Question

One of the earliest estimates of the speed of light was made in California in 1926 by shining a beam of light from the top of Mount Wilson to a mirror on the top of Mount San Antonio. The mirror reflected the light back to Mt. Wilson, and an innovative contraption measured how long the light took to travel the distance, Researchers measured the speed of light 100 times and used the average as their final estimate of the speed of light (299,796 kilometers per second). When more accurate techniques were developed, it was discovered that 80 of the measurements were higher than the true speed of light and 20 were lower.

Part A: Which was most likely?
a. we cannot say.
b. The final estimate was too high.
c. The final estimate was too low.
Part B: Which bias influenced this estimate?
a. Selection bias
b. Non-response bias
c. Response/measurement bias

Answer 1

The final estimate of the speed of light in the 1926 experiment was most likely too high due to response/measurement bias. This bias can occur from systematic errors in the experimental method.

In the 1926 California experiment, researchers aimed to measure the speed of light by reflecting it between Mount Wilson and Mount San Antonio. The average of 100 measurements yielded a final estimate of 299,796 kilometers per second. However, when more accurate techniques emerged, it was discovered that 80 of the measurements were higher than the true speed of light, and 20 were lower.

Part A: The conclusion that the final estimate was too low can be drawn from the fact that a majority (80%) of the measurements were higher than the accurate value. Therefore, option c, "The final estimate was too low," is most likely.

Part B: The bias influencing this estimate is indicative of a response/measurement bias. The discrepancy in the measurements, with a significant portion being consistently higher, suggests a systematic error in the measurement process rather than a random variation. This could result from flaws in the measurement apparatus, observational techniques, or environmental conditions. Hence, option c, "Response/measurement bias," best characterizes the systematic error in the experiment that led to the final estimate differing from the true speed of light.