The American Biology Teacher

are present in key variables, including standard deviations of 34.72 points for content knowledge, 29.46 points for acceptance of evolution, and 21.53 points for understanding of NOS.

Discussion

Regarding some variables, the demographics in this sample closely align to those of the larger population of teachers in the United States as well as the state where the study took place. In the sample, AP teachers appear to be well trained, highly experienced, and aptly certified in their fields, which suggests a comfort level and background that should support greater understanding and acceptance. However, in a 2009 study of science teachers as they taught evolution, Goldston and Kyzer (2009) observed changes in teacher persona during the teaching of evolution that suggest that discomfort and misunderstanding transcended years of teaching experience and background, which suggests that traditional indicators are not necessarily applicable when it comes to controversial topics (Sinatra et al., 2003). In the present study, the majority of participating teachers had a wealth of experience in the classroom – almost 90% having ≥6 years of experience and 60% having ≥11 years of experience – yet the levels of understanding and acceptance were still much lower than desired, aligning to results suggesting that experience is not an indicator of accuracy of teaching evolution (Nadelson & Nadelson, 2010). As such, it was not only in content that we saw breaks in logic between our expectations and the reality of the situation. In fact, even with lengthy teaching experience, controversial topics can override logical expectations (Sinatra et al., 2003; BouJaoude et al., 2011).

Ron Johnson (1986) noted in an early acceptance study that
biology teacher performance was more in keeping with acceptance
and understandings of evolution held by nonmajors in biology
courses. Furthermore, Johnson (1986) characterized biology majors
as being more open in their thinking and their willingness to con-
sider evolution than nonmajor students. Decades later, studies indi-
cate similar results despite the implementation of standards to align
practices (Rutledge & Mitchell, 2002; Bowman, 2008; Glaze et al.,
2015). It stands to reason that teachers of science should be more
open to scientific explanations; however, that is not always the
case, and frequently teachers exhibit a willingness to divest them-
selves of the responsibility to accurately teach controversial topics
in science (Bowman, 2008; Goldston & Kyzer, 2009).

In terms of the key measures deployed in the present study, participant teacher scores somewhat paralleled those of other science teacher samples but were notably lower in others (Rutledge & Warden, 2000; Rutledge & Mitchell, 2002; Nadelson & Sinatra, 2009; Nehm et al., 2009). Overall, acceptance of evolution in this study was moderate, with participants scoring an average of 68/100 on the MATE. This is noticeably lower than that of other teacher samples in the United States, where teacher acceptance means were documented at 77.59 in Indiana (Rutledge & Warden, 2000), 85.9 in Oregon (Trani, 2004), and 87 in Ohio (Korte, 2003). Compared with moderate to high levels of acceptance, understanding of NOS in our study was on the borderline between very low and low at a mean of 52.72/100. This is only slightly lower than what has been reported in Indiana teachers, whose average on the same NOS instrument was 59.49, also in the low range (Rutledge & Warden, 2000). Despite higher occurrence of graduate-level training and advanced certification in the sample, the level of content knowledge demonstrated among AP teachers in our study was also low, with a mean score of 56%. This is a good measure lower than the content knowledge reported on the same instrument in Indiana teachers, whose average score was 71% ( Rutledge & Warden, 2000).

Each of the evolution measures in this study demonstrated a range of variability among responses, with standard deviations ranging from 21 to 35 points. The greatest variability was found in content knowledge and the least in NOS understanding. It is not uncommon in studies of evolution understanding and acceptance for there to be higher values for standard deviation as a result of both high and low outliers that must be addressed prior to statistical analysis (Trani, 2004; Glaze et al., 2015). This is often explained in terms of content knowledge by the variability in required coursework in teacher education, where courses specific to evolution are often not a part of the plan of study (Glaze & Goldston, 2015). Furthermore, existing tensions in the southeastern United States around evolution often result in teachers and students actively avoiding discussions of the subject, resulting in future teachers coming into their program with existing gaps that are not sufficiently addressed in survey courses (Goldston & Kyzer, 2009). Outlying cases in the area of acceptance are also common, as previous studies have demonstrated that there is very little correlation between actual knowledge of evolution and acceptance, which is often influenced by extrinsic and intrinsic factors (Glaze et al., 2015). It is not uncommon for a person to be wholly accepting of evolution but to reflect very little actual comprehension and understanding of the processes of evolution. Conversely, it is also possible for an individual to be highly knowledgeable of the processes and facts surrounding evolutionary theory, but to actively elect rejection of those parts that are not in alignment with their worldview (Glaze et al., 2015).