ࡱ> =?<@ %jbjbqq Hlttttt$t{ 8 : : : : : : ,k  f f  8 6>68 T, ,  GAJtt, , { { , / / , MATH 116 ACTIVITY 6: Understanding Relationships between data - scatterplots and correlation WHY: Much of the use of statistics involves relationships between variables. It is important to understand the terms and ideas involved in describing such relationships in the same way that it is important to understand the terms and ideas used in describing distributions of single variables. LEARNING OBJECTIVES: 1. Learn the terminology and ideas for looking at association between two quantitative variables 2. Be able to find and interpret the correlation coefficient for a set of bivariate (two-variable) data. 3. Work as a team, using the team roles. CRITERIA: 1. Success in completing the exercises. 2. Success in working as a team and in filling the team roles. RESOURCES: 1. The Statistics Handbook chapter on Organizing an Describing data - especially p.6 2. The notes on "Association between variables" attached (as pages 4 & 5) 3. Your calculators and the sheet with instructions for your calculator (different for different individuals) 4. The team role desk markers (handed out in class for use during the semester) 5. 40 minutes PLAN: 1. Select roles, if you have not already done so, and decide how you will carry out steps 2 and 3 2. Read through the handout defining the basic terms and work through the examples given there [Note: the examples do not show how to find a general solution.] 3. Complete the exercises given here - be sure all members of the team understand and agree with all the results in the recorder's report. 4. Assess the team's work and roles performances and prepare the Reflector's and Recorder's reports including team grade . EXERCISES: 1. In a study of diet and exercise, data have been collected on the lean body mass (in kilograms- body mass excluding all fat) and metabolic rate (calories burned per 24 hours) of 12 women . The researchers hope to show that lean body mass [which is relatively easy to measure] is useful in predicting metabolic rate [which is harder to measure]. The data are given here: Subject # Mass Rate Subject# Mass Rate 1 40.3 1189 7 41.2 1204 2 33.1 913 8 50.6 1502 3 36.1 995 9 42.0 1256 4 54.6 1425 10 42.4 1124 5 48.5 1396 11 34.5 1052 6 42.0 1418 12 51.1 1347 a.) Which variable (metabolic rate, lean body mass) should be considered the explanatory variable, and which should be considered the response, [or does this distinction not make sense in this situation]? b.) Plot the data. Be sure to label the axes, indicating which variable is on which axis (Mass ranges from about 30 Kg to about 55 Kg. Rate ranges from about 900 calories to about 1600 calories). c.) Describe the association between the variables: Is it positive or negative? What is the form (linear, curved, complicated)? Does the association appear to be strong (a clear pattern, points close to a line or curve) or weak (no clear pattern or lots of scatter away from the pattern)? d.) Find the correlation coefficient r [use your calculator - do not work from the formula]. Does the value seem reasonable for the graph? 2. On the attached sheet there are 5 scatterplots, labeled a f. The correlation coefficients for the plots are 0, -.3, -.65, .8, and .4. Which diagram corresponds to which correlation coefficient? How can you tell? CRITICAL THINKING QUESTIONS:(answer individually in your journal) 1. Give an example of two variables (other than those in the text) that you would expect to show a positive association [increase in one would go with an increase in the other]. Is either of your variables a cause for the change in the other? 2. This scatterplot shows the Percent of High School seniors taking the SAT in each state on the x-axis and the mean SAT Verbal score on the y-axis. The correlation coefficient is r = -.884 a.) What characteristic of the graph (answer in terms of visible features of the picture) would tell you that r is close to - 1 (fairly strong negative correlation)? b.) What characteristics of the graph (the picture) does r not describe?  SKILL EXERCISES Exercises for Chapter 1, 9-11 Graphs for Correlation exercise a.) b.)   c.) d.)  e.)  Association between variables, correlation Two variables are associated if certain values of one tend to occur more often with some values of the second than with other values. If we wish to use changes or values in one variable to explain changes or values in the other, we refer to them as the response variable - measuring the outcome (think of y in our usual use of function and graphing notation) - and the explanatory variable - measuring the cause or explanation (think of x in our usual function notation). For two quantitative variables, the standard picture is a scatterplot - explanatory variable on the x-axis, response on the y-axis, plot the (x,y) - pairs. [If there isn't an explanatory variable - we aren't explaining one by the other - either variable can go on either axis]. Patterns are described by form (linear, type of curve - general shape of graph), direction (positive - variables increase/decrease together; or negative - when one increases, the other decreases; sometimes there is no fixed direction) and strength (how closely packed around the basic pattern are the data points?). Correlation is a measurement of the amount of linear (y = ax + b) association between two quantitative variables. The correlation coefficient r is given by  EMBED "Equation.3" \* mergeformat  - but we will calculate using the Linreg command on our calculators, or using computer software. The formula says: 1. Convert all values to Standard units (subtract the mean, divide by the standard deviation of the variable) 2. Multiply each standardized x by the matching standardized y , 3. Add the results 4. Divide by n-1[to eliminate the effect of more or fewer points being used] . (That is: we multiply the standardized values of x & y for each point, add them, and divide by n-1 to get an average). The value of r will be between -1 and 1. If r is close to 1 , the points are close to making up a line with positive slope If r is close to -1 , the points are close to making up a line with negative slope If r is close to 0 , the points show no linear association between values of x and values of y [no association, or the association is not linear]. See the examples on the other side of this sheet showing sets of points with different levels of r . What happens: If large x-values are with large y-values (and small x's with small y's) then the multiplications will give mostly positive numbers (+ times + or - times - ) and r will be positive - showing a positive association. If large x-values are with small y-values (and small x-values with large y-values) the multiplications will give mostly negative values ( + times - and - times +), so that r will be negative - showing negative association. If there is no such patterns, the signs will be mixed, causing cancellation so that r will be close to 0 - showing no linear association [r does not measure quadratic, cubic, etc correlation or matching to complicated curves]. In practice, r is not calculated directly from this formula, but with a calculator or computer. The calculator work will be the same for calculating the regression line equation [coming up] For TI-82,83: Enter the x-values in one list (using the Stat Edit commands) and then the linreg command [see the instructions for your calculator) - r is displayed as part of the results (TI-83's may require turning this on - see the notes or ask me) For TI-89 enter data in two columns with the Data Editor, choose calculation, then choose LinReg command For TI-86 Enter x-values in xstat, y values in ystat (all fstat values are 1), use the linreg command For TI-85 x-values as x1, x2 etc, , y-values as y1, y2,etc use the linreg command. For Casios- use 2-variable data entry, linreg command. Some scatter Plots, showing the corresponding values of the correlation coefficient r  `d /T%*".7Vcfghjkmz{|})= :B,7`|.RjB EHUVjB UVEH6 jJ%U jdU jU jU j U jU>*<5M_`no  n p - .  1$ `1$^``1$1$_`no  n p - .  + , R67UHI6WXglm{~<\GV"W"$$;%s%t%%%%%%^ + , R67UHI h1$^h`$h1$^h` 0 p@ P  T<$h1$^h` 0 p@ P  T1$1$I6WXglm{~<$a$1$ $1$a$1$1$\GV"W"$$;%s%t%%%%%%$a$2!!W""t%%%%% j,U>*5* 00hP/ =!"#`$`%| HH@R),, P `,-^&' Dd`h<  C ABT `=;{j9Prkmʖ0 D T( `=;{j9Prkmʖ 0(xXkT+JhBQBTRZyUG)+T$5v|i}l-E^KSߛ֥ٝ~~of޼y3?}Aimׄ]@ !Wٺ|z ‚/$)veXkmmşGs<ߍdcDYT"6z$gݪK{3 M.;c*d{'z$} b4qޘ#Y5vOGwfcZR6P=٢ygQDB`>;kbtƂv92 ڬ_b/y<[޵6A:tQhhHG1U~ʚ437}.@aG1U1Q p_.y.5u2O"dؼGνP7HgDl4[*dlӰWo$6o+~ HKlPa#"s, HJmHˤ>tɫ5T (J] | v>p&tE 8çU zS\nyn ΋C>CV0hr]Bg?rn0Ra~g\ܣB5>Vvo@Jև79'ED(B Nu (c1wCL~ 'GƏCΏuA ]RyqՔy"ҥoޖAlߠ#VZ20L;BMEmn+`DoB7f QW`ÄE]p|*\D\E pȼ؅Qu 4" j`ov8iFZc[gqsh!f͂ӹ2a1FMA.t|3/I؈C Wua}9Ad*^^rfo1zM4\e(Y3➓@5cEf8smqBeBT`^zLFXz#{ ~M(&m+vU#-苒㹟7J&h]W X܁69sUqxm$03}:ML1EIzX#VbEVm-΃ٍf$  W]+h_ףm#B5c^ABNX\6v+Z۔;%a#.-$;& Yܢ>mB x@ɨ-3}αv @.~ 'օTG߽\T}%ґёU uxYиIt[td1r!D 3t&;D@BﳨH=z^'tމpu)cd-޻nܡo_ݒni/;]Ѳ5u]Sr!u+?+ΐ3=ܼSR>؇su5uܖctEƯ_ǧZmwm ;kaadz+l#6` Ƚԩԓx@!,ya@4لtA7tk Zh=f|lpC7cAs DFج б3#}q&(&vio]AyLvil]Ik]"XG|H+"[CˋHtc[㊶@S{iZlއg,^?}In53kK'3>ssZ^>xMVO|{Dd <  C ABT`L%R#*H0( T(`L%R#*HUDF)xwXVeoȒHɌhA{آ=lA;h(&M(m Q+R CѨ4ZJ+}?Yȥ{]yy}wa bc2Fbn=NU gcĮ+>b5 b^{F1- [l%GEq"EDX6Vd$^HH?F5*"$=D":HHGFH'W-ds"]DSEl5[!SHo>*Z7מ"EY/pz"2rH?MHp,%h1"cm48Fx~/ "E'YUɪ)"t"D[|T=3E-?G8O" D/_l\ՠ!2ri5\=EWXRQWsHZuv5pHFM"E(\BVm"Ew4jT EWH~x@UAC"E3mvDF?!RV3",=ٽY?gg:HHH#GkWtEEDnE,;YHU`*xHXR`w"7ZMUK-EE6+]͊{""{F`>$/4FY(?F?V?N?^Vl6q "D'"O-+T?I_&ҟ,ҟbS5:EF)ҟ&/*骒W"E|#ҟaf'үiVf۬94*\"jիm}kJK sr?dfk3??JM5`\XT2.,R2.,r2,c`Wq26x-~Ѫf5"3qȌȌ̸~5lJBk'3!ӭ)!v):d.١P#:7e']'MhM@R8(򵧨O6~qU~5Q[:uWVb -tq8?tmdxrx:7r[(!jŸ^[X7&Ъgj)q-pJvՠ!Tl<LXwkjJʊs*{la}fAuBypmytL]',nh; 04qU`-n&]8t`4wk+Qo+1şt7~1n0p|7~;%<Lz^p5YŦ6GtvlJ|vi{Z܈EhoS ssi;hoVj.V-\/#38] *FKwp[`/sb0ܯP372~-ZDRBn˂ϖGt*/_{} Dd <  C AB:U4yCY <&I0  TU4yCY <&I0XG< *<xݗwXUuǿ%hA{آho=lkjTZ"Qj%IeCqSYX)xM@P&}9wy=9yypϽ$L̘Hφ` >A[02w)эSkmȳV#\RX"cs-dU'RMUU"#Eۈ{lT QVSdf;~H{A"OiUpf(I&V ղH7"=D{ڮ{iWۻљ"D[8@Y$O,?Ī D!?R#'?VqHxpfEDGH$"L=4{&?]HL[%pV"sDo\ݠ%2ra5\3"- \(KE\.ҿB \_-ҿFH:6 @пA&"[l6 D!ҿS?f ܥYwH"D9:~"#?,vxT7HȌ'L1P0"\vHHi3VDD""gm͂"{z`iNjL(I?VUџ$EdUd~xM?E_$]F5ig`-Gc3D""Y"VyE"#?,f |YL@ds Kۂ~HLPHk[5PEӯ/#V]Cbϸ5VrsKsr,㲌 ii j*PUX XU?Vg*X_b CȊ6kV]dE6f&YOWW54aVY߹ְ&Zam5f v!C7E :ŵ_V>!EE|Xg 8z-Qy.}>,߭gDlW㡻b$)v@ Td]U=^4>@F/ >*tx TUUД )5˲ylyyhN$+mҢy,Ԃ@-R),R{߽VZֽ߽|gs}IxxF0 Ntܬau@8]tT9" 1:"VB;qPSRY@Y(ҏo.$²FK"D "DHҬwUMVo#)%-vl+'/OingnyHGN"EXU]U^Vj tH@T/DF駈kY"D(?H*$?LqH[pVE(GHh`1zzlj/?AH$%pvɫSDO.?êjO9+֙9"!Uy"E\ ҿP.h/_&Ҹ\hWy\-ҿFyNH}7olnѪyuH6";D;N]"#w+ҿϪܯ*@?LdOPkxPg?"T?\)F#4"'DO=Xf\j\l\h,RXX#`XXz2PRf&"3ڪqȌDfEd_EfMd"3iU{5˶LkXn Ҭac5!h η o7E GS\P#|L7 Y -E O.uQ.48Cf6Ym9\/tV"\`T񤪙fAd-l[p9sy &)ţ9 &3fxAE] fGnk9-;cB3ql9hUʁ:J|ȫM-bfh \kn"EA5p]Bnᾀ;(;-""3Ow P@"A"l6 xJ?-,FH9 Qu9"\~HgHy[0Lp Iz_dFhG,E xM?Z?FduX8=0^H@?Af5L;#Wk74 ESDSEDӭ*0CUT$__,ҟ)/فY"#?闊?Uh?#v|.r}[/i~!v |MZH[|wv> B!¢"EYYPfqIF4Ҥ$2*.?V'* X_+`+`+BEY&aȊjV\%jYYqȊka]*ְ!jac5lʰ4kK"|)Z1ޭm:HKd;N  |9u-ب޿YXbnFmn;VB5+Bt,Zs3Nr;*wy1׺:y7ֺu\(Z^Xֹ)^x;+j6D'^H5M1*G5#ͅGjhɡO8Л-U` w­Ca2ǒݚH\/꼜Dv-AT$j61:ZҌbvkyP'fg@3Zge61FWVD Ss\{uQ* Oi~bM~iLsk!ԠuDZݥEABCDFGPYIJKLMNOQRSTUVWXRoot Entry FuaM@Data %<WordDocument HObjectPooluaMuaM_1108268428FuaMuaMOle PIC TPICT   !"#$%&(,-012345689:;=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnoprTm< &dxpr  &"& currentpoint " +r)=(1(n)-) 1"(Bx +i (M-) x"V(Js +x"@ ( : * (: ( ] * (] (ky +i (u-) y"(ss +y"i ( c * (c (  * ( (,Z30 dict begin currentpoint 3 -1 roll sub neg 3 1 roll sub 4512 div 1216 3 -1 roll exch div scale currentpoint translate 64 39 translate -9 665 moveto /fs 0 def /sf {exch dup /fs exch def dup neg matrix scale makefont setfont} def /f1 {findfont dup /cf exch def sf} def /ns {cf sf} def 384 /Times-Italic f1 (r) show 249 665 moveto 384 /Symbol f1 (=) show 840 426 moveto 384 /Times-Roman f1 (1) show 591 958 moveto 384 /Times-Italic f1 (n) show 863 958 moveto 384 /Symbol f1 (-) show 1122 958 moveto 384 /Times-Roman f1 (1) show /thick 0 def /th { dup setlinewidth /thick exch def } def 16 th 567 566 moveto 738 0 rlineto stroke 2052 426 moveto 384 /Times-Italic f1 (x) show 2225 522 moveto 224 ns (i) show 2407 426 moveto 384 /Symbol f1 (-) show 2719 426 moveto 384 /Times-Italic f1 (x) show 2708 109 moveto 183 0 rlineto stroke 2319 958 moveto (s) show 2475 1054 moveto 224 ns (x) show 2009 566 moveto 914 0 rlineto stroke 1810 399 moveto 384 /Symbol f1 (\346) show 1810 976 moveto (\350) show 1810 757 moveto (\347) show 2935 399 moveto (\366) show 2935 976 moveto (\370) show 2935 757 moveto (\367) show 3365 426 moveto 384 /Times-Italic f1 (y) show 3527 522 moveto 224 ns (i) show 3709 426 moveto 384 /Symbol f1 (-) show 4020 426 moveto 384 /Times-Italic f1 (y) show 4010 109 moveto 171 0 rlineto stroke 3624 958 moveto (s) show 3780 1054 moveto 224 ns (y) show 3323 566 moveto 890 0 rlineto stroke 3124 355 moveto 384 /Symbol f1 (\346) show 3124 1020 moveto (\350) show 3124 801 moveto (\347) show 4225 355 moveto (\366) show 4225 1020 moveto (\370) show 4225 801 moveto (\367) show 1361 752 moveto 576 ns (\345) show end dMATH r=1n-1x i -xs x ()y i -ys y ()   mFMicrosoft Equation 3.01ELO CompObj 'RObjInfo)OlePres000 *(Ole10Native +Equation.3 r=1n-1x i -xs x ()y i -ys y ()   Equation.3 Oh+'0Ole10FmtProgID .1Table</ SummaryInformation(/DocumentSummaryInformation87@u۟K)pW堦ԻTp &XZPc+O[=iԘhP\q.KU0UEPcZ"Q0Lyy(HVB- 1=iԘK78_2zHݓ f,E{\k_SP˲,hxf%Wy4K`u W`mIS[NwҨ0ao`WW~$?%DIQX =j0v*tf;}rvs&wRn%~06҉%|)f96~HRˡns!zxtTlsK$oЎ,MhX'pd'L˂ٌSTURἣެ*kUY@*0^"ةL#3'2AWg͜x65Đl7\~㪳gܦ8 )z<^EM$3{7c!Ń\jн$7]az2ӓ4_7NLJ|Cv1ͿA~>"tJ$O[ot5WJ`'ffq&ǷlȊWJXc˾g<[v Wwn5 sX.`R=i+r䁫{t udmqZHWs)nYs`I;bG+.bn`11Z3"8o7&c,X z-K˗c[v!Q|j'bJezifnT+hPf{23Vi(:} > ; kCr}҃ ,_[7=]#::^>6l6W6fW e'&yuQcLG!*qu GKar7С! EO3.RK{lb?+w!*y7l_jW٣Rp4?Hp<"g(w%Q0za~4l8+{`t}ʯMEp5oL*b8n2s.Cbk9JY1[. 㹞`+]WHV놼h$&r-ŜОT5xRjRIIc^ZG}$Fk둩Gvº&_'fG`8Y D!`"s%Age^pFpo\9H 20w{E.^#zҵ0Ȧxޥ7J~2P(i4tR2=Nķv\t!tů;}ѕ.pkTǭbHP/VͩJBRnkA+O:w1Q ޝ81o&ӘT?ɎF)`y@M<QٕVWzqAԜI<N&E{-*(pvid sM#? cUsBf͒5CTk`Ѡ]ƹ3,RIDq$'Rax 629Y[+@T#<^xWDjAH**n,ڛ&FP>>O?s۳yR=$Jt(6I0 ?`o-NE q"@PNe}x/Jֺ$>kއ.:>NOc+KUx24ձN0GQ85mVt-mNOF|>d'Ui[۟3sznۏMF7)O1F_Ȉ`6^VLn$]~#+1p37،%'a&Ef^҅]Ƣ;O|^Z_p*:(pshl:e0:$7 O u4)|-;.]=4^⛎ߛVzr}a}+< ۊ0sʕڀ[ HdFԓ+5.i2=>xW +{3,f#JSjCseD6epԦ6nM0א?nkҝBJM-ުգ eL~~E1`zIƘr:ƻy.R lNZ}@S/36t֑9"xhK, 9+Cy "}0:_Ϲ,:_g4-Rb Z\Fm,4?zwa@xV6zEQ1+1LAcSB3"[8#)4XY}2Pt ?>Kq*A,]:0Ffj>uhtd^F&e>lc2% ݋ 1~50w$Ӟ L5b2 BWBb!f_6Xw?xou]%:\sSu^˗۹cu MP *w]) f9` &x7zk8ѧiS ,(oZ<\P8]ӼշOjIejƭP_uK)^)4%#n['GVcShC8RVd?yeͼ(f׾mcT#)c/P<t0;|@L6GI/K@4l;ϙL *]9cEk52 c1zs\DwB~:Er&xc~}IrG `~ K29a %gvM`β [2r_ݡd V܃[mwٰbJ Fr'\)-%ǧ3Dr0 Y.5W׃ X{N}3Fxyy\:aS٨>/l%{lK:Fӵ^z4 م,> *t [H?NrI1K&f,yҮJU A{oȤ-NZ2 ӓJFcC*TbQ(YjVFh)`|5m0]av@HR% nEeN%/靘|&b0 T8F'AnIY-W9,eQPJ;uHZ[xԕڙ"ۧSzj.2n&]VȟP9&;f^{ԽZnEPTwp[(?ɪ q]V-%⺆:~73;.zg8jWOϙɢ83+ km8%ӵqϬpB*)͝ p^1tGr)I}@nqγoKfO^ Ѷ4ܙZ^#k^SR  !_kKHӊmW}J<ĵ t)1vڟ"fv3zo4x~m)a6"tjpD'b>]9@