The results of this experiment show that students have more trouble reasoning about the loss of traits than the gain of traits. They also have greater difficulty reasoning about the loss of traits between species than within species. But in reasoning about the gain of traits, there is no difference between the interspecific and intraspecific situations. Similarly, students have more misconceptions about the loss and gain of traits between species than within species. The hardest problem for students to solve, said Nehm, is the loss of traits between species. “If students can handle that, that’s the highest level of competency. But do we ask those questions? No.”

Also, students use more key concepts in solving problems involving familiar animals than unfamiliar animals, but this trend is not seen for plants, all of which seem to strike students as unfamiliar.

These surface features have a remarkably powerful influence, said Nehm. “If you want to show your class is doing great, I can design an assessment for you. If you want to show your students are failing, I can design an assessment for you. All I have to do is manipulate surface features because students’ reasoning is so tied to these features. And yet we pay no attention to this in any textbook or in any assessment.”

The bottom line is that “surface features matter, and we need to be more precise in our instructional strategies to deal with these.” Because misconceptions are surface-feature specific, instructional examples must be carefully chosen. Furthermore, assessments of competency must include authentic production tasks, such as explaining how evolutionary change occurs, not just fragmented knowledge selection tasks.


In 2007, Nehm reported on an introductory biology course that was changed so that every topic included evolution, while a parallel course was taught using a traditional curriculum (Nehm and Reilly, 2007). The outcomes were not substantially different. “It’s an awful downer at this conference,” he admitted.

However, one single study is not enough to draw broad conclusions. For one thing, students have difficulty learning evolution, so teaching it in the same way is probably not going to lead to progress. “If you have a problem with A and you give lots more A, the chances are it’s not going to lead to a substantial improvement.”

Also, as students work through the biology curriculum, they move from naïve models to mixed models to scientific models, but progress is very slow—25 percent of students who have completed a course on evolution and additional coursework still used mixed models.

Determining the conditions under which students can effectively

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement