Replication is a central aspect of doing science. When an experiment is completed we analyze the results to determine significance. For example, were the differences between the mean scores on the dependent measure for the control group and the experimental group large enough for us to conclude that they could not have been due solely to chance differences between the groups? In the majority of cases significant results mean a real effect has been found. But, even though we try to be very careful, it is possible that occasionally a finding will pass our statistical analysis though it is a false positive. In other words, everything indicates it’s a real effect, but in reality it is not.

Replication is a further safeguard to ensure the integrity of scientific research. If a study can be repeated successfully, perhaps with some slight modifications, that lends a great deal of support as to the legitimacy of the findings. However, failure to replicate calls the original findings into question. Were the experimental protocols followed as described? Were the results correctly recorded? Were the statistical analyses correctly applied and carried out? Were the findings simply due to a rare instance of chance variation between the groups? Or in the most serious scenario, were the results simply fabricated? And additional failures to replicate add to these doubts exponentially.

As these articles point out, there are many studies that cannot be replicated yet are often cited.

The Replication Crisis (Links to an external site.)

Replication Failures in Psychology (Links to an external site.)

The Replication Crisis in Psychology (Links to an external site.)

This is a real problem. People assume an often cited article represents accurate information. But that may not be the case. Unfortunately, most people do not have well-honed critical thinking skills and are unlikely to investigate whether particular findings have been replicated. As noted in the first article, one study claiming a link between vaccinations and autism, that was later debunked, is still being cited and influences the decisions people are making regarding their children’s vaccinations.

Often these studies present interesting results. Sometimes that means they are aligned with suspicions people already harbored. Sometimes that means the results were surprising and counter-intuitive. In either case, they end up being cited often and become part of popular culture. And again, few bother to double check the facts. It should also be noted that studies that fail to replicate a well known study are themselves less likely to get published.

For this assignment I’d like you to find another such example, an article that is often cited, often regarded as accurate, yet no replications have been successful. To find one I’d suggest accessing one of the numerous articles discussing the replication issue and then tracking down an example noted within. The idea here is to get you to realize how prevalent this actually is and that nothing should be taken at face value, even if presented as science. In other words, I’m looking to foster some critical thinking skills.

For the assignment, discuss the study of interest and the basic findings. How many attempted replications failed? At some point was the original study formally discredited, like the vaccination and autism study? What do think made it interesting and thus often cited? What did you learn from this assignment? Although I’d prefer articles related to psychology, critical thinking applies to all branches of science, so anything you find interesting is acceptable.