Medical researcher: At the Flegco Corporation, all employees whose work involves lifting heavy objects wear back belt...

Skylar on September 13, 2019

@Ryan-Mahabir Happy to help.

A researcher compares employees at a company who wear preventative back belts (which, as it states, are all employees whose work involves lifting heavy objects) with employees at a company who do not wear back belts (therefore, employees whose work does not involve heavy lifting). The researcher finds that the employees who wear the belts are actually more likely to have back injuries. From this, the researcher concludes that the back belts do not help prevent injuries.

This logic should raise a few red flags. Mainly, we should notice that the two groups that are being compared (those wearing belts and those not wearing belts) are not equal. Only the group with the belts does heavy lifting, so regardless of the presence of the belt, they will always have a significantly higher chance of back injury. In other words, since the two groups work different jobs with different levels of risk (for example, think of comparing heavy lifting jobs in a warehouse with desk jobs in an office, all working for the same company), the study is not well-controlled and the conclusion is not valid. The effectiveness of the back belts in preventing injury can only be ascertained when observed in two equal groups. This is why answer choice A is correct.

Answer choice D is one of those answer choices that is worded in a complicated way to try to trip you up. Don't fall for it- instead, take a step back and try to break down what the statement really means. In this case, it is essentially claiming that the argument is confusing lack of causation with preventative causation. This Is not the case, so it is incorrect.

I hope this helps to clarify. Please let us know if you have any additional questions!

Abigail-Okereke on November 18, 2022

Hello Skylar,

Thank you for your explanation. I still can't pinpoint what type of flaw this is. I believe it's a sampling flaw based on the stimulus; is it a biased sample?

Emil-Kunkin on November 27, 2022

Hi, this is indeed a problem with the sample. The author concludes something about a group without adding a control group. This sample is unrepresentative of the general population. I wouldn't worry about specifically categorizing a flaw, flawed arguments are often flawed for more than one reason.