finalsoln

# finalsoln - S TAT 8 200 F inal E xam Decmber 9 2 010 S HOW...

This preview shows pages 1–4. Sign up to view the full content.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
See the attached SAS program graders.sas and its output, graders.lst. The analyst who wrote graders.sas was interested in testing whether or not the graders all graded to the same standard and in estimating the mean score given by each grader, but was not sure how to implement the appropriate analyses, so he tried more than one approach in graders.sas. b. (6 pts.) Write down an appropriate model for the analysis of these data and write down the hypothesis that the analyst is interested in testing in formal terms (that is, in terms of model quantities). Test this hypothesis at a = 0.05 and state the appropriate conclusion. OJ ' ,/-or: _ /." } 1 v' -::: re J'~-/j 5c (~jrc;",-J. e C;:'i.ve~ j ".1 ItC; (-' ':7rooIC/ iJ~ J e '":'J.-( I ' lI. . II·' 0 . 7 //' U c,) i I) t I c~,?I'lt~ u. ~) 25" -c::! ~o ('-., C?ft C''- C I /lO /'Z. .t J. - e X{Z11Vl ') (':; ,) '. - . ) .J .j J ..' ) i \ (nD! oj/ DosS I .;/e (r:.l,"Vt~i.tlC;.-lJ1J 01' () J c:·.re <5 6.st/ Ve cL) /\/J. I!L I 7 .,.' J (' 2.. ) '_ I .. LD ~ . r -, . , t. l ,I -'~~. . ~ -- i ./' i ' '.' _ -+ J.+ 2>' + e. '-' 1 E':,' 'J r---J N u [; 2- or -= 0 L. & --= 0 .-- ,M <:71 [ r J lJ) lJ )') . I ). I J ft LI c. ,j ///1,. /'7 I .' ..J I /'-,' ,( OV'0.~" I ~ /'= SIl. c,de/' . \..) e eXt:"',";'" /~d J I('i) 3. \, /0'Jcll1.s 15, ': H; d 1;=-'" :::: o{z.-::C 0 ,~~v"- ~ (if!. {/ rU,.1 r-/ 'J C J Uf) i//~ 0.- J e feci E t KL Lj . A ::::- -.' := A.A } <- Dr 1,o~l"'\I- / £-,. . -~ L(, '1 f b A 2 L( 7{; d.l LJ LJi2 ',U =- /llA'f c>« ~;-0:3 ({C,Jz/ l , F / - j D -< .. 0 0 01 _~ ) ;2e;}2 C -; lb, ,b-r c 1.<.J£.r s } y;/-/ 1/ , v:v~£:~-;eI eXcc.V'o15 J (') /JDJ ';( f';-,. [)} e -h ,/'1.( J&vc>( Sl:,:V\.d~d" j CY } "J ) :'/ ~- (C~ 01'"\ C.j If c. (4 pts.) Estimate the mean score that grader number 3 assigns when j, c;; v{' Jc;~Vh€. grading writing samples of this kind and give a 95% confidence interval for /J ~ ) this quantity. /l/[.cL/{ SC,L/e - ;: b:J)( 62 / L/ C' JJJ cL of (0,(D)~ C7,95/Y) > .-.~r" 2
so (J/'/sl d. (4 pts.) Why do you think that 30 writing samples were used for this experiment? That is, assuming that each sample is evaluated by 5 graders and without changing the total number of graders (25), could we have used the same experimental design with fewer than 30 }yriting Ja~ples?

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 22

finalsoln - S TAT 8 200 F inal E xam Decmber 9 2 010 S HOW...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online