Butler, R., & Nisan, M. (1986). Effects of no feedback, task-related comments, and grades on intrinsic motivation and performance. Journal of Educational Psychology, 78(3), 210-216. doi:http://dx.doi.org.cmich.idm.oclc.org/10.1037/0022-0663.78.3.210
You know the reaction people have to the sound of microphone feedback — cringe, shrink, scream? Reading the results of this study may cause the same reaction when there is NO feedback. Butler and Nisan provided definitive answers about the importance of feedback in an educational situation through their research. Studying sixth grade students in Israel in a school setting with a middle-class background, their experiment manipulated the level and type of feedback students received on work that they completed. Some students received written comments with one positive and one negative item. One group received only a letter grade. The third group received no feedback at all. Children engaged in three sessions of activity completion, with the first on one day and the second and third sessions several days later, but on the same day at different times. Findings demonstrated some logical results, but the degree was both statistically significant and rather astounding. The survey administered afterward showed that students who received feedback attributed results to effort and interest, whereas the grades group suggested it was their own mood and the group with no feedback or grades stated that it was dependent upon the examiner’s mood.
Students who received comments on their work demonstrated the following improvements: 68% more short words, 32% more long words, 43% higher scores on Task A and 31% on Task B, increases in fluency (17%), flexibility (22%), elaboration (70%), and originality (58%) measures. Students who received grades on their work demonstrated the following: 71% more short words, 8% more long words, 32% higher Task A scores, but decreased in fluency (7%), flexibility (52%), elaboration (58%), originality (59%), and Task B scores (32%). Students who got no feedback or grade on their work produced 42% more short words, but declined in every other measure on both tasks: 47% fewer long words, 26% lower scores on Task A, 21% less fluency, 46% less flexibility, 53% less elaboration, 69% less originality, and 38% lower Task B scores. Care was taken to analyze the initial results based on gender, but differences were not statistically significant, so the results were aggregated for presentation. The construct of the investigation was sound, keeping all groups exactly the same except for the variable of the type of feedback. If replicated, one possible area for adjustment would be that the tasks were expressed as a unique request for a special project, whereas the observation of standard classroom examples with a regular teacher would be beneficial. If any fault is to be found, it would be in the discussion of the results not being presented as clearly and dramatically as possible.
As a classroom teacher, I must share that the basic findings did not shock me, knowing that students (as well as my debaters in competition) prefer specific feedback for their work over the numbers which seem cold and arbitrary. It was the magnitude of difference which chastised me for every time students did work without a feedback mechanism. For students to demonstrate 69% less originality in response when they could be producing with 58% greater originality is a conviction of practice. I never thought that a few words from me or the lack of them could cause a 127-percent pendulum swing! Honestly, the day after reading the study (which was published more than thirty years ago and was never mentioned in any education course, professional development session, or conference I attended), I shifted practices in my own classroom to assure that students received feedback for their work at all times, some graded and some not, but always with commentary.
Teacher Takeaways
Students perform significantly better and internalize their success when provided with feedback in the form of comments over grades or none at all.