When asked, "Human beings, as we know them today, developed from earlier species of animals", only 43% of randomly selected Americans answered "True" (National Science Board 2006). Furthermore, no more than 48% of the public could choose an accurate definition of evolution (People for the American Way Foundation 2000).
The woeful state of evolution comprehension is not surprising considering that formal education ends in high school for most of the American population. The Current Population Survey estimates that 28% had attained a bachelor's or higher degree (National Center for Education Statistics 2008). Although some high school graduates may have attended college without matriculating, there are no indications of how many of these students completed a college science course, particularly in the field of biology. Therefore, high school teachers provide the last formal instruction most students will receive in evolution. There have been many assessments of teachers' religious beliefs and the acceptance and teaching of evolution (see, for example, the University of Minnesota Biology Program faculty publications webpage at http://www.cbs.umn.edu/bioprog/publications for a compendium of Randy Moore's work). However, there have been very few measures of their conceptions of evolution and none attempt to determine the source of misconceptions.
Therefore, James E Platt (of the University of Denver) and I developed the Classroom Test of Evolution (CTER) aimed at learning more about secondary teachers' understanding of evolution. It is based upon recent research on evolutionary misconceptions (for instance, see Brian Alters  for an excellent treatment on general misconceptions; Mier and others  for problems reading phylogenetic trees; and the NCSE website at http://ncse.com/evolution/education/problem-conceptsevolution for Eugenie Scott's discussion regarding the terms "cause, purpose, design, and chance"). The assessment was reviewed by evolutionary experts for content.
The CTER consists of 26 multiple-choice items designed to measure an understanding of the following aspects of evolution:
- natural selection
- acquired characteristics
- vestigial structures
- convergent evolution
- reading phylogenetic trees
- common ancestry
- transitional fossils
- and the nature of evolution.
Almost all of the items are paired in the style of Lawson (1994): questions are sequenced in such a way that the answer to the second question identifies the reasoning behind the response to the first question. For example, in one set of questions, the first asks the respondent to identify which of a group of organisms is most closely related to lungfishes and the second inquires why biologists accept this to be true. Respondents must answer both questions correctly in order to receive credit.
The test was administered anonymously online. The respondents also provide demographic information, such as location and educational background. Over 700 high school biology instructors nationwide completed the survey. The majority were from the western or midwestern public schools, although the schools were roughly equally distributed between urban, suburban, small city, and rural locations.
Out of a possible 13 points, the mean score was just a 6.35. Yet unlike the American public, most teachers could correctly identify the evolutionary relationship of humans to other, related species. They performed well on questions concerning acquired characteristics and vestigial structures,as well.
However, several items asking teachers to interpret phylogenetic trees received no better than a 52% correct response rate, and for one set of questions, this rate dropped to a dismal 10.5%. Just under 35% were able to identify homologous characteristics in two organisms as being derived from biological features in their most recent common ancestor, and a mere 17.3% were able to define transitional fossils. Other problematic areas were discerning the effects of mutations on the genome and predicting the results of a selection event on a population.
Rutledge and Mitchell (2002) show that student knowledge about and understanding of evolution are affected by that of their instructors and our analysis of CTER data reveals that teachers possess many misconceptions. We continue to analyze these data with an eye toward developing science education curriculum to help correct teachers' misunderstandings. In the meantime, we need to be aware that students may be graduating high school with the baggage of their teachers' misconceptions.