Hypothesis tests or confidence interval for population mean usually require the use of a t- procedure. Requirements for using a t- procedure are • Samples may be regarded as SRS's. • Populations are normally distributed. (Actually, with r-115 it is enough that there are no outliers or strong skew. With r-‘40 anything goes.) Walden University claims that its faculty spend 11.0 hours in classroom teaching per week. You work for a student newspaper and are asked to test this claim at the 0.05 level of significance. A sample of classroom hours for randomly selected faculty members is:
11.8 8.6 12.6 7.9 6.7 10.4 13.2 9.1 10.8 9.8 10.4
1) Con we use t-procedures? These data are nartclorrily selected, so it depends on whether the population is cistributed normally or approximately ncrmaly. Make a sternplot:
Is Itin distribution opproximotely norrnal? SO we con test Oils using a t-lest of population mom, 2) Stale alternate and null hypotheses (hypotheses are always about mot/inters):
H,• : 1A.)
Ho:,....!..4
31 Enter data into LI. Use 1-VarStats to find-.
:.¦
4l x is the test statistic. Use a formula to find r. the standardized test statistic.
r=
-po
5) Sketch a density curve for the sarnpfing dstnbution for this situation. assuming that the nu hypothesis is true. Locate and tabei values of both the mean of the cirsinbutiOn and I on the horizontal axis and shade an area amid to this test's P-value.
61 Use STAI>>11.515-21-Test to rind the I Choose "Data," since the data are in t,,, then In appropriate Information and enter "Calculate"
I¦ P 2 •