FALLING BEHIND IN THE SOUTH: COMPARING EVOLUTION UNDERSTANDING OF THE SOUTHEAST TO THE REST OF THE USA

In order to improve education of the theory of evolution in the US, educators and researchers must better constrain current gaps in and misconceptions of the public’s understanding of evolution. However, there have only been a few studies assessing the US public’s understanding of evolution. States in the southeastern (SE) US are of particular interest because they have higher rates of conservatism, which correlates with lower evolution acceptance. In addition, SE states historically have had more lax evolution education standards. Here, I compare evolution understanding between SE US states with the rest of the US.

Using Amazon Mechanical Turk (MTurk), a survey of eight multiple choice evolution understanding items was administered to 305 participants, 63 from SE states and 242 from non-SE states. MTurk is an online survey tool wherein registered MTurk workers complete tasks for pay. The demographics of MTurk participants are slightly more diverse than an average sample of Internet users and significantly more diverse than typical American college samples.

A factor analysis resulted in six of the eight items being valid within one construct (factor loadings ≥ 0.27). Participants from non-SE states answered a significantly greater number of items correctly than participants from SE states (3.5 vs. 3.0; p = 0.01), with no differential item functioning (DIF). In addition, a greater proportion of non-SE participants answered each of the six individual items correctly. The item resulting in the most drastic difference between the two groups was, “Which of the following processes is the primary mechanism responsible for the evolution of the many varieties of domestic dog alive today?” The correct response was “Artificial selection” with a primary distractor of “Natural selection”. Fifty percent of participants in the non-SE states answered this item correctly, but only 33% in SE states answered correctly. Based on linear regression models, religiosity and level of education had small effects on the difference in scores between the two populations. No effects were found due to gender, ethnicity, income, or age. These data suggest that improving the US public’s understanding of evolution goes beyond overcoming socioeconomic backgrounds, and may lie in improving the education in certain portions of the US.