Article Review: Pretesting Enhances Learning in the Classroom

I am a teacher that absolutely loves to read research articles focusing on education. I just love it. And nothing spikes my inner nerd more than a well-written article studying a learning strategy that is directly applicable to my classroom and the students I teach. I was recently made aware of such an article (thank you, Brad Busch) by Drs. Nicholas Soderstrom and Elizabeth Bjork.The article, Pretesting Enhances Learning in the Classroom, takes a look at pretesting and its implication on learning. I have previously written about the pretesting effect here. 

One reason I love this article, which actually has little to nothing to do with learning in my classroom, is its accessibility for teachers. The average classroom teacher is not trained in reading somewhat complex journal articles. Every once in a while, I will find an article that is written in such a manner that teachers can read and use the findings of the study in their classroom or in a professional development setting. I can easily imagine a group of teachers reading this article and discussing it in a ‘book talk’ kind of manner.  From start to finish, the article is easy to understand and comprehend as a teacher, when considering how the research may or may not relate to the classroom. Love it. 

To all of the writers of education researchers who are considering teachers as a portion of their audience, I say a massive thank you. There may only be a small percentage of teachers directly interested in your writing, but we care a lot and appreciate when you write in a manner that is more easily understandable for the untrained eye.

Now, let’s get down with the get down. What exactly did this research take a look at? What are the findings? What are the implications for the classroom teacher?

So, in a 10 week college course that met once a week, students were pretested on items before lectures in weeks two, four, and eight. The pretests were given at the beginning of class and asked about information that had not been covered in class prior to that class meeting’s lecture. The pretest asked four multiple-choice questions with five possible answers (A-E). Interestingly, the lecturing professor was not present for the administration of the pretest and had no idea what questions were to be asked during the pretest. Also, there was no feedback for the students. They had no idea what they answered correctly/incorrectly, and there was never a time where the specific pretested items were discussed. 

On a final, cumulative forty question test at the end of the term, there were two types of questions based upon the pretests; six questions that were identical to a question appearing on one of the pretests and six questions that were related to a question asked on the pretest. Also, there were six control questions that appeared on the test. These eighteen questions were randomly mixed with twenty-two other questions that had not appeared on any previous unit tests.

Results? 

1. Understandably, students performed poorly on the pretests, with just under 35% of the questions answered correctly.
2. There was no significant difference between identical and related questions on the final exam (.86 vs. .85).
3. There was a significant difference between identical and control questions (.86 vs. .77).

So, from the standpoint of academic performance on the final exam based on pretested vs. not pretested (control) items…the pretest appears to show significantly more learning, almost a 10% increase. That’s pretty big. Also, interestingly enough, the items that were related to the pretested items also showed a significant increase. I mean, from the standpoint of the classroom teacher, that’s enough to convince me of the legitimacy of pretesting in my classroom. 

But, there’s another aspect of this study that is also quite interesting and, in my eyes, further justifies the use of pretesting in the classroom. The researchers also conducted a six item questionnaire with the students in the course. You can see the questions and the results of the questionnaire below:

After viewing the questions and results, here are a few items that stood out to either myself or the researchers:

1. Students took the pretests quite seriously, with student average response at 5.16 out of 7.0.
2. Students overestimated how many questions they believed they answered correctly on the pretests.
3. From questionnaire item 3, only 14 % of students guessed while answering while a much higher percentage either used a process of elimination (76%) or chose the most recognizable answer (41%), which is the availability heuristic. 
4. From questionnaire item 4, 50% of students didn’t consciously consider the pretest items any further after the pretest, while the remaining 50% of students thought about, discussed, looked up, and/or made a concerted effort to study the pretest items later. 
5. From questionnaire item 5, students appeared to be aware of the pretest items when the professor discussed them during the ensuing lectures. 
6. From questionnaire item 6, most students did not pretest themselves in other courses. 

So, what do the results from the questionnaire tell me as a classroom teacher? The big standout for me is how much the pretest items seemed to steer student attention during the lecture. It was as if the pretest items stood out as important guideposts during the lecture and the students maybe saw this material as being important and, therefore, paid more attention to that information. And then, when students studied later, they again saw this content as being particularly important because it was pretested and subsequently mentioned during the lecture. I am speculating here, as the present research did not test this. 

In addition, it appears students who took the pretests seriously then used that information to self-regulate their studies. If they acknowledged they saw/heard the professor discuss the items after already seeing them during the pretest, they may have adjusted what/how they studied to ensure they covered the pretest material again. This is somewhat evident, as students indicated on the questionnaire they looked up the answers to the pretest after they were administered, made a particular effort to study the items found on the pretest, and/or talked with other students about the items after the pretest.

Another rather interesting aspect of this study is the lack of feedback after the pretest. My intuition tells me that I should give students feedback after the pretest; going over the material and discussing answers. But, perhaps by not providing feedback, students are more curious to find out during the lecture whether their estimations are correct or not? Maybe this impacts attention during the lectures also? As a teacher, I want to tinker a bit with this and maybe provide students with a pretest in precisely the same manner as the researchers did in this study, but also tell my students there will be a posttest after the lecture that will ask the same questions and that posttest will, hopefully, show growth in knowledge and understanding of the material. Or, better yet, maybe the posttest won’t be until the next class meeting to allow for student forgetting and the spacing effect. Plus, I can see where pretesting can lead to some really nice classroom discussions of the material to be introduced during class. There are several directions I could go with this, and I’m just brainstorming ideas right now. 

Well, what’s the bottom line? What does this research tell me as a classroom teacher? It tells me I need to incorporate more pretesting in my classes. How rare is it to find a classroom strategy that (1) increases scores on assessments, (2) appears to focus student attention during class, (3) seems to improve student self-regulated study strategies, and (4) is so easily applicable to a wide range of material and student ability levels? It’s a win-win-win…win. Why in the world would you not want to include pretesting in your classroom?

---

Feature image by Engin Akyurt: https://www.pexels.com/photo/light-bulb-beside-books-on-shelf-2767814/