Science Literacy and Belief in the Paranormal-An Empirical Test
Ryan, Travis J., Brown, Jessica, Johnson, Angela, Sanberg, Colin, Schildmier, Megan, Skeptic (Altadena, CA)
A STUDY PUBLISHED IN SKEPTIC (Vol. 9, No. 3) by Walker et al. investigated the relationship (or lack thereof) between scientific literacy and pseudoscientific belief. The authors found no significant relationship between college students' performances on a test of basic science knowledge and their belief in superstitious, pseudoscientific, and non-scientific ideas. They did not, however, account for potential confounding factors in their research design, such as science background, level of education, or experience in science. Towards the end of their article, Walker et al. were equivocal as to whether a scientific background was sufficient to "ward off [pseudoscientific] belief." As a part of a course on Science, Pseudoscience, and Superstition, we designed and carried out a follow-up to the Walker et al. study.
We were interested in the same question as Walker et al., namely: does scientific literacy lead to increased skepticism of pseudoscientific beliefs? To that end, we repeated major elements of the Walker et al. study. We randomly pulled 10 questions from the Praxis Series National Series Exam in order to gauge basic scientific literacy. These questions come from the same source as those in the Walker et al. study, and though some questions on the exam may lack absolute accuracy (see SKEPTIC Vol. 10, No. 1), they do likely reflect a basic scientific literacy (hereafter, SL).
The questions in the SL portion of the test were multiple choice, each with four to six responses to choose from for each item. Respondents also completed a survey where they were asked to rank their belief in pseudoscientific ideas. We did not use the same questions as published in the Walker et al. study, but they were of a similar content and demeanor (e.g., "Some people can move objects with their minds" and "Some injuries can be cured by placing magnets on the skin near injured areas"). We also presented the respondents with 15 statements, and asked them to rate their agreement on a scale of 1 ("disagree completely") to 5 ("agree completely"); a response of 3 indicated respondents neither agreed nor disagreed with the statement. For each respondent, we determined the number of correct responses on the SL test and the average score on the pseudoscientific belief (hereafter, PB) survey.
The only major difference in our study from that of Walker et al. is that we asked respondents to identify their academic major and class standing (freshman, sophomore, junior, senior) in order to determine whether these factors (i.e., primary area of academic interest and amount of training or experience) influenced the relationship between SL and PB. We then grouped students into four classes: freshman science majors, freshman non-science majors, senior science majors, and senior non-science majors. In total, 144 students completed the SL test and PB survey. We used two-way analysis of variance with academic major (science or non-science) and class standing (freshman or senior) to detect differences in SL and PB. We also used linear regression to determine whether SL and PB are significantly related. We expected to me an inverse relationship between SL and PB scores if basic scientific literacy also teaches critical thinking and wards off pseudoscientific belief.
We found SL scores to be significantly higher among science majors ([F.sub.1,142] = 54.63; p < 0.0001). The scores were not significantly different between freshmen and seniors ([F.sub.1,142] = 0.05; p = 0.822), but there was a significant interaction between major and class standing, with SL scores increasing with class standing among science majors and decreasing with class standing among the non-majors ([F. …