A research project was undertaken to evaluate two focimeters. Each of 20 lenses of varying powers was read once on each focimeter. The measurement differences were then calculated, where each difference was Focimeter A €“ Focimeter B. Assume readings are normally distributed.

a. Using a t-test on these paired differences and an a = 0.01 determine whether the corresponding population mean difference is significantly different from zero.
b. Construct a 99% confidence interval for the mean difference in focimeter readings.
c. Explain what both inferential procedures indicate about the differences.
d. If an enterprising experimenter performed this same test using a = 10% what would the outcome be? Offer comments about proceeding using these ground rules.

Transcribed Image Text:

0.016 0.013 0.009 0.000 -0.005-0.015 -0.006 2 -0.016 -0.022-0.006-0.020 0015 -0.017 -0.010 0.003 0.0 -0.012 0.008-0.005-0.009