What type of error occurs when a treatment effect truly exists, but fails to be detected?

A Type 2 error occurs when a treatment effect that truly exists is not detected in a study. This means that the study concludes there is no effect when, in fact, there is one. Type 2 errors are associated with failing to reject a null hypothesis that is false, which can happen for various reasons, such as insufficient sample size, low statistical power, or variability in the data.

In contrast, a Type 1 error involves detecting an effect that does not actually exist, leading researchers to conclude there was a significant treatment effect when there was none. Selection bias refers to a systematic error arising from how participants are chosen for a study, which can distort the true effect of a treatment. Measurement error encompasses inaccuracies in how variables are assessed or recorded, which can affect the validity and reliability of the results but does not specifically relate to the detection of an existing treatment effect. Thus, the essence of a Type 2 error directly pertains to the scenario where a genuine effect is overlooked, making it the correct choice in this context.