that appears within lecture videos, typically after a key concept has been explained. In-video quizzes allow for lecture videos that are more interactive, dynamic, and personalized. In addition, these quizzes facilitate retrieval-based learning and enable students to test their understanding on the spot. Research shows that such interactivity plays a critical role of the efficacy of videos in e-learning environments (Zhang et al., 2006).In addition, research shows that even simple retrieval questions have significant pedagogical value. For example, in two papers in Science, (Karpicke & Roediger III, 2008; Karpicke & Blunt, 2011) show that activities that require students to retrieve or reconstruct knowledge produces significant gains in learning - much more so than many other learning strategies.
Click here to watch a video overviewing about Coursera’s in-video quiz functionality:
For more information about in-video quizzes, please check out this support doc.
The community forums allow students to collaborate, share ideas, and give feedback in an immersive, dynamic, shared digital space. There are numerous applications of the forums in flipped classrooms, including:
You can access more information about community forums on our support docs.
Coursera’s peer assessments are a useful tool in flipped learning environments. Our peer assessment interface allows you to create organized, linear peer assessment activities, and has many advantages over in-class peer grading activities. These include:
For more information on peer assessments, check out our support docs.
Editable wiki-pages enable students to collaborate, share ideas, and curate content. This resource is particularly useful when trying to integrate student generated content into your course.
Autograded assignments allow professors to more efficiently and effectively collect and respond to student performance. The Coursera platform allows for a number of different autograded assignment types including multiple choice, math expressions, short answers, and computer programs.
What are flipped classrooms?
In flipped classrooms, technology enables instructors to more actively engage with their students. This is typically accomplished by using in-class time for active learning exercises and out-of-class time for watching recorded lectures.
Why flip your classroom?
Flipped classrooms allow for an unprecedented level of interactivity and insight into student learning and performance. By moving passive learning experiences outside of class, professors are able to use in-class time to engage students in activities such as small group problem solving, peer grading, and experiential learning activities; such active learning exercises have been shown to significantly improve educational outcomes (Deslauriers, Schelew and Wieman, 2011; Topping, 1998; Mazur, 2009). Additionally, autograded assessment data helps instructors target their lessons and address student comprehension gaps while refining their teaching strategies for subsequent course offerings.
By moving lectures online, instructors are able to make modular, high-quality lecture videos that can be reused repeatedly across classes. Video lectures also allow for considerably more interactivity than their brick-and-mortar counterparts: dynamic in-video quizzes give students an opportunity to test their understanding while improving learning outcomes (Agarwal et al., 2012; Karpicke and Grimalidi, 2012; Roediger et al., 2006). Coursera’s autograded assessments allow instructors to efficiently evaluate student comprehension while facilitating the use of mastery learning techniques (Bloom, 1984; Kulik et al., 1990). Perhaps more importantly, these autograded assessments give instructors valuable data related to student learning and performance.
Blended learning models
Flipped classrooms can vary greatly depending on the attributes of the class, student needs, and the level of involvement required by the instructor. In this document, we presented two broad categories of blended learning environments: student- and instructor-led.
In a student-led flipped classroom, the instructor’s involvement in the day-to-day running of the course is relatively low. Students have more flexibility and freedom in navigating the course content, and instructors have the option of using a variety of lower-touch strategies to interact with their students, including: answering student questions via email or on the community forums; offering feedback on student work; providing motivational nudges or reminders throughout the course; periodically holding open office hours; or identifying students who are struggling with a concept and holding a facilitated discussion (perhaps in a small group) on that topic to help them get over the hurdle.
In an instructor-led flipped classroom, the instructor is much more involved in the coordination of the course. Outside of class, students engage with the more passive elements of the course; inside class, students participate in active learning activities coordinated by the instructor. Such activities might include group discussions, role-playing games, case studies, group problem solving, team projects, etc. Instructors can also employ Just-in-Time Teaching strategies by using student assessment data to determine the concepts or questions that students are struggling with and that should be addressed in class.
Just looking for the gist? This flipped cheat sheet overviews each of the major elements of this resource.
Why flip your classroom?
Common step-by-step process for flipping your classroom:
Tips and tricks for first-time flippers:
A list of some engaged learning activities for in-class time:
A list of some collaborative Learning Activities for in-class time:
A brief summary of Just-in-time teaching (JiTT):
Just-in-Time Teaching provides instructors with valuable student-performance data that inform the content and method of their in-class instruction. This typically involves students responding electronically to autograded assignments due shortly before class and the instructor using these data to better address student needs in lecture. Some best practices for JiTT include maximizing student participation on pre-class assessments, using autograded assessments, giving timely feedback, and systematically tackling comprehension gaps.
A brief summary of differentiated instruction:
“Differentiated instruction” refers to the practice of varying the ways in which students engage with course materials, depending on their students’ unique attributes. Coupling technology with increased opportunities to engage with students during class in flipped classrooms allows professors to more easily differentiate their instruction. Instructors can differentiate a number of features of their instruction including the classroom content, their presentation of new concepts, their students’ expected product (e.g. evaluations, reports, projects, assignments, and assessments), and their students’ learning environment.
A list of some Coursera tools that support flipped classrooms and blended learning environments:
Check out the academic and news articles below to learn more about flipped classrooms and blended learning.
The list below features some academic articles relevant to flipped classrooms and blended learning:
A.A. Gokhale, (1995). Collaborative Learning Enhances Critical Thinking. Journal of Technology Education:
This study demonstrates that students who learn collaboratively develop better
critical thinking skills than students who learn individually.
Bloom, B., (1984). The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective as One-to-One Tutoring. Educational Researcher, Vol. 13, No. 6, pp. 4-16
This paper examines the effects of “mastery learning,” a pedagogical strategy in which
students are encouraged to master each concept before proceeding to the next. This practice has been shown to increase student performance by about one standard deviation over more traditional forms of instruction.
Bowen, W.G., Chingos, M.M., Lack, K.L., & Nygren, T.I. (2012). Interactive Learning Online at Public Universities: Evidence from Randomized Trials. ITHAKA:
“We find that learning outcomes are essentially the same—that students in the hybrid format "pay no price” for this mode of instruction in terms of pass rates, final exam scores, and performance on a standardized assessment of statistical literacy. These zero-difference coefficients are precisely estimated. We also conduct speculative cost simulations and find that adopting hybrid models of instruction in large introductory courses have the potential to significantly reduce instructor compensation costs in the long run.”
Crouch CH and Mazur E (2001). Peer instruction: Ten years of experience and results. American Journal of Physics, 69: 970-977.
This study explores the efficacy of peer instruction.
In this study, Deslauriers, Schelew and Wieman compared two large sections of an
introductory undergraduate physics course: one section was taught as a traditional lecture by an experienced, highly-rated instructor; the other was taught by an inexperienced instructor using active learning strategies. In the active-learning group, student engagement nearly doubled, attendance increased by 20%, and average scores on assessments increased from 41% to 74%. Read an excellent summary of the study from The Economist here.
Hake R (1998). Interactive-engagement versus traditional methods: A six-thousand-student survey of mechanics test data for introductory physics courses. American Journal of Physics 66: 64-74.
This study demonstrates the efficacy of interactive engagement methods. Students exhibited improved educational outcomes almost two standard deviations higher than those observed in the traditional courses.
Lovett, M., Meyer, O., & Thille, C. (2008). The Open Learning Initiative: Measuring the effectiveness of the OLI statistics course in accelerating student learning. Journal of Interactive Media in Education:
“In this study, results showed that OLI-Statistics students learned a full
semester’s worth of material in half as much time and performed as well or better than students learning from traditional instruction over a full semester.”
Marzano, Pickering, & Pollock (2001). Classroom Instruction that Works: Research-based Strategies for Increasing Student Achievement. ASCD.
This book includes a number of evidence based pedagogical finding, including that students who meet in learning groups at least once a week show significant improvements in learning and that the process of breaking complex problems into smaller bits has been shown to improve learning outcomes.
Redfield, D. L., & Rousseau, E. W. (1981). A meta-analysis of experimental research on teacher questioning behavior. Review of Educational Research, 51(2), 237-245.
“Results show that gains in achievement can be expected when higher cognitive questions assume a predominant
role during classroom instruction.”
Roediger, H., Karpicke, J. (2006) Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention. Psychological Science:
This study demonstrates that the process of testing students on learned material is considerably more effective in
improving retention than traditional studying methods.
Kearsley, G., Shneiderman, B., (1998) Engagement Theory: A framework for technology-based teaching and learning. Educational Technology, v38 n5
“The fundamental idea underlying engagement theory is that students must be meaningfully engaged in learning activities through interaction with others and worthwhile tasks. While in principle, such engagement could occur without the use of technology, we believe that technology can facilitate engagement in ways which are difficult to achieve otherwise. So engagement theory is intended to be a conceptual framework for technology-based learning and teaching.”
Scheines, R., Leinhardt, G., Smith, J., & Cho, K. (2005). Replacing lecture with web-based course materials. Journal of Educational Computing Research, 32, 1, 1-26:
“This article overviews the efficacy of various features of the online experience and student learning strategies. The researchers found that students who entirely replaced going to lecture with doing online modules did as well and usually better than those who went to lecture. In addition, the study demonstrated the effectiveness of incorporating frequent interactive comprehension checks into the online material (something that is difficult to do in lecture). The results also show that smaller, more interactive educational environments are effective than large, passive lectures.”
University of Wisconsin-Madison. Flipping Courses: Transitioning From Traditional Courses to a Blended‐Learning Approach, 2013
A fantastic high-level overview of flipping the classroom from the University of Wisconsin-Madison.
U.S. Department of Education. Evaluation of Evidence-Based Practices in Online Learning: A Meta-Analysis and Review of Online Learning Studies. Washington, D.C., 2009.
This meta-analysis from the Department of Education is based on 45 studies and shows that online learning is as effective as face-to-face learning, and that blended learning is considerably more effective than either.
<iframe style="width:100%;height:24990px;border:none;" src="https://docs.google.com/document/d/1arP1QAkSyVcxKYXgTJWCrJf02NdephTVGQltsw-S1fQ/pub?embedded=true"></iframe>Back to the top