By Yong ZhaoFoundation Distinguished Professor, School of Education, University of Kansas; Professor in Educational Leadership, Melbourne Graduate School of Education, University of Melbourne. You can find Dr. Zhao online at or on Twitter @yongzhaoEd.

This has been adapted from a blog post at

Going online seems to be the first and instinctual choice for schools to continue to offer education when schools are forced to close during the COVID-19 pandemic. Deciding to shift to online education seems an easy one because there does not exist many alternatives, but deciding what online education to offer is a harder one because there is no one online education.

Online education can take a wide variety of shapes and forms due to the numerous ways to combine the basic ingredients such as technology platforms (learning management systems, broadcasting platforms, social media platforms etc.), media modality (text, animations, videos, audios, etc.), temporal arrangements (synchronous and asynchronous), instructional approaches (direct instruction, inquiry-based, product-oriented, flipped classroom etc.), student arrangement (small groups to massive groups), teacher roles, and more such as frequency of interactions among students and instructor. Different combinations result in different forms of online education.

Hence, when making decisions about what forms of online education to use, it is important to examine the evidence of effectiveness of specific configurations of each program and the measured outcomes rather than accepting or rejecting a program or model. When examining the evidence of effectiveness of online approaches and programs, we need to consider both effects and side effects, that is, desirable and undesirable outcomes.

My book What Works May Hurt: Side Effects in Education (Teachers College Press, 2018) makes the case for why we must consider both effects and side effects of educational policies and practices in order to improve education. Furthermore, I analyze why side effects occur and what questions we must ask when faced with different choices of educational practices. These questions can be used to guide our decision about online education.

What Works for What?

Online education (or education) can have many different outcomes: political, financial, societal, and educational. Different online education models have different effects on different outcomes and thus suit different purposes.

Broadcasting live or recorded lectures, for example, can be very effective for the purpose of simulating traditional teacher-centered schooling for students: attending class at specific times, receiving uniform information from the same teacher at the same time as their peers, and carrying out the same tasks as their peers at the same time. But it is not effective for attending to individual differences, accommodating different learners, or addressing students’ social-emotional needs.

Having students work on projects individually with teachers providing feedback and support to individual students or small groups asynchronously can be effective in engaging students in more authentic work, have more autonomy and more personalized experiences, and receive more individualized attention. But it may not be effective in uniform knowledge transmission and thus would be a poor choice for the purpose of ensuring coverage of the planned curriculum.

What Works for Whom?

The same program can have different effects on different stakeholders in online education. Some programs may work very well for political leaders but may not work well for teachers. Some may work for teachers but not students. Some may work for some students but not others. No one program can work equally well for all students.

It is well known that students vary a great deal as learners in many aspects: academic abilities, personalities, interests, motivations, and experiences. These differences can affect the effects and effectiveness of educational programs, online or face-to-face. When students are physically in front of us, these differences may be more obvious than when they are out of sight. Thus we need to be mindful all the time that these differences remain in online education and keep asking how online offerings affect different students.

More important, online education has the potential to amplify the impact of student differences, in particular factors associated with student background. We already know that home environment plays a significant role on student learning. Schools are supposed to be an equalizer by mediating the differences in family backgrounds. When schools are closed, home environments play a much more significant role than when they come to school. Thus when examining the effects of online education, it is even more important to ask the question what online programs for what kind of students.

What and Who Gets Hurt by What Works?

In the pursuit of what works, we must keep in mind what works for some students can do harm to other students. For example, some programs may work for students who have more experiences with online activities but can hurt those who are just learning to get online.  Likewise, what works to improve some outcomes can hurt other outcomes. For instance, some programs (e.g. requiring students to attend online class following the same school class schedule) may help students have a sense of attending school regularly but may hurt the visions or increase their anxiety.

In summary, it is important recognize that online education, just like face to face education, has a wide array of different practices and an equally broad array of potentially competing goals. It also serves student population with variations in ability, interest, personality, home environment, and access to technology. There cannot be one online program or practice that serves all students all goals equally well. When we are looking for something that works, we should always keep in mind: What Works May Hurt.

Featured image by Chuck Underwood on Pixabay