2. Many statisticians argue that hypothesis-testing proceduresdo not provide muc

Question

2. Many statisticians argue that hypothesis-testing proceduresdo not provide much useful information. Rejecting the nullhypothesis tells us only that it is unlikely that the sample camefrom a population with a specific mean. However, procedures likethe z-test do not give much information about the "true" value ofthe mean from which we are sampling. Means may be statisticallydifferent, but there is no indication of whether the size of thedifference is large enough to be considered important in real worldterms.

How does this relate to results in Question 1 above? (Besure to discuss power, effect size, and statistical significancevs. practical significance .)

When we reject the null hypothesis, have we always found animportant effect ? What other considerations may be important?

For several of the sampling exercises we used a one-tailed testwith alpha of .05 (z= 1.645). If we used a two-tailed test with thesame alpha value (z = 1.96), would you be able to reject your nullhypotheses more or less often? Explain

Explanation / Answer

When analyzing data, your goal is simple: Youwish to make the strongest possible conclusion from limited amountsof data. To do this, you need to overcome two problems:

Statistical analyses are most useful when you arelooking for differences that are small compared to experimentalimprecision and biological variability. If you only care aboutlarge differences, you may follow these aphorisms: If youneed statistics to analyze your experiment, then you've done thewrong experiment.If your data speak for themselves,don't interrupt!

Related Questions

Navigate

• Browse (All)
• Browse 2
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.