From 2015... Reductionist evidence in science debate by laypersons is viewed as more explanatory & conclusive than comparable evidence from macrolevel processes; the preference for reductionist information does not go away with education

From 2015... The Influence of Reductionist Information on Perceptions of Scientific Validity. Rebecca Rhodes. PhD Thesis, Michigan Univ., Psychology, 2015. https://deepblue.lib.umich.edu/bitstream/handle/2027.42/111352/rerhodes_1.pdf?sequence=1&isAllowed=y

ABSTRACT
The ability to reason scientifically about evidence is an important skill for many everyday decisions, ranging from whom to choose in the next presidential election to whether or not to immunize your child. Evidence for the importance of scientific reasoning can be found in continued attempts to improve the teaching of reasoning skills within the educational system. However, in order to develop effective strategies for reasoning about scientific evidence, it is important to understand lay conceptions about what it means for something to be “scientific”. In the present work, I examine the influence that reductionist evidence – that is, evidence that comes from micro-level processes, such as biological or neurological processes – has on perceptions of scientific validity. Across eight experiments, I demonstrate that reductionist evidence tends to be viewed as more explanatory and more conclusive than comparable evidence from macrolevel processes, such as psychological processes. Interestingly, the preference for reductionist information does not go away with education. In fact, people with greater scientific literacy are even more likely to assume that reductionist information is superior to macro-level information. I interpret this finding as evidence that the preference for reductionist information is not irrational, but instead an expected consequence of traditional science curricula. I demonstrate several important implications of reductionist preference. For example, this preference increases the likelihood of making causal inferences from the results of research studies that suggest micro-level – as opposed to macro-level – mechanisms, and it can decrease the size of the sample one needs to feel confident in accepting conclusions from these studies. I relate these findings to current pervasive issues in the scientific community, such as publication biases and the prevalence of underpowered studies utilizing reductionist approaches. I also discuss educational strategies that could encourage holistic thinking about science – specifically, emphasizing science as a tool for thinking strategically about everyday phenomena, regardless of the level of analysis, rather than a collection of discrete facts obtained by the use of technology and equipment.