Week One Chapter 8 Homework

8.48 A sample of 20 pages was taken without replacement from the,591page phone directory Ameritech Pages Plus Yellow Pages. On each page, the mean area devoted to display ads was measured a display ad is a large !lock of multicolored illustrations, maps, and te"t#$ %he data in s&uare millimeters# are shown !elow' 0 2(0 )5( *0) 5)( 0 2(+ )( *2+ 5)( 2(+ )9( *(9 5)( 1(2 ))+ *0) 5)( 5)( 1)0 (a) Construct a 95 percent confidence interval for the true mean. mean , )*($5 standard dev , 1-0$)-+ df , 19 t (95 interval) , 2$09)0 t (99 interval) for the later part , 2$+(09 (!) Wh" mi#ht normalit" !e an issue here$ .ormality might not !e a fair assumption, since we don/t know if the pages were taken randomly, for e"ample$ %his data set has some outliers, for e"ample, the data points of 0$ (c) What sample si%e would !e needed to o!tain an error of &'( s)uare millimeters with 99 percent confidence$ 0 , t1stdev23#42 , 2$+(0911-0$)-+210#42 , 2)-5$92 5ound up to' 2)-( pages (d) *f this is not a reasona!le re)uirement+ su##est one that is. this is not a reasona!le sample si6e, since that might even !e more pages than the phone !ook contains7 8e should either lower the confidence re&uirement or raise the accepta!le error$

+$(* 9iting an un-popped kernel of popcorn hurts7 As an e"periment, a selfconfessed connoisseur of cheap popcorn carefully counted --) kernels and put them in a popper$ After popping, the un-popped kernels were counted$ %here were +($ (a) Construct a 9( percent confidence interval for the proportion of all kernels that would not pop. 3 , 61 p-hat2s&rt n## 3 , 1$(*51s&rt: +(2--)# (+-2--)#2--)# , 0$011)1 90; <$=$ , +(2--)#-0$011)1 , +(2--)#>0$011)1# (!) Check the normalit" assumption. pn, +(2--)#--)?5@ n&, (+-2--)#--)?5 .ormality assumption holds (c) ,r" the -er" .uick /ule. 0oes it work well here$ Wh"+ or wh" not$ AB5 ,p > and - 12.4$5 0$0-52+0$1*-222 (d) Wh" mi#ht this sample not !e t"pical$ %his is one !atchC$not a random sample from the producer$ Dou cannot !e sure that appropriate sampling si6e and selection# were followed$