4h: MEASURES OF DISPERSIO We have studied various measures of central tendency such as mean, median, m in the previous chapter. But they are not adequate to describe the distributi For instance, consider the following sets of observations : Set A:35, 37, 34, 38 and 46 with mean 38 : Set B : 10, 90, 45, 12 and 33 with mean 38 Set C : 38, 38, 38, 38 and 38 with mean 38 - All the sets have the same mean 38, but if the values in the sets are. obse carefully, it can be seen that in set C, the average 38 completely represents the: distribution; in set A, only one value is represented by the average and in set average 38 represents none of the values. Also, the variation of the itemsis nil i set Cand is maximum in set B. Thus, it is quite clear that in addition to averages, some additional information aboiit the variation of items is required, to know the extent to which the val a vary from one another and from central value. Z ‘A measure of spread of scatter of the data is called a measure of varial dispersion. : ‘The measures of dispersion can give us idea about reliability of the aver: ‘When the dispersion is less, the average is more reliable so thatit isa better| ofthe population average, and if, the dispersion is more, the average is nota representative of the data. The measures of dispersion determine the extent of variation in the da! which, some steps can be taken to control the variability. For instance, in facto quality control techniques can be applied to control the variation. The measures of dispersion can be ised to compare two or more distribu The one with less dispersion is more consistent or homogeneous and the one more dispersion is less consistent. “ The study of dispersion leads to further advanced techniques in analysis sit Statistical Quality Control, Cost Control, Inventory Control, etc. Scanned with CamScanner; zes of Dispersion of a Good Measure of Dispersion a Ershould be easy to understand and easy to caloulete, pesould be rigidly defines, fr should be based on all the observations, a igsbanld be capable of further algebraic trecament, a Er should not be affected much by extreme (op) Bstould have sampling stability f quero ere tao types of measures of dispersion, ee Absolute Measures giving actual extent A scatter f the bate wna (@) Bdative Measures expressed as pure numbers, indesrendecrs Of the writ | i Conespending, to cach absolute measure Oh Cispersion, 2 relalive measure cen We defuei which can be used to compare two or mee distriesions, Now, yeas tuaky | hesemezeres. valuss, 1. RANGE | Range for Set. A = 46 -34-=12 | Page or 52 B=W-16=0) | Pse he Sac=3%-23=0 & for he sash, BC, mes Fel fe means seh hare te re Pe Wing ae the prices in % of Gittxent trands of thevision war, Find the etn P29, 199, 799, s9159, i, SI) i iP. Rang» = Veaxismuem Price - Miniroum Price ls, fhe pa, 180 SOO = 39) i "Fast A prices is% 3M, E Scanned with CamScanneri no 2. CORFFICIENT Jcisa rotative menmire of diapernion and th dettned as, ~yeitto! ‘i Maximum Mininuns CootMeient of Range 5) Mii urn vale, hus, olifteret 0 sinvony sacl sino Thus, it is a ratio ofthe difference and the nun of nie eanuredl, Roing.a ratio, tig independent ofthe unit in whieh the original variable 147 Mastration 23 Find the eoetficient oF range for the following date relating, to prices of AhArEs OF "ABC" company during Ww F2S,75, 23,10, 23,80, 29,25, 2 Solution ; The maximum price = € 29,25 The minimum price © € 2 Maximum» Minimum * Maximum + Minimum 29,25 = 23,10 O15 Meee ance = 01N75 20.25 6 23.10 5235 ‘The range and coefficient of range are used to measure the variation in prices commodities, shares, temperatures, rainfall, ete, over a time period, The range! used to measure variations which are highly sensitive like gold and silver pa In day-to-day life, question like "How many answer papers are assessed in a or How many marks are expected in a particular paper ? or How much profit a shopkeeper make per day?" are always answered in the form between, two eal value. . But statistically it is represented by single number which ix interval bet maximum all minimum value called as range, Coefficient of Range = Merits of Range (i) Wis easy to und (ii) Mis rigidly defined, (iii) For small sample size, range is a good measure, (iv) Itis affected by extreme values and its value changes from sample Limitation (i) Tt can not be calculated for open end classes, Gi) I does not take intoaccount the deviations of inlivicluabs te? fron” of central tendengy, (ii) JLis not based on all the observations, and and easy (o compute, yor Scanned with CamScannereasures of Dispersion 3. SEMI-INTER: RAN . -QUARTILE- OR QUARTILE DEVIATION = tris defined as follows. Semi-inter-quartile Range _ The semi-i nter-quartile range considers on! ly the mi set b jiddle 50% of th i and it ignores the first and the last quarter. It is an absolute nabs ee deviation also measures the average amount by which the two quartiles Sea 2nd Q; differ from. median. Ser ee SEATS, 4, COEFFICIENT OF QUARTILE DEVIATION | {risa relative measure andis defined as i Ga Coefficient of QD. = Q, + Q\ itis a ratio, it isa pure number, so that it can be used to compare two.or more istributions. | Illustration 3 : Find the quartile deviati | student. Also find the coe’ 4 Set A %) : 34, 45, 53+ 42, 39 ‘on for the following data of pocket allowances () of 15 fficient of quartile deviation. 35,40, 51, 57, 52,47, 62, 55, 50; 63 | Solution : {| Arrange the sets in ascending order of magnitude as | Sea: 34, 35, 39, 40, 42,45, 47, 50; 51, 5253; 55.57 62,63 _ Here, n= the total no, ofobservation= 15” th 1 3 ] observation = 4th observation Lhe : ssecyation = [2th observa 2G S52 8 18 Scanned with CamScanner122 Business Statistes (1 aw Qs -Q Coefficient of Quartile Deviation = a Y a 3s * Oe Now consider the quartile deviation for grouped data. Mustration 4: 00 Calculate quartile deviation for the following distribution of dges of 8 eee Also find the coefficient of quartile deviation. Age in Years 20-25 | 25-30 | 30-35 |3: No. of Persons 50 70 100 Solution : As it is a continuous distribution, cumulative frequencies and to locate the re first prepare the following table to obtain equired class intervals. ‘Cumulative Frequency (less than) ge in Years] No. of Persons Here, N = Zf= 800 (a) For Q,, consider than 200, the required class for Q, is 30-35. Now /, = 30, /,= 35, f= 100, cf= cumulative freq. of previous class = 120) N (lp - oft a f = 3 + G5=30)(200 - 120) 100 = 30 +, 2280) _ 49 , 400 100 To 34 yrs. N a = 200, As220is the first cumulative frequency grea Scanned with CamScannerh pleasures of Dispersion 123 3N Pe Fr Q- consider — = 600. As 670, is the first cumulative frequency exceeding 60. whe required class interval for Q, is 45-50. = 120, cf= 550 120 So. the quartile deviation is 6.54 years and coefficient of quartile deviation is (R163. Mezizs of Quartile Deviation | @ Wissimpie w understand and easy to calculate. Gi) Asx isbesed on middle 50% of the observations, it isnot affected by extreme © ¥alnes. So, it is useful in erratic data. GB) com be calculated for open end classes. @) The quantile deviation can be obtained for qualitative data which can not be mezsured but can be ranked. _ Limitetions @ bis nox based en all observations. | ® Binet copsble of farther mathematical treatment. GD wis ated by sampling fluctuations. i” 4 does not consider the observations in the first and last quarter. 2 gives 2 rough idea about the scatter of the data and itis a better average Scanned with CamScannerBusiness Statistics 124 EXERCISE 1 mean by dispersion ? ¢ of a good measure of dispersion? and cocflicient of range in practice. ) of two companies during the last week of 1 expect similar tendency to continue if the 1s shares would you like to buy if you T° @) 1. What do yo 2: Whatare the requisites 3, Explain the use of range The quotations for shares Gn April, 92 wereas follows. Ifyou decks to come which company" wrest (0) aspeculator 7 ee ‘Company B 885 [ ‘Ans. Coefficient of range for A= 0.0224, Coefficient of range for B= So, (a) company A.and (b) company B 5, The student of the groups A and B obtaine Mathematics. Find which group gave more col ‘on coefficient of range. a : Marks of Group A #35, 47,42, 45, 40, 39, 38, 46, 41,43 Marks of Group B 35, 70, 95, 65, 60, 42.73, 80, 53, 57 ‘Ans. Group A is more consistent.” 6, Discuss the meritsand demerits of 7. Find quartile deviation for the following set of Il milk for 11 days bya dealer... 34, 37, 32, 25, 40,45, 50, 42, 39, 47, 55 ‘Ans. Q.D.=6.5 lit. | 8, The temperatures recorded at the end of every 4 hours during @ day att cities Bombay and Nagpur are given below. : ‘Temperature (in degree Celsius) Bombay - 28, 30, 34, 31, 28, 27 Nagpur - 25, 28, 34, 40, 32, 26 Use.appropriate measure of dispersion and comment on the Ans. Cotte of range for Bombay = 0.1148 and for Nagpur it lemperature at ‘Nagpur has more variation. .d following marks in a test of nsistent performance, based frange as a measure of variation. bottles of 1 litre of energy variation. 430.2308 Scanned with CamScannerb-~ of Dispersion - - “Find the coefficient of quartile deviation for the following data : Monthly income | No. of Farmers ‘ in? 1500 - 1700 1700 - 1900 1900 - 2100 2100 - 2300 2300 - 2500 2500 - 2700 “ans, Q, = 18.60, Q, = & 23.20, Coefficient of Q.D. = € 0.1100 10, Find the quartile deviation for the following data ; Age 10 20 30 40 50 60 70 80 (less than) No. of 15, 30 53 15 100 0 | 115 125 Persons Ans. Q, = 20.5435, Q,= 47.5, QD. = 13.4783 years 11. Find the semi inter-quartile range for the following data : Daily Wages] 0-10 } 10-20 | 20-30 } 30-40 | 40-50 | 50-60 | 60-70 | 70-80 ing No. of 10 7 26 30 33 25 12 9 es [ © Ans. Q, = € 25.19, Q, =% 52.2, Q.D. =¥ 13.5039 _ 12. Find the quartile deviation for the following distribution. x mates 14 | 15 ete 18 | 19 | 20 § f | 8 [45 | 17 | 21 | 30 |. 23 | 15 | 8 | 3 Ans. Q,= 14,Q,=17,Q.D.=15 , 13. Discuss the merits and limitations of quartile deviation. 14. Find the semi-inter-quartile range for the following data : Marks No. of Students Ans. Q, = 14.7368, Q, = 34, Q.D. = 9.6316 Scanned with CamScannere Business Statistics | 15, Find the coefficient of quartile deviation, 37,6667, Coefficient = 0.2454 Ans. Q, = 16.7647. Qs = jate the three quartiles and the quartile deviatio 16. From the following data calcul and its coeff ‘Ans. -21.16 years, 34.14 years, 47.81 years 13.325 years, 0.3864 ATION 5. MEAN ABSOLUTE DEVIATION OR MEAN DEVIATION (M.D.) “The earlier two measures of dispersion viz. rane and quartile deviation are not tesed on all the observations. Also, they do not jndicate the scatter of the Gbservations from average value. Mean Deviation (M.D. overcomes both these Jier measures of dispersion. drawbacks of the earl ition : M.D. is defined as the ‘Arithmetic Mean of the absolute deviations of Defini wpe observations from any suitable constant, say ‘A ‘Thus ifa variable X can assume values2x,.% ~-%,and A'isany arbitrary constant qa ineen absolute deviation or mean deviation from A is defined as 1 M_D. from A= => Z|x-A] For a frequency distribution, where class-intervals and frequencies are given 1 MD. from A= 5; B/1x-Ah and are the class-mark and frequency, respectively of the where x, ‘class and =f ‘The corresponding relative measure of dispersion is Coefficient of M.D. from 4 = pL Dnome, ‘Note: ‘ yh 1 Though ‘A"is an arbitrarily selected constant, generally for statistic! att : is = ae one of the measures of central tendency such as mea i mode, Iis observed that M.D. from /4"isminimam when A=" Scanned with CamScannerNe asures of Dispersion ‘ 127 M.D. is the simplest measure of dispersion that takes into account all the yalues in a given distribution. But it has some limitations also. First, as it takes into account the absolute deviations (i.e, does not consider signs of the “ geviations), it is unwieldy in mathematical operations, Secondly, it is jnfluenced by extreme values, MD. is useful in statistical analysis of economic and social phenomena, using ‘small samples. © ence, forthe observation x), x,, z M.D from mean = n bi - | MDfrom median = ©! . n n > |% - Modd| MD from mode =. <1 i n Ifthe observations 2,,.X,,...«5X, have frequencies /;,f, ..-.,Fespectively, M.D from Mean: MD from Median = © $ fila, — Model E sibs ~ Model M.D from Mode = =~ N The corresponding relative measures of dispersion are M.D. from Mean ne : M.D. from Median _ Coefficient of M.D. from median = yy Coefficient of M.D. from mean = Scanned with CamScanner128 Coefficient of M.D. from mode = ‘Mustration 5 = Find mean deviation and coefficient of mean devi data representing ages of 7 children. 5, 6,9, 11, 12, 13 and 14 Solution : It is raw data. To calculate mean deviation from mean, first compute ™ _ srergen ewer 2 = E 7 7 roy + Ut 101 a -#1= 15-10) + 16-10 412-101 + [13-101 + [14 - 10| 2434+4=20 ean aS Now, iI a544titit Zh - x; - | _ 20 MD ftom mean=2U = = 2.8571 ” ; M.D. from Mean Coefficient of M.D- fommean=———yyean 2.8571 = = 0.2857 | ent of M.D. for tHe Ilustration 6 : from median and corresponding coeffici Calculate M.D. following data = 100, 150, 200, 2: _ Solution : 50, 360, 490, 500, 600, 676 observation = 5! observation = 360 feta . yh nil N=9, Median = M= 2) iw [a Pa [50 Ten [0 | pefete tile Le] [wo Scanned with CamScannerbat _ M.D. from Median Median Ma 360 7 0.4833 ulustration 7: Find MD. from (i) mean, (ii) median, (iii) mode and (iv) 21 for the following data © and verify that M.D. is minimum when taken from median : 17, 19, 18, 22, 19, 18, 49,21, 19, 20, 24, 20, 23, 25, 21. © Solution : : X: 17 19 18 22 19-18 19 21 19 20 24 20 23 25 21 : EX _ 305 —() Mean= =~ = [> =20.33 [X= Mean] : 3.33, 1.33, 2.33, 1.67, 1.33, 2.33, 1.33, 0.67, 1.33, 0.33, 3.67, 0.33, 2.67, 4.67, 0.67 = |X = Mean| _: 27.99 a i279. (i) For median, we arrange the data in ascending order as follows : 17, 18, 18, 19, 19, 19, 19, 20, 20, 21, 21, 22, 23, 24; 25 <. M.D. from Mean = = 1.866 ...(1) ee pee = Median =( 2 |X— Median | : 3, I, 2,2, 12,1, 1, 1, 0,4, 0,3, 5,1 =|X = Median| _ 27 n 15 th } observation = 8" observation = 20 :. M.D. from Median = = 1.80 ..(2) Gi) Mode = 19 |X—Mode |: 2, 0,1, 3, 0; 1,0, 2,0, 1, 5, 1,4, 6,2 E\X — Mode! _ 28 = ig (3) 7” 15 ~ Gy) [X-21|:4,2,3, 1, 2,3,2, 0,2, 1,3, 1,2,4,0 2 Xr 30:5 4 i = 4) From (1), (2), (3) and (4), we can see that M.D. is minimum when taken from median, -. M.D. from Mode = *. M.D. from 21 = Scanned with CamScannerrowing lable, ry from the fol ei 2 LAG a oe N 37 ‘Now prepare two more columns in the table representing ber 3] and fils; -4 to get sum of the last column. z Sify - | = 47.62 M.D. from Mean Mean Coefficient of M.D. = ustration 9 : aM from mean and corresponding coefficients of M.D. for the flo Fepresenting daily wages (in %) of workers in a factory. Scanned with CamScanner"solution * “Pally Wa ‘ges [No. of Workers| “ 63-67 68-72 B17 78-82 83-87 88-92 93-97 98-102 103-107 = 2 _ 3830 _ 97.7333 = 77.73 75 _ syle — Mean) _. 495.03 _. 6 6004 MD .foin Mean = a me M.D. from Mean _ 6.6004 ‘efficient of M.D. from Mean = Manct Gp Tia 0.0849 Itis rigidly defined. Itis easy to understand. Itis based on all the observations. N) Its value is minimum when calculated from ¥) It is less affected by extreme values. (Vi) As it takes into account deviations from averages it is moré scientific than tange or quartile deviation. ‘Amitations (). Ttrequires more calculations for continuous variables. ati : ; It ignores the negative signs for deviations and only absolute values are considered. median. A a6 Scanned with CamScanner% . ; ae Be “sain and also 1, Find mean deviation from arithmetic mean for the following dat the coefficient of mean deviation, , a 1357, 1454, 1389, 1405, 1485, Ans, 41,2, 0.0291 i ind mean deviation from mean for the following data : 47, 49, 22, 24, 49, 66, 51 Ans, 12 3. Calculate mean deviation from mean and median for 100, 150, 200, 360, 490, 671, 500, 250, 600 Also, find the corresponding relative measures. Ans, 174.44 units, 173.44 units, 0.47, 0.48 . fe 4. Calculate M.D. from (i) A.M. Median, (iii) Mode for the following data representing marks (out of 25) of 9 students : 7, 4, 10, 9, 15, 12, 7, 9,7. Withe marks are converted out of 50, will the M.D, in marks increase ? Ans. 2.34, 2.56, 2.56; If marks are doubled M.D. from mean is also doubled. 5. Calculate mean deviation from A.M. for the following ungrouped frequency distribution : x to [1 [12 [13 [14 f 3 | 2 [is [re] s Ans, 0.75 units 6, Find mean deviation from mean for the following data: the following data + 1.2 years 7. Find mean deviation from mean and its coeffici representing age distribution of 50 boys, ient for the following da! Ans, 8 No. of Employees Scanned with CamScannerypasires oF DQ = ans Ment = 27.O8ST years, M.D, = 2.96 yours © a. Cateulate mean deviation trom the toa anelits event fe the followin data Cafculateaverage length of illness and corresponding M.D. ‘Ans. 24165 days, 13.64 days —_—_—_—_—_——__—___—— 6. STANDARD DEVIATION arithmetic mean of the squares of the Itisdefined as the positive square root of th tic mean, It is denoted by 6 (sigma). deviations of the observations from the arithme Itisan absolute measure. (a) For Ungrouped Data x, are n observations then x ) For Grouped Data are n distict values of a variable with frequencies f, Scanned with CamScanner= jg integer 2nd (ii) sed in any case. But ative formula ti ities for value of z. They are (i X is not an integer ne of the two formulae can be us usually, the first formula is used when = is an integer and the altern is used when x is not an integer, to simplify the calculations. Now consider some examples. Mlustration 10: Find standard deviation of the following : ‘ ‘Set 12, 17, 15, 11, 25 Solution : =. WHIT 1S +11 + 25. _ 80 21g Hote, X= As x = 16, is an integer we calculate standard deviation (9) using the formula | _ eaw- 2? ¢ : | n Consider 3 (x- x)= (12-16)2+ (17 - 16)? (5 167+ (1 - 16) + (25 = 16)? =164 141425481 = 124 = Pad _ a JAB = 49799 | 1 = Hence the standard deviation is 4.9799. i Iustration 11 : ’ Find standard deviation for the following data : 21, 16,13, 11, 9, 14, 8, 14 Solution : ide Consider X = — ‘ n i ® FG +IZ+ N+ 9 +14 +8414 _ 106 _ 4325 ae 8° As % =13.25, is not an integer, we use the following formula to calculate stam deviation. Re " ; L re Scanned with CamScanner‘Measures of Dispersion ‘ : 135 x 2 cs - ys 2 Now, “ se = (21)? + (16)? + (13)? + (11)? + (9)? + (14)? + (8)? + (14)? = 441 + 256 + 169 + 121+ 81+ 196 +64 + 196 = 1524 2 ps 1524 Sa ges (2A 2 , 7 gee neae = {190.5 - 175.5625 = 414.9375 = 3.8649 Hence, the standard deviation is 3.8649. Jlustration 12 = . Find the standard deviation for the following distribution. remota [7 * 19 | fie orrewons [50 | Solution : To find =, prepare the following table upto fx. Here the size of shoe is the discrete variable x and Size of Shoe x 7 8 9 10 11 no. of persons is the frequency f 50 ‘As x isan integer, use the following formula. 2 ou pad E a Tofind Ef (x - H* -Prepare two £ (x - 3)”. in the above table. 2 = 60 and N=50 additional columns of (x - %) andthe product Now, Ef (x - X) Scanned with CamScannerSolution : Tt is a discrete variate case. To calculate x, prepare the _ product term fx, following table for the Ek _ 716 g- = = = 10.08 Now gy cl | ‘As x is not an integer, we use the alternative formula for calculation of standard deviation viz. : Sox? a : To find 6, prepare an additional column of fx? in the above table. Note that fr can be obtained as the product of f and x? or as the product of fx and x. We 2?f" the second way and obtain fx? as fx multiplied by x because fx values are al calculated. ‘Now, from the table, Efx? = 7380, N= 71, x = 10.08 Substituting these values in the formula Ei? 22 _ [7380 ee age a - 2 7 5 7 (10.08) 5 % o= Scanned with CamScannerpasures of ‘Dispersion 137 | ffgn545T - 101.6064 = 2.3373 = 1.5288 ndard deviation is 1.5288. ence the st 14: yustration d deviation for the following data : ind standar' 40-50 interval 12 able is a continuous variable so x represents the class marks of the class The vari ex values are 5, 15,25, 35 and 45. intervals i.e. X ble for the product term fx, fx? Prepare the following tal to obtain mean and SD. _ rT 215 15 3375 15625 = 23.4286, N= 70 Now, Uf? = 48150, X in the formula Substituting these values, Ex? N - 23.4286)" - = 687.8571 - 548.8993 = 138.9578 =. 11.7880 * So, the standard deviation is 11-788- 48150. 70 Scanned with CamScanner138 Illustration 15 : 4 Find standard deviation for the distribut Fwragmrines [2005 [a0 [3035] He ofcondeal | 16 ana Solution :. ne ait We prepare the following table to calculate the product terms fx ate aes jon of weights o x Six = 3010 and N=90, B2= 1041125 = - EH 300 = 53.a4as N ~The standard deviation is given by sik? s= {= -z N . fiostaas - 33. 2 90 = 4f/1156.8055 - 1118.5278 = 438.2777 =-6.1869 Hence, the standard deviation is 6.1869. It.can be observed that if the class marks are expressed as fractions or if the class intervals are more, the calculations become lengthy and tedious. The working ca be reduced by change of origin and scale as follows. q Short-Cut Method or Change of Origin and Scale ~s wherea and c are any constants. Ustsl! ervale We define anew variable wasu = ~ ais chosen as the middle-most class mark and cas the length of the class int The values of u can be calculated easily from the values of: f t : 1 they a" the form -3, -2, -1, 0, 1, 2, 3, ....and so on. abe In gener OY The arithmetic mean and standard deviation are given by the following rest!"* Scanned with CamScannerures of Dispersion af Disp 139 e + <(28) Webave prresey’ ra the above formulae in the previous chapter to calculate the mean. fstandard deviation of x is denoted by ox and standard deviation of u is denoted by Ou, then ox=cou or S.D.ofx=ox=cx Zfu If we substitute ¥ = yp then ox = © x [ustration 16 : Find the mean and standard deviation for the following data representing the distribution of no. of units consumed by 200 consumers. No. of Consumers 0-500 2 2 500-1000 18 1000-1500 27 1500-2000 32 2000-2500 35 2500-3000 38 3000-3500 30 3500-4000 il Solution : - ‘As the number of class intervals is more, We will use the short cut method. We prepare the following fable containing columns for x- class marks - 250, 750, 1250, 1750, 2250, 2750, 3250, 3750 u=new variable values f= product of frequency and u value fi? = product of fu arid u value wa _ x- 2250 ==, i 500 re a= 2250, one of the middle clas | The corresponding values of w for various values of are We define u = smarks, c= 550, width ofthe class intervals -4, -3, -2, -1, 0, 1, 2, 3. Scanned with CamScanner500-1000 1000-1500 1500-2000 0 r | 2000-2500 2500-3000 3000-3500 3500-4000 Here, N=Zf= 200, a= 2250, e= 500, Efu=-45, Efu? = 703 j oS Si 45 Now 8 = gop 7 075 sata = 2250 + 500 (-0.225) = 2250 - 112.5 = 2137.5 | : | 2 _. pefu? > SD.ofx =e - 8 703 2 = 500 x 4J— - (-0.225 ‘ 500 * 999, ~ ©0229) | | = 500 x ¥3.515 - 0.050625 a | = 500 x 43.4644 | = 500 x 1.8613 = 930.65, the mean and the standard deviation are 2137.5 and 930.65 re Hence, spective Iustration 17 : For the following data, find the: standard deviation, Solution : % ‘ ‘The class intervals are inclusive type. So we find the class-boundaries #S ae We have difference between lower class-limit of the second class inte™ upper class limit of the first class interval as 1200 - 1199 = 1. - Scanned with CamScannerfeasures of Dispersion at F di he faction to be added or subtracted = THEE 9,5, So, the cass bounds are calculated by adding 0.5 to upper class limits and subtracting 0.5 from lower class limits and the class boundaries are 399.5 - 799.5, 799.5 - 895.: and so © on. The following table is prepared to calculate fu and fu. fu? ‘Amount No. of Class Class fu int Workers | Boundaries | Marks Bs x 400-799 100 399.5-799.5 599.5 800-1199 150 799.5-1199.5, 999.5 1200-1599 200 1199.5-1599.5 | 1399.5 1600-1999 275 1599.5-1999:5 | 1799.5 2000-2399 150 1999,5-2399.5 | 2199.5 2400-2799 80 2399,5-2799.5 | 2599.5 2800-3199 45 2799.5-3199.5 | 2999.5 1000 Note : a is chosen as 1799.5 and c- the width of class interval is 400 and not 399. Here N = Zf= 1000, a = 1799.5, ¢ = 400, Efu = -355, Df? = 2575 Efu 355 my ee ee 20,355) Rew e oN 1000 : 2 SD.ofx = ef : N = 400 x [2225 - 0.385)" 1000 = 400 x V2575 - 0.1260 = 400 x 2.449 = 400 x 1.5649 = 625.96 ‘Hence, the standard deviation is €.625.96. Algebraic Properties of Standard Deviation “(The sum of the squares of deviations from mean is minimum. | (i) The standard deviation is always greater than or at the most equal to mean deviation from mean. (ili) Ifthere are two groups containing n, X, and standard deviation 6, and ©, respectively, 6 of the combined group of ny + My is given by and n, observations with mean X, and then the standard deviation Scanned with CamScanner142 where dy) = Fy «Red = Ky K and ¥ fsthooombined snean onto" samples, given by the formula mye + yy, g- nip my ty Ilustration 18 + There are (wo groups of children having 50 and 70 children respectively: ' yy arithmetic means of weights of the children in the two groups are “9 keys anid 40 ind the combined kgs. with standard deviations 1G kgs, and 5 kgs. respectively, "i mean and standard deviation of the entire group ‘dontnining, 120 entre: Solution + The given information can be written as Group} n= 50 %, = 30 ys a,= 6 keys. 7 19 follows + group IE Ng ” hy * 40 kgs. 0,3 5 keys No.of observations Mean sp We calculate combined mean = nyhy * Dah, my M2 7 50% 30 + 10% MO 1500 4 2600 a5 e3kys, » 50% 70 420 ’ = Fy %* 30 « 3583 = “5B, = M0 - 35.83 = 457 rd deviation 018 given by Now, dy dy = Fy °F ‘the combined standa In, (or + dt) 4 1 (4 4 3) ny Ng 43)") + 1005 + (417)") 504 10 2 536 & 33,9989) f TW25 4 V13HB9) 7) 036 4 (Be ) Scanned with CamScannerres of Dispersion 18 yoo ee [50 (69.99989) + 70 (42.3889). pe 120 yp9.945 + 2967223 _ [6467168 a 120 ~ ¥" 120 2 453.8931 = 7.3412 kgs. he combined mean =35,83 kgs and the combined standard deviation = are of the standard deviation is known as the Variance or the square ‘Note The sau ‘ance is called standard deviation. © oot of the Vari k “thus, Variance = (0)° pases 7. COEFFICIENT OF VARIATION ig, to standard variation is the relative measure correspondin} - The coefficient 0! /, and itis expressed as percentage. © deviation. It is denoted by C. i Standard Deviation 4 99 Mean ‘two or more distributions: s calculated for the distributions -e consistent. On the or less It is used to compare variability or consistency of! ‘Tocompare variability, the coefficient of variation i and the one with less coefficient ‘of variation is said to be more ~ other hand the one with more co¢) ficient of variation is more variable | consistent. - IMustration 19 = "The following are the ru © Decide who should be selected for Runs by batsman A : 56, 58, 60, 62, 59 Runs by batsman B : 70, 62, 50, 35, 69 ~ Solution : : _ Wefind the means, stand: ~ AandB. © @ -ForA, 3x=295,n=5, x= 59 L(x - =941414940=20 ns scored by two batsmen A and B in 5 test matches. the coming England tour. jard deviations and finally the coefficients of variation for + 2 Scanned with CamScannerMean. %y = 76.5 Ky = 76,25 144 ah sp = =" Se Sey n 3 sD Coefficient of Variation = =~ % 100 2x 100 = 2% 59 59 =572 , (b) For batsman B, =x = 286,n=5, % x2 = 4900 + 3844+ 4761 +2500 + 1225 = 2 sp =o= ea . n ee 7220 « (s13y" = {3446 - 3271.84 = 174.16 = 13.1970 = SD x 100 Coefficient of Variation x 17320 37. ents of variation. than GV, for A= 3.3898 stent than batsman B Bue 100 = 23.0716 So, comparing the coeffic C.V, for B= 23.0716 is greater Hence, batsman A is more consi for the England tour. Illustration 20: The mean marks of two groups of students in a certain respectively with the standard deviation 16.97 and 5.82 for the two groups: which group is more consistent or uniform, so he should be selected test are 76,5 and 16. vind Solution : We have the following information : Group I Group II SD 6,=1697 = 5.82 CV. for Group! = 2b x 100 = 16.97 = = x 100 H i Fens to eee CV. for Group tt = 22 x 100 = 5.82 a saog % 100 = 7.6328 Scanned with CamScannerures of Dispersion 145 the values of coefficients of variation, as 7.6328 is less than 22.183. group js more uniform. Merits of Standard Deviation @_ Itisrigidly defined. @. Itis based on all the observations. G Itisnot affected much by sampling fluctuations. Gv) tis amicable to further mathematical treatment. Demerits (i) Itisnot easy to understand and easy to calculate. (i Asitconsiders the sum of the squares of deviations of items from the mean, the items away from the mean get more weight than those near the mean. So the standard deviation gives more weight to extreme values. ‘Absolute and Relative Measures of Dispersion We have seen Range, Quartile Deviation and Standard Deviation as absolute measures of dispersion. These are expressed in the units in which the original variable is measured. : We have also seen the corresponding relative measures of dispersion sucit as coefficient of range, coefficient of quartile deviation and coefficient of variation. These are expressed as ratios so that they are pure numbers, independent of the unit in which the original variable is measured. This is precisely the reason why they are used in comparison of two or more distributions. In some situations, the comparison of two distributions of different variables is ~ required, For instance, if weights and blood pressure are considered for a group of persons to find whether the weights are more variable or the blood pressure, then the main problem is that standard deviation of weights is expressed in kgs and standard deviation of blood pressure is expressed in pure numbers. Because of the difference in units of measurement, the values can not be compared. Soa relative measure is required. ‘We may have two distributions with the same units of measurement. For instance, * we may compare the salaries of the employees of two different firms. Let the averages be % 1580 and & 925 respectively for the two firms and let the standard deviation be % 250 and 175 respectively. Then we can not directly conclude that there is less variability in the second firm as the standard deviation 175 is less than in the first firm, i.e. % 250. Here the relative degree of variation is to be Considered by calculating the coefficient of variation. CV. for the First Firm = 22% x 100 = 5.82 1580 + CV. for the Second Firm = z x 100 = 18.92 Scanned with CamScanner146 Business Statistics (RY. As coefficient of variation is more in the second firm, the variabilityis second firm, Thus, we can conclude that relative measures must be use two or more distributions. thom Some additional problems are solved below. Illustration 21 : fo groups: The arithmetic means of marks of tw 4 deviation of the second group i * 60 and 45 respectively. The standar‘ combined standard deviation of the 300 students is 69 . Find standard deviation containing 100 and 200 students are is 4 and the of the first group. Solution : We have the following information about the two groups : Group I Group IL No.of observations n, = 100 n, = 200 Arithmetic Mean Ky = 45 marks sD nyky + ngXp ny + M2 Consider * = Substituting the values, 100 x 60 + 200 x 45 or 100 + 200 6000 + 9000 _ 15000 -, $000: 9000 = 50 300 300°, So combined mean = is 50. 4, = %, -% = 60-50 = 10, = 45-50 = -5 dy = %- Consider 2 ay dt ny(o? + dp) + (09 + 42) ny +p. Here all values are known except 0). So substituting the other values, 100 +200. * Scanned with CamScanner147 | quaring both thesides, 100(67 + 100) + 200(41) : 300 “ross multiplying and simplifying, 1006, + 10000 + 8200 20700 = 1000,3 + 18200. “= 20700 - 18200 = 1000, 22500 = 1000, 2500 = —— = 25 ot 100 OSs oct So, the standard deviation of the first group is 5. Illustration 22 : : ‘An analysis of the monthly wages paid to the workers in two firms A and B, belonging to the same industry, gives the following results. FirmA FirmB No. of Workers 550 650 ‘Average Monthly Wages % 500 2450 Standard Deviation of Wages: © 90. & Ji20 Answer the following questions : (@ Which firm A or B pays out larger amount as monthly wages ? (i) In which firm A or Bis there greater variability in individual wages ? (iii) Find the combined mean and standard deviation ofall 1200 workers of the firms taken together. Solution : Here n, = 550, n, = 650, ¥, = 500, ¥ = 450, 0,= y90,0,= Vi20 () Tofind the total wage bill for each firm; we use the property that the product of mean and number of observation gives the sum of the observations. ice. total wage bill of firm A= n,X, = 550 x 500 = 275000 and total wage bill of firm B= nyX7 = 650 x 450 =% 292500 _ So, the firm B has larger wage bill, since the amount of % 292500 paid by the firm B is more than the amount of 275000 paid by firm A.” To compare the variability of the wages of the firms A and B, we calculate the coefficients of variation. Scanned with CamScanner148 = PP. 199 = 24868 x 100 : = = 1.8974 Coefficient of Variation for Firm B oy == x 100 =. 120 10.954451 x 100 = x 100 = = = a 0 2.1909 Hence by comparing the coefficients of variations, 2.1909 is greater than 1.8974. -. there is greater variability in wages of firm B. (iii) To find combined mean and standard deviation, consider 1% + 05%) = 550 + 650 __ 275000 + 292500 _ 367500 _ gry 99 1200 1200 00 - 472.92" = 27.08 = 450 = 472.92 = -22.92 2. z, ny 7 Ba | 350(90 + (27.08)") + 650(120 + (22.92)) ay 550 + 650 550 (90 + 733.3264) + 650 (120 + 525.3264) < 1200 559 (623.3264) + 650 (645.3264) : 1200 952829.52 + 41946216 _ f= [452829.52 + 41946216 _ JP O” = 1200 1200 2 76,9097 = 26.9613 = 26.96 Scanned with CamScannersures of Dispersion 149 the combined mean and standard deviation are % 472.92 and & 26.96 respectively: - | uustration 23 : © The mean and standard deviation of a group of 100 observations were found to be gqsand 18 respectively. Itwas afterwards detected that one observation was wrongly jaken as 486 instead of 468. Find the corrected mean and standard deviation. Solution : © wehave n= 100, x =475, S.D. c= 18 Now, DX =n = 100 x 475 = 47500 s = 2x - wrong value + correct value = 47500 - 486 + 468 = 47482 Corrected x _ 47482 Corrected Mean= = T= = 474.82 Hence, Corrected Mean = 474.82. 2 Now, 6 = y= 7 Substituting in the equation the given values of mean = 475 and $.D.= 18, wehave 2. I _ (475, 109 7 squaring both the sides, 2 324 2. (475)? 100 xe 324 = =— =.225625 100 2 =x = + 324 = 225949 100 225625 3. + 2x? = 225949 x 100 = 22594900 Corrected Sx? = Ex? - (wrong value)? + (correct value)* = 22594900 - (486)? + (468)? = 22594900 - 23196 + 219024 = 22577728 Scanned with CamScanner150 ‘Now, the corrected standard deviation is z Corrected = [Corrected 2x" _ (Corrected x)” a 225777: [22577728 _ (474.82)? 217.9792 100 = {22577728 - 22545403 = 4323.25 data are 4748 Hence the corrected mean and standard. deviation of the respectively. We have seen meas and Mode. We have al Mean Deviation or Stan Now let us study the s Skewness and Kurtosis. Mi sures of Central Tendency oF Location such 45 Means ti go seen Measures of Dispersion uch as Quartile Devise dard Deviation. d ‘of concepts like shape of the distribution with the help 2 8. SKEWNESS ution has the values of all 3 measure: nd its mean i.e. the right and the left parts ‘A symmetric distrib symmnetrical arow identical. Its curve is s (ails) of the curve are exactly same. On the, contrary, skewness is 2 measure of the asymmetry of the probability ec of two types Positive Skewness and distribution of 2 sandom variable. Itcan Negative Skewness. If the tail on the right side of the _ indicates Positive Skewness. On th curve is longer or flatter than its rig) There are two measures of skewness. or flatter than the left side, it f thie tail on the left side of the Negative Skewness. curve is longer e other hand, it nt side, it indicates ji) Karl Pearson's Coefficient of ‘Skewness : Mean - Mode_ _ 3 (Mean -— Median) Standard Deviation Standard Deviation (ji) Bowley's Coefficient of Skewness 3 Ou Median) - (Median - 21) _ Qs + Q, — 2Median (Q3.- Median) + (Median — Q)) 0,- 2 For a symmetric distribution, the above coefficients are zero. ed distribution, coefficients are > 0. For a positively skew ed distribution, coefficients are <0. e Foranegatively skew Scanned with CamScanner- Positively Skewed Distribution Negatively Skewed Distribution ner ie BE 9. KURTOSIS (A eisai te sziisics Kurtosis is the concept associated with the ‘peakedness’ of the sectility distribution of a fandom variable. Some jdea about width of the peak fine dicribution can be obtained by observing the shape of the frequency curve. |A Geribation with zero excess kurtosis is called 2 *Mesokurtic' distribution. ¢2- Normal probability distribution. : isis called a "Leptokurtic' distribution. Ithasa very acute peak around its mean. : . ‘Adiscibation with negative excess kurtosis is called’s ‘platykurtic' distribution. ‘Hhesa more wide and flat peak around its mean. Telooks like plateau, representing flat area at the top of a mountain. Leptokurtic Curve Mesokurtic Curve Platykurtic Curve Scanned with CamScanner152 Its measure is Karl Pearson's moments. Ba, = ‘ where; i is the fourth moment about mean and es ois standard deviation B, is also known as the fourth standardised moment. — EXERCISE Write the merits and demerits of standard deviation as a measure of dispersion. 1. 2. Explain.the need of defining relative measures of dispersion. 3. Why relative measures of dispersion are used ? 4. Describe the uses of different measures of dispersion. 5, Distinguish between an absolute measure and arrelative measure of dispersion. 6. ‘Find the standard deviation for the following sets : (i) 15, 20, 17,8, 9; 12, 18, 10 (ii) 652, 672, 670, 639 642, 670 ‘Ans. (i) sd = 4.2112 (ii) sd = 13.7568 7, Find the standard deviation for the following sets = (() 85, 35, 43, 75, 42, and 41 (li) 52, $7, 49, 48, 35, 37 ‘Ans. (i) sd = 19.1289 (ii) sd = 7.867 : 8, From the following distribution, find its standard deviation. Ans. sd = 1.5657 9. From the following distribution, find the standard deviation. Ans. sd = 17.1890 10, Find mean and standard deviation for the following : No. of Tickets No. of Persons Ans, 12.2, 2.242 Scanned with CamScannersures of Dispersion 153 _. From the following data find coefficient of variation. 15-20 [20-25 [25-30 [30-35 |35-40 13, | a [16 | 8 3 No. of Students ‘ans. 36.52% 2. Findmean and standard deviation of ages. 10- | 10-20 | 20-30 | 30-40 | 40-50 | 50-60 7. 12 10 ‘Age in Years No. of Persong 27.89 years, 12.546 years Ans. 13, -Find the standard deviation. Marks: 5-15 15-25 | 25-35 | 35-45 | 45-55 55-65 No. of Students 7 at 12 20 14 8 2 Ans. 12.75 14, The daily wages of 69 workers are given below. Find the standard deviation of wages. 70-80 | 80-90 Daily Wages | 30-40 ing No. of Workers Ans, sd =% 13.61 15. Find the standard deviation of the. following data. Also find the coefficient of variation. Class Interval 120-140 13 4 Frequency Ans, sd = 28.4345, CV = 36.5988 16. Find the standard deviation and coefficient of variati data: ion for the following 3000- 3500- | 4000- es Daily Collection 2500- in (00) 3000 3500 4000 4500 5000_| 8 [ No. of Agents 1 15 is [ 4 } Ans. sd =% 6280.4106, CV = 16.99 Scanned with CamScanner154 e Business Statistic “) ; les mM 17. The following data gives returns, expressed in percentages, of investments A and B over a period of 7 years, nee yes # DOE consistent return ? (use coefficient of variation) nt OP Type A 18 | 13 ; Type B 15 | 22 Ans. Type A is more consistent. 18. Find in which of the following subjects, there is more variation in marks- [57 [27 [ot | 39 | [2 [is [78 | 70 | Ans. V,, = 65.76, Vi, = 54.05. -. Subject A is more variable, 19, Find coefficients of variation for the following sets representin two groups of students. Which group is more consistent? (i) 85, 83, 87, 90, 65, 75, 57, 70 (li) 84, 83, 72, 79, 75, 10, 67, 80 ‘Ans. (i) CV = 14.3494 (ii) 7.6401, second group is more consistent. 20, The mean and standard deviation of a group of observations are 25.5 and 10.87 respectively. For another group of observations of the same type, the: mean and standard deviation are 37.5 and 4.89. Which group is more variable? ‘Ans. Group I is more variable. D1, The following is the distribution of weights in gms of articles produced att factory. Find the mean and standard deviation. No. of Articles ‘Ans. Mean = 298.70679 gms, SD = 4.5 gms 22.- The following is the data representing profits in thou: companies. Find coefficient of variation. | Subject A Subject B g marks of sands of & of some Ans. CV = 16.3486 a | 23. The distribution of payments toa number of salesmen is given below fi i standard deviation and coefficient of variation. Scanned with CamScannerfeasures of Dispersion 155 Paymentin® | No. ofSalesmen | Paymentin® | No. of Salesmen F 100-120 4 200-220 . 30 \ 120-140 10 220-240 32 ¥e 140-160 16 240-260 3 we 160-180 29 260-280 17 180-200 52 280-300 7 "ns, Standard deviation = 39.70, CV = 19,3721 ‘4, Find coefficient of variation for the following data : ‘Amount | 500- | 600- | 700- | 800- | 900- [1000- | 1100- | 1200- int 699 | 799 | 899 | 999 | 1099 | 1199 | 1299 No. of a | 355 | 70 |.62 | 50 | 35 | 1 Workers ‘Ans. CV = 20.8793 , 25. Find coefficient of variation for the following data’: Twe in ke 50-54 | 55-59 | 60-64 | 65-69 No. of Persons| 3 10 15 Ans. 9.49 , 26. The following are some particulars of the distribution of weights ofboysand girls in aclass. Find the standard deviation of the combined group. Ans. 6 = 7.57 kgs. 27. Find the combined mean and s.d. for the following : Number Mean Weight S.D: Also find coefficient. o! consistent. Ans, = = 164, 6 = 6.03, CV are 2.94, 1.25, so female are mor f variation for each group and decide which is more e consistent. Scanned with CamScanner_156 ‘ Business Statistics (F.Y.B.) 28. Find combined mean and standard deviation. Also find the grot variation. 70 90 15 82 wea 7 83, CVI = 5.33, CV2=8.54, so group ILis mor h mean 15 and ean 15.6and SD vi Number Mean Weight SD. variant. standard Ans. % = 78.9375, 0 29. The first of the two samples had 100 items wit deviation 3. Ifthe whole group has 250 items with m find the standard deviation of the second group. Ans. SD =4 Number ‘Average Daily Wage sD. ‘The standard deviation of the combined group of 900 workers is «f37 « Fin the standard deviation of the second group. Ans. & 3.2557 31, The first of the two groups has 100 items with mean 45 and standard deviation 7. Ifthe combined group has 250 items with arithmetic mean 51 and variance 130, find the mean and variance of the second group. é Ans. 55, 12 32, ‘Themean and stan ‘as 300 and 25 respectively. It w wrongly considered as 290 and 30: mean and sd. Ans. 301, 25.18 33. Calculate th items from the following. . [| eon | No. of Items 50 js were | calculated dard deviation ofa group of 30 observation: ns were as later found that two observatio '5 instead of 305 and 320. Find the correct” ‘e mean and standard deviation of the combined grouP of 200 Ans. Mean = 225, SD = 19.1572 Scanned with CamScannerires of Dispersion ‘Measures of Dispersion ve 4, Calculate the standard deviation of th Fie following date, © combined group of 300 items from’ 200 87 9 ‘Ans. SD= 11.9 \ 35, Find mean, standard deviation and coefficient of variation for the following data. [Height in ems 100-110 | 110-120 | 120-130 | 130-140 | 140-150 [[No. of Children 8 10 15 3 2 Ans. Mean = 120.75 cms, sd = 10.93 cms, CV = 9.0507% © 36. Thearithmetic mean and the standard deviation of 100 items are found to be 40 and 10 respectively. Ifat the time of calculations one item was wrongly taken as 30 instead of 3. Find the correct mean and correct standard deviation, ‘Ans. Corrected mean = 39.73, Corrected sd = 6.3732 37. The arithmetic mean and standard deviation of a group of 200 items were 150 and 19 respectively. It was afterwards found that one item was wrongly considered as 155 instead of 125. Find the corrected mean and standard deviation. ‘Ans. Mean = 149.85, sd = 19.0782 38. The values of mean and standard deviation for 50 observations were 475 and 25 respectively. It was observed that one item was wrongly considered as 400 instead of 500. Find the corrected mean and sd. Ans. Mean = 477, sd = 22.8254 : 39. Themean and standard deviation of group of 25 observations were 42 and 8 respectively. Two values were wrongly recorded as 25 and 20. Find the corrected mean and standard deviation after deleting the wrong values. Ans, Mean = 43,6957, sd = 5.7513 ; 49. The mean and standard deviation of a group of 30 observations were respectively 125 and 11. It was afterwards found that two observations were - "wrongly considered as 105 and 107 instead of 135 and 137 respectively. Find the corrected values of its mean and sd. Ans. 127, 10.049 - 41. The mean and standard deviation ofa group of 30 observations were 93 and ‘ Trespectively. It was detected that three observations were wrongly taken as ' 89, 65 and 73. Find the corrected mean and sd after deleting wrong values. © Ans. 94.106, 5.044 i Scanned with CamScanner_ 158 workers is 480 and& 726.4. The arithmetic mean of 400 workers out of them is€ 450 and 10. Findt remaining 600 workers. o Ans, Mean = % 500, sd =@ 12 4 43. The coefficient of variation of a group of observation: the mean was 57.2. Find the standard deviation ofthe BroUP- Ans, Siandard deviation = 13.1970 : 4 44, The coefficient of variation for a group was 30.908 and mean is 13.625. Find the standard deviation. Ans, sd= 4.2112 45. The standard deviation and coefficient of variation of __ 13.7568 and 2.0923. Find the mean. ‘Ams, Mean = 657.4965 46. The following data gives the means a - workers. s was 93,0716% and a distribution are rd standard deviations oftwo groups of (Which group has a larger wage bill ? Gi) | Which group is more consistent 2 (iii) What is the combined mean and standard deviation of all the workers taken together ? ‘Ans. (i) Group II (ii) Group I (ili) x = 2480, 0= 226.951) 47. The following is the age distribution of 125 persons. Find tl standard deviation. ‘Age in Years No. of Persons Ans. Mean = 48, Find the coefficient of variation for the following : Salary in® 1000-1199 ]1200-1399| Ans, x = & 1463.3, s=% 231.02, CV= 15.8% he mean and 5,02 years, sd = 19.76 years Scanned with CamScannerof Dispersion ges of 159 4 kgs. jard deviation and , For the following distribution of weight the arithmetic mean is 16. "Find the missing frequency. Hence calculate the stand 9 coefficient of variation of the distribution, 17-19 19-21 No. of Children i “gps. 28, 24658 ke, 15.0352% | g0, Ifthe mean for the following data is 56, find the unknown frequency. Also calculate its coefficient of variation and sd, [Wages int | 30-40 No.ofPersons| 10 50-60 | 60-70 | 70-80 | 80-90 40 16° 8 6 ns. 20, % 12.69, 0.0226% 51, The following is the distribution of marks of 80 students in a class, Find the standard deviation of marks. Also find the coefficient of variation. Marks below 50 40 30 | 20 f 10 No. of Children 80 65 46. 25 12 Ans.-sd = 13.1434, CV = 49.59% 52, Find the standard deviation and coefficient of variation for the following data, Marks No.ofStudents] 16 | =9.1072, CV =31,902% 53. Find the standard deviation and coefficient of variation for the following | Ans. SD = 266.294, CV = 18.0752% 4 Find mean and standard deviation for the following data. 65-170} 170-180 f 15 5 Scanned with CamScanner» SD = 5.7589, CV 3.7476, - C= 6, A= 136.5, the class intervals are 109.5-115.5, 115. The following is the distributi the standard deviation and coeffi ‘fue values of arithmetic mean and standard deviation of the followin frequency distribution of. continuous variable obtained by change of or and scale are 135.3 sind 9.6 respectively. Determine the original class interval uw 3 | 2-1 0 1 2 fees" | 3 [as-| 22 [3-8 157.5) MULTIPLE CHOICE QUESTIONS Choose the correct alternative from the following : The suitable measure of dispersion to indicate extreme variations in thed is (a) Range (b) Mean Deviation (©) Both (a) and (b) (@)_ None of these ‘The sum of the absolute deviations of observations from the following is minimum (2) Mean (b) Median (©) Mode . -@_ None of these The following measure of dispersion considers middle 50% of observatios (a) Range (6) Quartile Deviation (©) Standard Deviation @) None of these The following measure of dispersion cannot be calculated in case of end class intervals. (2) Standard Deviation (6) Mean Deviation (©) Both (a) and (b) . @)- None of these wl The combined measure for two or more groups can be calculated {°° following measure of dispersion. (a) Mean Deviation (&) Range (c) Standard Deviation (@ None of these Scanned with CamScannerFant of Dispersion 161 JL values in a set of 20 observations are identical then its standard deviation Ital is ot (b) 0 @ 035 (d) None of these fall values in a set of 11 observations are identical then its mean deviation from mean / median is @ 1 /() 0 @) 05 (@) None of these Tocompare two or more sets of variables with different units of measurement, the following measure of dispersion is best suited. (a) Coefficient of Variation (b) Quartile Deviation (o) Range (a) None of these 4, Ifthe value of coefficient of variation is more, the consistency of the data is (@) more (b) less (©) same (a) _ none of these 10. Ifmean and standard deviation of a group of 11 observations are 8 and 4 respectively, the coefficient of variation is @ 2 (b) 20 © 50 (d)_ None of these Ans. (1) =(@), 2) = (6) GY ©), 4)- ©, =O. O- OD -O), @)- @ (9) - (), (10) - (©) . Scanned with CamScanner