0% found this document useful (0 votes)
1K views5 pages

Chapter 11 Practice Free Response Answers

The document provides summary data from the Johns Hopkins Regional Talent Searches test between 1976 and 1978. It shows the mean scores and standard deviations of males and females on the verbal and mathematics sections of the test. A t-test is used to determine if male scores are significantly higher than female scores on the mathematics section. The t-test results in a p-value less than 0.05, so males score significantly higher in math.

Uploaded by

ChitraBanarjee
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
1K views5 pages

Chapter 11 Practice Free Response Answers

The document provides summary data from the Johns Hopkins Regional Talent Searches test between 1976 and 1978. It shows the mean scores and standard deviations of males and females on the verbal and mathematics sections of the test. A t-test is used to determine if male scores are significantly higher than female scores on the mathematics section. The t-test results in a p-value less than 0.05, so males score significantly higher in math.

Uploaded by

ChitraBanarjee
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 5

Chapter 11 Practice Free Response The Johns Hopkins Regional Talent Searches gives the Scholastic Aptitude Test

(intended for high school juniors and seniors to 1!"#ear"olds$ %n all& 1'&((! )ales and 1'&'!* fe)ales took the tests +et,een 1'(- and 1'(.$ The )ean scores of )ales and fe)ales on the ver+al test are nearl# e/ual& +ut there is a clear difference +et,een the se0es on the )athe)atics test$ The reason for this difference is not understood$ Here are the su))ar# data$
1. Group Males Females

x
416 386

s 87 74

(a 1eter)ine if the )ale scores are significantl# higher than the fe)ale scores at the 2 -$-1 level$ State #our h#potheses& check conditions& identif# the significance test #ou are using& and then find the test statistic and the P"value$ 3rite #our conclusion in nontechnical ter)s$ Step 14 H - 4 male = female )ale scores are not significantl# higher than fe)ale scores H a 4 male > female )ale scores are significantl# higher than fe)ale scores

Step .4 Since is unkno,n& ,e ,ill use a t,o sa)ple t test$ 1$ 3e are not told that either of the sa)ples are fro) an SRS& therefore ,e )a# not +e a+le to generali5e to either population$ .$ 6oth Sa)ple si5es are large$ 7ur results should +e accurate$ X1 X . s1. s. . + n1 n.

Step !4

t=

= !*$-8

Fro) our T% (!9(: using degrees freedo) !(*(;$'; our p value is P (T !*$-8 = -

Step :4 Since - < -$-; ,e ,ill reject our null h#pothesis and conclude that )ales score significantl# higher in )ath than fe)ales on the Scholastic Aptitude Test$

(+ =ive a ''> confidence interval for the difference +et,een the )ean score for )ales and the )ean score for fe)ales in the population that Johns Hopkins searches$ %nterpret the interval$ Step 14 3e ,ant to esti)ate & the )ean difference +et,een the )ean score for )ales and the )ean score for fe)ales in the population that Johns Hopkins searches ,ith ''> confidence$ Step .4 Since is unkno,n& ,e ,ill use a t,o sa)ple t interval$ 1$ Checked A+ove .$ Checked A+ove

Step !4 ( X 1 X . t

s1. s. . + n1 n.

( .*$'1;&!.$-(; using !(*(;$'; degrees freedo)

Step :4 3e are ''> confident that the )ean difference +et,een the )ean score for )ales and the )ean score for fe)ales in the population that Johns Hopkins searches is +et,een .*$'1; and !.$-(;$ 2. The level of dissolved o0#gen in a river is an i)portant indicator of the ,ater@s a+ilit# to support a/uatic life$ Aou collect ,ater sa)ples at 1; rando)l# chosen locations along a strea) and )easure the dissolved o0#gen$ Here are #our results in )illigra)s per liter4 :$;!& ;$-:& !$.'& ;$.!& :$1!& ;$;-& :$(!& :$:-& ;$:.& 8$!(& :$-1& :$88& .$(*& ;$*!& ;$;; Construct a ';> confidence interval for the )ean dissolved o0#gen level for this strea)$ Follo, the %nference Tool+o0$ Step 14 3e ,ant to esti)ate & the )ean dissolved o0#gen level for this strea) ,ith ';> confidence$ Step .4 Since is unkno,n& ,e ,ill use a one sa)ple t interval$ 1$ 3e are not told that our data are fro) an SRS therefore ,e )a# not +e a+le to generali5e to the ,hole population$ .$ if n is +et,een 1; and !' dra, graph (can +e ske,ed +ut still ,ith no outliers and nor)al pro+a+ilit# plot ()ust +e linear =RAPH SH73B %B CCASS 7ur graph of the data is s#))etric and the nor)al pro+a+ilit# plot is linear in for)$ 7ur results should +e accurate$

? Step !4 X t

s n

(:$.;&;$.'

Step :4 3e are ';> confident that the true )ean dissolved o0#gen level for this strea) is +et,een :$.; and ;$.'$ 3. %n a stud# of the effectiveness of ,eight"loss progra)s& :* su+jects ,ho ,ere at least .-> over,eight took part in a group support progra) for 1- ,eeks$ Private ,eighings deter)ined each su+ject@s ,eight at the +eginning of the progra) and 8 )onths after the progra)@s end$ The )atched pairs t test ,as used to assess the significance of the average ,eight loss$ The paper reporting the stud# said& DThe su+jects lost a significant a)ount of ,eight over ti)e& t(:8 2 :$8(& P < -$-1$E (a 3h# ,as the )atched"pairs t statistic appropriateF 6oth )easure)ents are +eing done on the sa)e individual and ,e onl# have one sa)ple$ (+ G0plain to so)eone ,ho kno,s no statistics +ut is interested in ,eight"loss progra)s ,hat the practical conclusion is$ This support progra) ,ill significantl# reduce #our ,eight over a period of ti)e$ (c The paper follo,s the tradition of reporting significance onl# at fi0ed levels such as 2 -$-1$ %n fact the results are )ore significant than DP < -$-1E suggests$ 3hat can #ou sa# a+out the P"value of the t testF The pro+a+ilit# that the results happened +# chance is ver# ver# s)all$ Practicall# 5ero$

4. The Bational Gndo,)ent for the Hu)anities sponsors su))er institutes to i)prove the skills of high school teachers of foreign languages$ 7ne such institute hosted .French teachers for four ,eeks$ At the +eginning of the period& the teachers ,ere given the Hodern Canguage Association@s listening test of understanding of spoken French$ After four ,eeks of i))ersion in French in and out of class& the listening test ,as given again$ (The actual French spoken in the t,o tests ,as different& so that si)pl# taking the first test should not i)prove the score on the second test$ Ta+le *$1 gives the pretest and posttest scores$ The )a0i)u) possi+le score on the test is !8$

3e hope to sho, that attending the institute i)proves listening skills$ Carr# out an appropriate test of this clai) at the = -$-; level Step 14 H - 4 difference = - attending the institute does not i)prove listening skills H a 4 difference > - attending the institute i)proves listening skills

Step .4 Since is unkno,n& ,e ,ill use a one sa)ple t test$ 1$ 3e are not told that our data are fro) an SRS& therefore ,e )a# not +e a+le to generali5e to the ,hole population$ .$ 7ur sa)ple si5e is .- ,hich is )ediu)$ 3e )ust sho, a graph of the data (an# shape +ut no outliers as ,ell as a nor)al pro+a+ilit# plot (linear =RAPHS SH73B %B CCASS The graph sho,s a ske,ed left distri+ution +ut there is one outlier and the nor)al pro+a+ilit# plot sho,s a curve$ 7ur results )a# not +e accurate$ X t = = !$(8 Step !4 s n

P (T !$(8 = -$---;

Step :4 Since -$---; < -$-; ,e ,ill reject our null h#pothesis and conclude that attending the institute i)proves listening skills$

You might also like