The relative importance of religion and education on university students’ views of evolution in the Deep South and state science standards across the United States

Two thousand nine hundred and ninety-nine college students were all enrolled at a large public university and exposed to higher education. Yet, when Rissler, Duncan, and Caruso ran their numbers, religiosity predicted whether those students accepted evolution better than anything their teachers had done. If education isn't the lever we thought it was, what does that mean for science classrooms? That question sits at the heart of a long-running debate. Across wealthy democracies, the United States stands out for its low public acceptance of evolution, and researchers have disagreed about why. Is it weak science education or strong religious identity? The answer shapes everything from how biology teachers approach their lesson plans to how state governments write curriculum standards. Rissler and colleagues decided to test both explanations simultaneously, and the answer they got was unambiguous. First, a distinction worth keeping in mind. Understanding evolution means knowing what the theory says — the facts, the mechanisms, the evidence. Accepting evolution means believing it's true. Those are not the same thing. A student can pass an evolution exam and still personally reject the theory. That gap between knowing and believing turns out to be exactly where this story gets interesting. The setting mattered. The Deep South is among the most religious regions of the country, and the survey data from this university bore that out. Seventy-nine percent of respondents considered themselves religious, seventy-six percent of those identified as Christian, and thirty-seven percent endorsed a young-earth creationist view — the position that the Earth is less than ten thousand years old. Forty percent accepted evolution only with the caveat that God initiated the process. These figures align with decades of Gallup polling. This wasn't a fringe sample. It was a portrait of a real student body. To measure evolution acceptance, the researchers used the Measure of Acceptance of the Theory of Evolution, or MATE, a validated twenty-item instrument. Factor analysis confirmed that a single underlying factor explained sixty-two percent of the variance across items, and Cronbach's alpha exceeded 0.96, meaning the instrument was highly internally consistent. They also included two knowledge items from the Knowledge of Evolution Exam. The core analytical tool was a structural equation model, or SEM — a statistical method that lets you estimate the simultaneous influence of multiple predictors on an outcome, in this case, a latent variable they called "evolution understanding," built from both MATE and knowledge scores. The SEM included five predictors: religious identity, religiosity measured by service attendance, academic level, high school science experience, and college major. Model fit was strong across every standard index. The results were not close. Religiosity carried a standardized path coefficient of negative 0.39 to evolution understanding, meaning more frequent religious attendance predicted substantially lower understanding. Religious identity as Christian carried a coefficient of negative 0.32. Both were statistically decisive, with z-scores above twenty-two. Now compare that to the education-side predictors: academic level came in at positive 0.033, college major at positive 0.095, and high school experience at positive 0.12. The religious predictors were roughly three to four times the size of the education predictors. Rissler and colleagues put it plainly: the relative importance of either religion variable was greater than all the education variables combined. The difference in MATE scores by religiosity made this concrete. Students who identified as not religious averaged ninety point ninety-three out of one hundred on the MATE. Religious students averaged sixty-two point ninety. That's a twenty-eight point gap — nearly three full scoring categories on the instrument's scale. And critically, when the survey asked separately what students believed versus what they knew the scientific community believed, the gap widened in a telling way. Questions about the scientific community's position shifted relatively little across religiosity levels. Questions about the student's own belief shifted dramatically. Highly religious students often knew that scientists accept evolution. They simply rejected it for themselves. That's not a knowledge failure. It's something else. The pattern held within every subgroup. Science majors outperformed non-science majors at every academic level. But within science majors, religiosity still dominated: those who attended religious services weekly averaged fifty-eight point seventy-six on the MATE; those who seldom or never attended averaged eighty-four point ninety-three. Non-science majors showed the same gap — fifty-two point thirty-four for weekly attenders and eighty-one point thirteen for seldom or never. High school instruction mattered too: students taught only evolution had higher acceptance than those taught both evolution and creationism or creationism alone. But high school experience primarily moved the knowledge needle. Acceptance among highly religious students remained low regardless. So, what happens when you put those students in a college biology classroom? Rissler and colleagues ran pre- and post-course surveys in three undergraduate biology courses — four hundred eighty-seven matched surveys in total. Two of the three courses produced statistically significant increases in evolution acceptance by the end of the semester. That sounds like progress. But the disaggregated picture tells a more complicated story. When the team separated students by religiosity, the gains collapsed to a single group. Only students who seldom or never attended religious services improved significantly after the course, with a chi-square of nine point eighteen and a p-value of 0.002. Students who attended services weekly or nearly weekly showed no significant change. The classroom worked, but only for the students who were least resistant to begin with. Non-majors introductory biology MATE scores rose from sixty-two point twenty-seven to seventy-four point fifty-eight, and honors biology rose from sixty-nine point forty-eight to eighty-four point sixty-nine. The human anatomy and physiology course showed no change at all. Two out of three courses moved the group average. None of the courses moved highly religious students. That finding from individual students maps almost perfectly onto the national landscape. Rissler and colleagues pulled state-level data on K-12 science standard quality, grades assigned by the National Center for Science Education, and correlated them with measures of state religiosity and educational attainment. The associations were significant and consistent. The percent of a state's population identifying as evangelical was the strongest predictor of weaker standards, with a correlation of negative 0.43 and a p-value of 0.002. The percent who said religion was "very important" to their lives correlated at negative 0.33. Meanwhile, states with higher educational attainment had stronger science standards: the correlation with percent holding bachelor's degrees was positive 0.37. The lowest-graded states — Alabama, Louisiana, Oklahoma, Texas, and West Virginia — are among the most religious in the country. The structural irony is hard to miss. The pipeline that would produce students with stronger science foundations is weakest in exactly the places where religiosity is strongest. Students arrive at university having learned less evolution in high school, and once there, the ones most resistant to change are the ones least likely to change. The system compounds the problem at every stage. Where does that leave science educators? Rissler and colleagues are careful here. They don't pretend there's an easy fix, and they don't endorse strategies that paper over real scientific conflicts — the age of the Earth, for instance, is not a matter of interpretation. What they do argue is that rigorous science instruction and genuinely high-quality K-12 standards are non-negotiable. Strong standards legitimize teachers. They protect evolution from being quietly marginalized under community or administrative pressure. And they shape the students who eventually sit in college biology courses. The harder truth the paper leaves us with is that education operates inside a cultural context it did not create and cannot fully overcome. Religion and religiosity, in this study, far outweighed every educational variable the researchers could measure. A biology class can move a student who is open to being moved. It cannot, on its own, dissolve a deeply held identity. That isn't a reason to stop teaching. It is a reason to be honest about what teaching can and cannot accomplish and to fight hard for the structural conditions that make good teaching possible in the first place. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.