Heart Rates The resting heart rates tity sanrencorba ta 7 bel per Minus witha tandard deve 459 beats per minute, What percent ofthe ‘resting heart rate of ttc) “a, greater than 80 beats per minute? “by between 70 and 85 beats per minute? | 35, Cereal Weight The weight ofall the boxes 7% tse ied by machine ae somes fern with a mean weight of 14.5 oz and a standard dew ‘of 04 02. What percent of the boxes will a a weigh less than 14077 b. weigh between 13.5 oz and 15.5 02? Telephone Calls A telephone com, ; company bas foun thle eth of ng Stans ep “normally distributed, wih a mean of 235 sand eta | dard deviation of 55 s. What percent ofits long distane Se Percent of ts long distance a longer than 340 5? _b. between 200 and 300 5? |. Rope Strength The breaking point ofa particular ype of ope is normaly sib, witha mean of 380 Ib and a standard deviation of 24 to, What is the j ‘probability that apiece of this rope chosen at random will have a breaking point of a less than 32016? », between 340 and 370 Ib? EXTENSIONS 1 In Exercises 43 to 49, determine whether the given state “met is true or false "43, The standard normal distribution has « mean of 0. '44. Every normal distribution is a bell-shaped distribution, 445, Ina normal distribution, the mean, the median, and the mode ofthe distribution are all located atthe center of. the distribution. “46. The mean of a normal distribution is always larger than the standard deviation ofthe distribution. 47, The standard deviation of the standard normal distribu tion is 1 [fection 4.5 Linear Regression ‘When performing research studies, scientists often wish to know whether two variables are related, If the variables are determined to be related, a scientist may then wish to find fan equation that can be used to model the relationship. For instance, a geologist might vant to know whether there isa relationship between the duration of an eruption of a SECTION 4.5 | LinearRegession and Corelaion (EM 38. Tire Mileage The mileage for WearEver tires is ‘normally distributed, with a mean of 48,000 mi and @ standard deviation of 7400 mi. What is the probability that the WearEver tires you purchase will provide @ mileage of more than 60,000 mi? », between 40,000 and 50,000 mi? 39. Grocery Store Lines The amount of time customers spend waiting in lineata grocery store is normally sistributed, with a mean of 25 min and a standard ‘eviation of 0.75 min. Find the probability thatthe time customer spends waiting is fa. Jess than 3 min De. less than 1 min, 40. 1Q Tests A psychologist finds thatthe intelligence quotients of a group of patients are normally distributed, With a mean of 102 and a standard deviation of 16 Find the percent of the patients with IQs above 114 b. between 90 and 118, Heights Consider the dataset ofthe heights of all babies born in the United States during a par- ticular year. Do you think this daa set is nearly normally distributed? Explain, 42. [AWB Weights Consider the dataset ofthe weights of all Valencia oranges grown in California during a particular year. Do you think this datasets nearly normally distributed? Explain. a 48. Ifa data valve» from a normal distribution is positive, then its z-score also must be postive. 49, Al normal distributions have a mean of 0. 50, a. Make a sketch of wo normal distributions that have the same slandard deviation but different means. 'b, Make a sketch of two normal distributions that have the same mean but different standard devi 51. Determine the approximate z-scores for the first (quartile and the third quartile of the standard normal distribution Linear Regression and CorrelationCHAPTER 4 | Statistics pin this determination i © ci ions. A frst data, Table 4.1» geyser and the time between eruptions. ATTN 4 pivariate data. F Some data. Data involving two variables and the duration of the scng fiat. a inne betwen two eros 42 variate data showing ene eruption for 10 eruptions ofthe geyser TABLE 4.11 ‘Time between eruptions Ainseconds),x 203 | 270 | 218 | 226 si | ss | 78 | 8 “an 85 shown Once the data are collected, a seater diagram or seatter plot can be drawn. crane ge E% i en. & <097.) E 0 Gono mo aon mo nom? — 030030220 20 290 0 20 2 Scns been stone sends between erpions FIGURE 4.15 FIGURE 4.16 ‘One way forthe geologist o create a model ofthe relationship between the time betwee? two eruptions and the duration ofthe second eruption is to finda line that approximates ‘the data points ploted in the scatter plot. There ate many such lines that ean be drawn 3s shown in Figure 4.16. Of all the possible lines that can be drawn, the one that is usually of most interes i called the line of best ft or the least-squares regression line. The least-squares regression line is the Line that fits the data better than any other line that might be drawn. The least squares regression line is defined as follows, The LeastSquares Regression Line ‘The least-squares regression line fora set of bivariate data is the line that minimizes the sur ofthe squares of the vertical deviations from each data point to the line. In this definition, the phrase “minimizes the sum ofthe squares of the vertical =v tions" is somewhat daunting. Referring o Figure 4.17. it means anor sible, the linear equation that minimizes the sum oe al Geass asad s dea,a SECTION 4.5 |unerRegesion na corenion EER i ray b isthe equation ofthe line of best fit In this expression, each d, represents the distance from data point nt the line are ee 3 Length of eruption (seconds) 0 200 210 220 230 240 230 260 270 * ; ‘Seconds between eruptions en : FIGURE 4.17 . ¢ ‘Applying some techniques from calculus, {is possible to find a formula for the least- squares line. The Formula for the Least Squares Line Tistrdtion tousethe symbol | The equation ofthe least-squares line fo then ordered pairs (pronounced of y in the equation of the least- Grose squares line. This helps distinguish | ig 9 = ax +b, where Treen the Ln’ ysales he abevcalailted fontieequ | (5.9306 Gi Ie) tow and the dat’ j-values he =7- ar sales given in the abe. ‘To apply this formula to the data for Old Faithful, we first find the value of each Bx= 286 Yy= 822 B= 573,560 Bay = 196,636 TAKE NOTE Next, we use these values to find the value of a “There ae other lines, such asthe qo M9 = E08) smedian-median line, tat canbe nde — (a fai sperms si rie = {2090196 636) ~ (2386)(822) (10)(573,560) — (2386)"~ °-1189559666 ‘We then find the values of and, ia = Bt, 2386 By _ 922 i go ean 2B _ns ‘and use them to find the y-intercep, b bay ak = 822 ~ 0.1189559666(238,.6) = 5381710637WR conrren 4 | srsrcs ‘The regression equation is $ = 0.1189559 sion equation and a scatter plot of the data are show” sone + 53 8171067. The rah ofterepe, below Length of eruption (seconds) (0 ao 20 220 230 240 250 240 270 * Seconds betwen eruptions ‘We can now use the regression equation to estimate the duration ofan eruption given the time between eruptions, For instance ifthe time between two eruptions is 200 secon then the estimated duration ofthe second eruption is $ = 0.1189559666x + 5381710637 = 0.1189559666(200) + 5381710637 =78 ‘The approximate duration ofthe eruption is 78 seconds As our example demonstrates, it can be challenging o calculate all ofthe values needed to find a regression line, Fortunately, many computer programs and calculators can pe form these calculations. The following example shows the use of a TI-84 to find the regression line forthe Old Faithful data, Enter the data from Table 4.11 into L1 and L2, as shown > aan atthe right ie 7 a3 536 lee 203 at 20 as 6. Press the [STAT] key, tab to [ERLE] and scroll to 4:. Then Press| Ser tote Sre Reg ine. rsh [key an ‘scroll to [Y=WARs} Press [ENTER] twice. =SECTION 4.3 | Lnearteyeson an coralaon EE [IES EERTER) twice. The lope a and yimercept 6 ofthe teepe-squares fine ae shown, You wil see two addtional alues, and r, displayed on the screen. We will discuss the meanings ofthese values later ‘The equation for the regression line is stored in YI. Using 200 seconds as the time between eruption, as we did above, we can calculate the expected duration ofthe erup~ tion as follows Pres the [TARE] key and scroll [Y=vaws) Press[ziTe) “°° 77 gasaon69 twice. Now enter "(200)" and press [ene] The predicted ‘uration of the eruption s approximately 78 seconds. Here is an additional example of calculating regression lines. Professor R. McNeill ‘Alexander wanted to determine whether the stride ength of a dinosaur, as shown by its fossilized footprints, could be used to estimate the speed ofthe dinosaur, Stride length for | fan animal is defined asthe distance « from a particular point on 2 footprint to that same | poiat on the next footprint of the same fot. (See the figure atthe left) Because dinosaurs fare extinct, Alexander and fellow scientist A. S. Jayes carried out experiments with many types of animals, including adult men, dogs, camels, ostriches, and elephants. Some of | the results from these experiments are recorded in Table 4.12. These data will be used | inthe examples that follow. TABLE 4.12 ‘Speeds fr Selected Stride Lengths i ‘Ault men fee] = [SST sop aya leased > Lo [ss Lee [oo] 7 Po Pa vw sana 1 [at a Da as eee 7 Lo [os | [ant on | ce. Camels omme] =|] l= [se ~e Tapa] sweating |) | Poo [ss Per [a Pe]MIEN ciate 6 | ststsncs EEO Find the Equation ofa Least Squares Line eS ae irs in Table 4.124 Find the equation of the least-squares line forte ordered pairs in Tab Solution Enter the data into a calculator or software proges Here are the results using a TI84 calculator. um that supports fegresson equations Gaeue ression equation is = 2.730263158x ~ 3.316447368, Find the equation of the least-squares line for the stride length and speed of camels given in Table 4.12c. It can be proved that for any set of ordered pairs, the graph of the ordered pair (Z, 5) is a point on the least-squares line that models the set. This fact can serve as a check. If you have calculated the least-squares line for a set of ordered pairs and you find that the point (9) ‘does not lie on your line, then you know that you have made an error. 0 X3.6, 65125) Running speed (meters per second) og aes 6 Stride length (meters) Once the equation ofthe least-squares line is found, it ean be used to make predic tions, This procedure is ilustated in the next example TEXAN Use a Least-squares Line to Make a Prediction ‘Use the equation of the least-squares line from Example 1 to predict the average speed of an adult man foreach ofthe following stride lengths. Round your result. est tenth of a meter per second, a 28m. bh 48m Solution 4. From Example the repression equation is $ = 2.73026315&e ~ 33164473658 Substitute 2.8 for x and evaluate the resulting expression, i aa $ = 2.730263158x ~ 3.316447368 = 2.730263158(2 8) ~ 3316447368 ~ 4328 The pret average speed of an adult man with side length of 28 min 43 ms b. From Example 1 the regression equation is $ = 2739263188, a sieas7 oe Sobe 1 fox and Cale eeu gree ee SO $ = 2.730260158r ~ 3316447368 2:730263158(4 8) ~ 3316447368 = 9.739 The predicted average ped ofan ad man with a ride length of 4m jo the nea: nmSECTION 4.5 | Linear Regression and Correlation = CCRC RTUTACITETESSRN Use the equation ofthe least-squares ine from (Check Your Progress Ito predict the average speed of a camel for each of the followin Stride lengths. Round your results to the nearest tenth of a meter per second. a 27m be4sm imENOTE ‘The procedure in Example 2a made wse of an equation o determine a pint betwee ple 2e, anes P : siven datapoints. This proceduce is refered to as interpolation. In Ex Seca” ton was used to determine a point to the right of the given data points. The process sof on are not reason __usng an equation to determine a point the right or left of given data pints i se fiionace i've who (oar extrapolatlon See Figure 418. Peeler x= 20m, meer uOsTSEr 3 SIGEST36S Sarees poe Ti pessspecd of hou S13 ms : (reso) ‘Bese tbe maximum stride (45,95) leh of adult men is consider {sp less than 20'm, we should not inst predicton. 12 3 4 5 Sie length of adult men (meters) FIGURE 4.18 loverpoltion and extrapolation Linear Correlation Coefficient ‘To determine the strength of linear relationship between two variables, statisticians use ‘statistic called the linear correlation coefficient, which is denoted by the variable r and i defined a flows CRSTORCAIES Xai Pearson (ison) spent | Iostf his career asa mathe- fais professor at University Coleg, London. Some of his "ar contributions concerned the Serecpment of statistical proce- des uch as regression analysis tcoelation. He was paticu- | terested in applying these | acon fo te tu) of ety. The term standard deva- {owas invented by Pearson, and eet inemtad by Pearson 2rd | gar eorelation coefficients pstv, the relationship Between the variables rebton, the emcl encawen | pas.a postive correlation. In this cae, if one variable increases, the other variable clog eleer corelton couficet | fends to increas. fr is ezative, te liner relationship between the variables has a neg {hePeeson product moment | tye correlation. this case if one Vatable increases, the oer variable tends todeerees snc of carlton. eason Figure 4.19 on the next page shows some scatter diagrams along withthe type af de ofthe stats! | nr gonelation that exists between the sand y variables. The closer iet0 Lethe oo ser te linear relationship is between the variables, Linear Correlation Coefficient For them ordered pais (51.1) (8,93) (35) wor Bn | coefficient ris given by ), the linear correlation se ny) ~ (By) Vat") ~ Bx? - Vat) — yyMEE carren «| sinstc * Lk be, Strong positive correlation, r= 08 «. Positive correlation, a, Perfect postive r=06 correlation, f. Little or 90 linear «: Nepaivecoreation, = correlation, r= 0 4, Strong negative ave correlation, r= -0.9 FIGURE 4.19 Lincar correlation [SXXCIERBA Find a Linear Correlation Coefficient Jent for stride length versus speed of an adult man Find the linear correlation coeffic wrest hundredth. Use the data in Table 4,12a. Round your result to the nea Solution The correlation coefficient r is displayed on the same soteen as the values for the regression equation. The seteen from Example 1 is shown again at the right. ‘The linear correlation coefficient, rounded to the ne est hundredth, is 0.99. ARTES Find the linear correlation coefficient for siride length versus speed of a camel as given in Table 4,12c. Round your result tothe nearest ‘hundredth. KESTEESIIOIM wat is the signitcance ofthe fact tha the linear correlation cocticiest 4s postive in Example 3? ‘The linear correlation coeflicient indicates the strength of a linear relationship between two variables: however, it does not indicate the presence of a cause-anuf relationship. For instance, the data in Table 4.13 show the hours per week that a student spent playing pool and the student’ weekly algebra test scores for those same weeks TABLE 4.13 [Algebra Test Scores vs, Hours Spent Playing Pool ‘ours per week went paying pool | |S ‘Weekly alsa test score 2 [eo 7” 8 TED ise pote co era inde ove craton wena ma re st nh ines peda ree ns nate THSECTION 4.3 | nea Reyesion and coreoion EEC me ton coefficient for the ordered pais in the abe r~ 0.98. Thus the ting Mane Positive linear relationship between the students algebra test scores an Ne time the student spent playing pool. This does not mean that the higher algebra test Teauch Mere caused by the increased time spent playing pool. The fact thatthe students roan es increased with the inerease in the time spent playing pool could be due tO Imany other factors, o it could just be a coincidence. : 4n Sour work with applications that involve the linear correlation coefficient r. i is ‘The linear correl ‘important to remember the following properties of r Properties of the Linear Correlation Coefficient |} The linear correlation coefcint ris akways areal number between —1 and 1, inclusive. Inthe ease in which * all ofthe ordered pairs lie on a line th positive slope, ris ® all ofthe ordeted pairs lie on a ine with negative slope, ris —1. For any set of ordered pairs, the linear correlation coefficient rand the slope of the least squares line both have the same sign ng the variables in the ordered pairs does not change the value ofr ‘Ths the value ofr forthe ordered pairs (X.9) (X,¥s-~» ne) is the same 4 the value ofr for the ordered paits (3.83) (08a) =v nS 4, The val of r doesnot depend on the units used. You can change the units of @ le frm, for example, ett inches and the value ofr will remain the same. 3. Imtercha Exponential Regression Earlier in this chapter we examined linear regression models. In some cases, an exponet tial function may more closely model a set of data. For example, suppose a diamond ‘merchant has determined the values of several white diamonds that have different ‘weights, measured in carats, but are similar in quality. See the table below. TAKE NOTE + 400ct 300et 200 175 1Het 125ct LOVE O7Set OSD Tre ale of diamond i gen Ay determined by is color cu hy, and carat weight. These stances ofa damon ate nor asthe four. tn the ‘We can use the data in the table to determine an exponential growth function that models ‘sample atthe right we have the values ofthe diamonds as a function of their weights, and then use the model to pre {sumed thatthe color, cut, and_—_ gg the valu ofa 35-carat diamond of similar quality. Using a graphing calculator the Sltiy fall ofthe diamonds are exponential regression equation is y ~ 4067.6(1.3816),, where x is the carat weight of ‘similar. This assumption enables th giamnond and y isthe value of the diamond, sto model the value of each amond as a function of just ts ‘514500 $1070 $7500 $7300 $4700 $4290 $5800 $5000 $400 Exores, sant wei Ue asx we 2 = W067. Gunes Bai Sereuie6 o> Saueplais 997573238WEEE cunrren «| sarin POINT OF INTEREST ‘The Hope Diamond, shown below, isthe wor’ age deep-bhe diamond weigh 52 carats f should not expect the func tion y = 4067.6(1.3816)' 10 Yel an accurate value of the Hope Diamond be reause the Hope Diamond isnot the same type of <éiamond asthe diamonds in the ‘example, and its weight is mach larger, Con ape eet ‘The Hope Diamond ison display at the Smithsonian Museum of Natural History in Washington, D.. of a3.5-carat diamond o esi the valde equation 10 regression : ‘Tose the regression alate quality, substitu 35 0" y= 4067. 611.3816)" y= 4067.6(1.3816)"° oo ‘According tothe model i $12,608. 4 35-arat diamond of similar function, the vale of tat 7, the coefficient od “This indicates that the equ mination, is about 0: yy = 4067,6(1.3816)" provid EXCURSION EXERCISE ws Earth’s atmosphe yn that imate, erie pressure Patan altitude of a modes the atmospheric Pressure as 3 tothe nearest tenth, the atmo 41, The following table sho meters Find an exponential functo function of altitude, Use the function 1 st spheric pressure at an altitude of 11 km Exercise sEeT 4.5 41. Which of the scatter diagrams below suggests the 1. strongest positive lin xvand y variables? strongest negli svand y variables? ¢ linear correlation between the 2. Which ofthe scatter diagrams below suggests ‘a. a early perfect postive linear correlation betwee the x and y variables? bs lite orno linea correlation between the x and 1 variables? Jation between the3, Given the bivariate data: a, Draw a scatter diagram for the data, b, Find n, Ex, By, 7, (Sx), and Say, ¢. Find a, the slope of the least-squares i ntereept ofthe astsguares ie ne 4. Draw the least-squares line on the scater diagram from part a. ¢. Isthe point (Z, ) on the least-squares line? {. Use the equation ofthe least square linet pred the value of y when x 4 o Led Find, tothe nearest hundredth, the linear correlation coefficient. 4, Given the bivariate data | 3] 6 3 [a [s ~ 1. Draw a scatter diagram forthe data, », Find n, Ex, Sy, Ex, (Bx) and En. . Find a, the slope ofthe least-squares line, and b, the | +intercept of the least-squares line. 4. Draw the'leastsquares line on te seatter diagram from part a Is the point (F, 7) on the least-squares line? {Use the equation ofthe least-squares line to predict the value of y when x = 7.3. Find, to the nearest hundredth, the linear correlation coefficient. 1 In Exercises 5 to 10, find the equation of the least ‘squares line and the linear correlation coefficient forthe sven data. Round the constants, a 6 and r1o the nearest hundredth, 5. (2,6), (3,6). (4.8) (6,11) (8, 18)) 6. {(2, 3), (3, ~4), (4, 9), (5s —10), (7, 12) 7. (3,118), (-1, 9.5). (0,86), 2,8. (5.5.4)} 8. {(-7, -11.7), (-5, -98) (3, -8Ds (1, 59, (@,-52)} 9. {(1, 4.1), (2,6.0), (4, 8.2) (6 115), (8; 162)} 10, {0 5), (3,7), (4, 8) (6, 11), (8 18), 9. 200) SECTION 4.5 | LncarResesson and coretaton [EEE In Exercises 11 to 18, use the statistics features of & sraphing calculator. ¥ Value of a Corvette The following table ives retail values of a 2010 Corvette for various s. (Source: Kelley Blue Book website) [TGansorrnting | Roaivane| ao [sar re aocoo | 851200 asoon | 8295 e 32,000 $49,075 1. Find the equation of the least-squares Fine for the data, Round constants to the nearest thousandth. '. Use the equation from part ato predict the retail price of a 2010 Corvette with an odometer reading, of 30,000. . Find the linear correlation coefficient for these data. ‘d. What is the significance of the fact that the linear correlation coefficient is negative for these data? 12. @ Paleontology ‘The following table shows the length, in cen timeters ofthe humerus and the total wingspan, in ‘centimeters, of several plerosaurs, which are ‘extinct lying reptiles. (Source: Southwest Educational Development Laboratory) Prerosaur Data ‘Humerus | Wingspan | Humerus | Wingspan 24 600 20 300 32 750 7 370 2 450 15 300 17 370 15 30 | 1B 270 9 0 44 e 4a s 32 33 29 7 1s 4 . 4 Find the equation ofthe leas squares line forthe data. Round constants tothe nearest hundredth ». Use the equation fom part ao determine, othe nearest centimeter, the projected wingspan of a pterosaur if its humerus is 54 cm. —REE carrer 4 | satsics 13. Health “The 18, Centers or Disease Con 'evention (CDC) use @ measure cal mass index (BMI w determine whether: peron ‘overweight. A BMI between 25.0 and 299 is considered ‘overweight, and a BMI of 30.0 or more is considered obese. The following table shows the percents of US. ‘ales 18 years old or older who were obese in the years indicated, judging on the basis of BMI. (Source ion) 2003 | 227 2 | 239 | 200s | 249 72006 | 253 | 2007 | 265 2008 | __266 200 | 276 Using 3 for 2003, 4 for 2004, and so on. find the equation of the least-squares line for the data, b. Use the equation from part ato predict the percent of overweight males in 2015. 14. @ Health The US. Centers for Disease Control ‘and Prevention (CDC) use a measure called body ‘mass index (BMI) to determine whether a person is ‘overweight. A BMI between 25.0 and 29.9 is considered ‘overweight, and a BMI of 30.0 or more is considered ‘obese. The following table shows the percents of U.S. females 18 years old or older who were overweight in the years indicated, judging on the basis of BMI (Source: Centers for Disease Control and Prevention) fa. Using 3 for 2003, 4 for 2004, and soon. find the equation ofthe least-squares Hine fr the data by. Use the equation from part ato predict the percent of overweight females in 2015. .e_ The following table shows Phone cee ante wines krone rie United States for recent years, subscriptions in the U.S, Wireless Telephone Subscriptions ‘Subscriptions, in lions ‘Source CTIA Semi Anal Wires Survey Midyear 2010 «the linear correlation coefficient forthe dat b. On the bass of the value ofthe linear correlation cratfcient, would you conclude, atthe [r| > 09 fovel thatthe data can be reasonably modeled by a Iinear equation? Explain, 16. @ life Expectancy The average remaining life mes for men of various ages in the United States are given inthe following table. (Source: National Int tutes of Healthy Average Remaining Lifetimes for Men Thee] Years [ Age | Years o | mo | 65 | 168 15 | wo | 75 | 102 as | #20 Use the linear correlation coefficient to determine ‘whether there isa strong correlation, atthe level jr] > 0.9, between a man’s age and the average remaining lifetime of that man. "7. oe Life Expectancy The average remaining life- times for women of various ages in the United States are given in the following table. (Source: National Institutes of Health) Average Remaining Lifetimes for Women ' Find the equation of the east-quares line forte lata. b Use the equation from part a to estimate the remaining lifetime of a woman of age 25. €. te pede nu anexamplef inter tion or extrapolation? *carrer 4|sinmey SEER Find the linear correlation coetficent forthe dat. bers the data given in the following table, : a sovsthereoomendel maria eee MES haste sigieance of he valu ound in prt for individual of the piven ages ¢. Find the equation ofthe least squares in. 4. Use the equation from part c to predict the maximum exereise hear rate fra person whois 72. 6 Is the procedure in part d an example of interpola tion or extrapolation? Maximum heart rate {y beats per minute) EXTENSIONS 19. Tuition The following table shows the. average 'b. Using 1 for 2009-2010, 2 for 2010-2011, and so on, annul tition and fees at private and public # ear cok find thelnearcorelaioncoeficient andthe qua- Jeges and universities for the school years: 2009-2010 ‘tion of the least-squares line for the tuition and fees through 2014-2015, (Source: National Center for Ei 21 public 4-year colleges and universities, based on cation Statistics) ee . Based on the linear correlation coefficients you Fouryea Colleges an Univeitis Tn and Fess found in pars wand bare the equations you wrote wer it parts « and b good models ofthe growth i tition 3 and fees at 4-year colleges and universities? 2009-2010 4. BAW The equation ofa least-squares ine is written io inthe form $= ax + b. Explain the meaning a ‘ofthe value ofa fr each equation you wrote in 2011-2012 parts aand b 2012-2013 20. @ ‘Search for bivariate data (in a magazine, in Tee ‘newspaper, in an almanae, or on the ae Internet) that can be closely modeled by a linear 2014-2015 equation, ‘a, Draw a scatter diagram of the data b. Find the equation of the least-squates line and the linear correlation coefficient forthe data, ©. Graph the least-squares line on the scatter diagram in part a 4. Using | for 2009-2010, 2 for 2010-2011, and so on, find the linear correlation coefficient and the equa- tion ofthe least-squares line forthe tition and fees at private 4-year colleges and universities, based on 4. Use the equation of the least-squares line to predict the year ‘range value fora specific domain value. The following table summarizes essential concepts in this chapter. The references given in the right-hand column list Examples and Exercises that can be used to test yo uur understanding of a concept Measures of Central Tendency Mean, Median, and Mode The mean fn mimbersis the sum ofthe | See Examples 1,2, and 3on pages 102 numbers divided by n. The median ofa ranked ts ofm numbers isthe | 10 104, and then try Eneeise | on page 158, ‘middle number ifn is odd, othe mean ofthe two middle numbers if n is even. The mode of a list of numbers is the number that occurs most frequently, ] | | | |