Guiding Principles Of Scientific Thinking: Correlation Vs. Causation
As defined in class (Huggon, 2018), correlation is when two variables are associated with each other, whereas causation is when one variable causes another. Often times, when two variables correlate, people jump to the conclusion that one variable must cause the other. However, they may have no causation whatsoever. In the scenario, they have identified that dairy products and intensity of autism have a correlation. From there, they jumped to the conclusion that dairy products make autism worse. This conclusion is a fallacy because the logic is incorrect. It would be a lot more logical if they had taken into consideration a possible third variable that causes both the variables. Ultimately, the best option would have been to find evidence that proves causation, rather than relying on correlation to prove it.Replicability
According to Lilienfeld et al. (2017), replicability refers to when a study’s findings can be duplicated consistently. We shouldn’t put too much trust in a psychological finding until it has been replicated. Upon evaluating the claim of autism becoming worse from dairy products, I asked myself whether other scholars or independent investigators have replicated the findings that support this claim. The answer is that they haven’t, which increases the odds that the original finding was due to chance. This means that the findings might be a one time only fluke, as it can’t be backed up with other studies. If it had been replicated by other scholars or independent labs, it would help disprove the chance that it was a fluke, and help prove that the study was reliable.Parsimony (a.k.a Occam’s razor)
According to Lilienfeld et al. (2017), the Occam’s razor rule says that when two explanations account equally well for a situation, we should generally select the more simple one of the two. In this scenario, their explanation for intense autism is dairy products. However, a simpler explanation that fits the data just as well could be something along the lines of dairy products being unhealthy in general. Blaming the intensity of autism and autism spectrum disorder merely on dairy products seems quite complicated and far-fetched. By shaving off needlessly complicated explanations and arriving at a simple one that fits the data just as well, it does a better job of accounting for the evidence.Warning Signs of PseudoscienceLack of self-correction
As defined in class (Huggon, 2018), lack of self-correction is when an idea remains unaltered, despite the fact that contrary evidence has been presented. It is not appropriate to lack self-correction because it is unethical to spread misinformation on purpose. In the scenario, the original claim is that dairy products worsen autism. However, contrary evidence is provided. The University of Texas examined 14 other studies that investigated the effects of dairy and dairy-free diets on people with autism. They found that the original claim’s overall study quality was poor and only the least scientific studies said that the diet improved behaviour. Completely disregarding this contrary evidence, the author sticks with the original claim. Had they taken the contrary evidence into consideration and altered their claim, they would be more reputable and reliable.
