Can children learn to read via fully online instruction? Not long ago, this question was mainly a scientific curiosity or the stuff of sometimes-rancorous political debate. With the coming of Covid-19 and extended stretches of widespread remote learning, it took on an urgency that demanded an answer. According to new research published in Frontiers in Human Neuroscience, the answer is yes—but with a number of important caveats.
In 2019, University of Washington (UW) professor Jason Yeatman created a two-week in-person summer program that was part reading boot camp for five-year-olds and part neuroscientific research project to assess the teaching method being used. While the method was touted as groundbreaking science, it just seems like good old fashioned phonics with a side order of brain imaging. It also made use of small student-to-teacher ratios, highly qualified and high-energy instructors, manipulable materials to reinforce instruction, and a strong social aspect to bond the young participants. Sessions alternated between full-group and smaller breakout sessions during the day. Think Sesame Street live. Initial results were promising, but hopes for another year of data were dashed in 2020 due to the pandemic. Yeatman and his team quickly adapted—and while no one got their skull scanned via Zoom, the researchers were able to morph their teaching method into a fully online version. Even the social aspects and session structures.
Participants were recruited from UW’s existing group of voluntary test subjects and had to be between five years old and five years and four months old, have no diagnosed neurological or sensory conditions, speak English as the primary language at home, and have demonstrated no ability to read prior to the start of the new online reading camp. Phonological awareness and letter-sound knowledge, the two best predictors of reading acquisition during the first two years in school as concluded by a 2020 meta-analysis from the National Institute of Child Health and Human Development, were the exclusive focus of the camp’s teaching method. There were 116 children participating in the study, with eighty-three receiving the two-week treatment and thirty-three receiving no reading instruction, having gone through the full recruitment process but ultimately declining to participate in the camp. This does raise questions about why families would choose not to participate—children staying with relatives, kids attending daycare/preschool elsewhere, etc.—and whether those reasons could have impacted the outcome, but no data were collected on that.
The two groups did not differ significantly on observable measures, however. For both groups, the average parental education in years was roughly equivalent to a four-year college degree, with a wide range extending from elementary- to postgraduate-level degree completion. Socioeconomic status was measured via income-to-need ratio. Both groups included families at or below the federal poverty line, as well as families into the upper levels of wealth. Each group was approximately half boys and half girls. No racial or ethnic breakdown was provided.
Treatment children received a kit prior to the camp, which included workbooks and manipulatives like stacking blocks and modeling clay. High speed Wi-Fi and enabled devices were provided to any participants who did not already have them at home. Sessions were administered by three teachers with a bachelor’s degree in either education, linguistics, or speech and hearing sciences and who had prior experience teaching English to young children in previous lab projects. They received two weeks of training prior to the start of camp. The camp was conducted in fourteen waves of six participants each over the period of fall 2020 to summer 2021. It is interesting to note that the children were receiving “various and constantly changing” levels of formal schooling during this time, due to the pandemic. The two-week camp was held five days per week for 2.5 hours per day via Zoom. Two teachers taught each daily session, with each teacher primarily instructing three children every day, and those small groups were mixed and counterbalanced on different days so that each child became familiar with all the others. Each day included at least one full-group activity. Parents and other caregivers were instructed to assist their children with logging in and preparing the daily materials to be available. They were instructed to stay within earshot in case their children needed assistance but were told not to provide or prompt answers and to keep siblings away as much as possible. There were some cases of parental overinvolvement, which resulted in email and phone-call reminders from instructors of the proper level of interaction. No specific guidelines were provided to the families regarding additional reading-related activities outside the camp; however, they were asked to report any such activities in a post-camp survey.
Achievement and progress were measured via two pretests and two posttests (one standardized, the other not) on skills related to reading and the effect of the training program on those skills. For the control group, the pre- and postsessions took place two to three weeks apart to match the timeline of the treatment group.
While treatment and control group students did not differ significantly in their baseline pretest scores, treatment group students showed significant boosts on the posttests in the specific skills that were taught in the online sessions. These include phonological awareness and lowercase letters’ names and sounds. Interestingly, the data show increases in some reading skills that were not taught directly—such as uppercase-letter sounds and pseudoword decoding—among treatment-group students. Control group students did show small but not-significant improvement on uppercase-letter sounds, lowercase-letter sounds, rapid automatic naming, and expressive vocabulary. No correlations were found between these improvements and parental education level, and only two skill increases were found to correlate with socioeconomic status.
It feels a little odd after more than two years of pandemic disruption to read a research paper whose findings suggest that maybe we can make this online learning thing work after all—especially when members of the research team go on to suggest that it may even work for kids who take online classes by choice. Quelle surprise. This model is live, small group, super interactive, conducted by highly trained teachers. In this limited form, it resembles high-dosage tutoring more than it does “school.” More and varied research on what makes for successful online teaching for more children is required, but we’ve got a lot more data available than we’ve ever had before, as well as many examples of how it should not be done. This is at least one model upon which we can focus to figure out once and for all how do it right for anyone who wants it for any reason.
SOURCE: Yael Weiss, et. al., “Can an Online Reading Camp Teach 5-Year-Old Children to Read?” Frontiers in Human Neuroscience (March 2022).
