Chapter 4 MEASURES OF DISPERSION Basic Terms used in statistics are well known to a common person (1) To develop the skills to analyze the data by finding its representative value. (1) To understand and interpret the distribution of the data. (iii) To analyze the data through measure of scatterings, skewness and kurtosis. (iv) To compare individuals in a data set by Z-score. 4.1 Dispersion The scattering of the values of a distribution of the data from an average is called dispersion. . 4.2 Measure of dispersion A measure of dispersion expresses quantitatively the degree of variation or dispersion of values of a variable about any average. OR any measure that indicates how the observations are spread out from an average is called a measure of dispersion. 4.2.1 Types of measure of dispersion There are two types of dispersion . (i) Absolute measure of dispersion and (ii) Relative measure of dispersion 4.2.2 Absolute and relative measure of dispersion The measures of dispersion which are expressed in terms of original units of a data are termed as absolute measures of dispersion. Relative measures of dispersion also known as coefficients of dispersion, are obtained as ratios or percentages. These are pure numbers independent of the units of measurement and used to compare two or more sets of data values. 4.2.3 Methods of measure of dispersion 01) Range (R) 02) Semi inter quartile range/Quartile deviation (SIQR/QD) 03) Mean deviation/Average deviation (MD) 04) Variance (S*) and Standard deviation (S) 4.2.3.1 Range Difference between maximum and minimum value of the data is called range. It is Given as R =X, — XoMeasures of dispersion Coefficient of range X,4N Example 4.1 Following are the values of a varlable X 5,3, 7, 9, 11, 15, 18, 19, 16, and 20, Find range and coefficient of range, Solution: X: 5, 3, Me 9, 11, 15,18, 19, 16, 20 NV, = 20nd Range = X,,~X,= 20-317 Coefficient of range: Co — R = 0,739 Example 4.2 Find range and coeffcent of a range from the following frequency distribution. Classes | 20 - 30 | 30 ~ 40 | 40 - 50 Frequency | 7 | 12 15 2 Solution: Classes _| 10- 20] 20- 3030 - 40 | 40 - 50 Frequency} 7 | 12 2 \ = upper class boundary of the highest class, ¥, — 50 1 = lower class boundary of the lowest class, V,, 10 =50-10=40 Coefficient of range: Co— R= paar l OA = 0.667 X, +X 50+10 60 Range can also be calculated by using two extreme mid points. 4.2.3.2 Quartile Deviation (QD) / Semi inter quartile range (SIQR) Half of the difference between upper and lower quartiles (Qs and Q:) is called semi inter quartile range or simply quartile deviation (QD). It is given as OD = 2-9 2,+0, Example 4.3: Calculate QD and its‘coefficient from the data Coefficient of QD = 108Measures of dispersion 6, 4, 2, 8, 10, 14, 12 Solution: Arranging the data: 2.4,6,8,10,12,14 n=7 a+ thyalue = Tn value =2" value Q, = 4 3(n+1 (a+) thvalue = 6" value Q, = 12 +1) 4 12-44 O,+Q 1244 — Example 4.4: Calculate QD and its coefficient from the following data Co-QD= Classes 0-9] 10-19] 20-29 | 30-39 | 40-49 | 50-59 equency 5 | 9) [7 15 7 2 Solution: Classes | f | CB of ~ 5 |-0.05-9.5 | 5 9 |9.5-19.5 | 14, | n/4 17 |19.5-29.5 | 31 15 | 29.5-39.5 | 46_, | 3n/4 7 | 39.5-49.5 | 53 2 |49.5-59.5 | 55 h(n), Beret] 9-95-2413 75-5) =19.22 g=1-A{ N\4 Q, 29509 (41 25-31) =36.33 Ob = Q=Q, _ 36.33-19.22 _¢ 55 2 2 109Measures of dispersion QQ MOSS 19.22.— Q.10, MO94 19:22 CO. 0D O31 4.2.3.3 Mean deviation (Average deviation) Mean deviation is defined as the AM of the absolute deviations of the observation fro, their any average (¢nean, median or mode), Cis also known as average deviation ae¢ sy [Xn It ane Coutticient of ap “2 yy m Example 4.5: calculate mean deviation from mean, median and mode and their Is given as MD, coefficients 3, 4, 7, 7, 8, 3, 5, 3. Solutio . x, x| |x- | | je #4 3 2 1S ‘0 4 1 0.5 jd 7 2 25 \4 7 2 2.5 \4 \8 3 3.5 \s 3 2 15 0 5 0 0.5 l2 3 2 15 lo EX 407 yy xf dra] 4 Le - 4] 16 Smarks ” Median: 3, 3, 3, 4,5, 7, 7,8 n=8 X=. Mode: \ =3 Mean Deviations and their coefficients:Measures of dispersion _Sx-8|_14 MP ren = = 1.75 marks ” 8 D_i. CO= MP yy = = HE 20.35 mean 5 x-X| 14 MD jedan = 2x41 => =1.75 marks n 8 D 2 Co-MD, -MP_ _175 _o 3 mertiany median 4.8 MD, we Co- MD, Example 4.6: Compute mean deviation from mean, median and mode from the following data of marks in statistics of BS (H) mathematics. Also calculate coefficients for each. eee | Marks 0-19 | 20-39 | 40-59 | 60-79 , 30-99 Solution: [Marks | |X | rx.-¥] | sy, 10-19 |6 (9.5 | 235.2 | 240 (20-39 | 10 | 29.5 | 192 200 (40-59 | 18 | 49.5 | 14.4 0 60-79 | 12 | 69.5| 249.6 _| 240 80-99/4 | 89.5] 163.2 1160 | yy 50 854.4 |840 | 865.74 Calculations of mean, median and mode discussed in «!\¢ previous chapter are ¥=48.1,¥ =49.5 and X= 50.93 111Measures of dispersion Mean Deviations = Togs MD. = dvb. =3] = Boe = 17.08 marks . vs 50 MD 17.08 Co- MD, 0.35 * mean 48.7 MD, ean) = 16.8 marks - i 0 CO= MD stun = ae = 168 _ 0.34 median 49.5 Q 1 N MD oo. = ae = 865.74 _ 17 31 marks MI 7. Co MD, eva! mode 50.93 4.2.3.4 Variance . Variance is defined as the AM of the squared deviations of the observation from thei mean. It is given as Yu -xF BS — For ungrouped data Dixy, - XY Sa > a For grouped data Direct method EE (ES). 0 Ree (Re) 112Measures of dispersion Short cut method . Nr pS py . Sp y=] S|]: Ss ve i Coding / Step deviation method Ne Be (Eel le, sae s n 4.2.3.5 Properties of variance (i) Variance of a constant is zero, Var(c) = 0 (ii) Variance is not affected by change of origin, that is If y- V+ & Then Var )- VarCX) (ili) Variance is affected by change of scale, that is if Y= «V then Vari) = ar Var) (iv) VartX#Y)-Var(X)+ Vary) if X and Y are independent (v) Combined variance: If a distribution consists of k components with 1,1, 1, Observations with Sin, =n having means * x, and variances S; Sj, then the combined Danse, ~ variance * of all n observations Is given by 5, Py Xb te ty (87 +X, — X97} Where 1g byMeasures of dispersion 4.2.3.6 Standard deviation Standard deViation (SD) is defined as the positive square root of AM of the squareg deviations of the observation from their mean. It is given as for ungrouped data y , for grouped data 4.2.3.7 Properties of SD: W) SD(C)=07 Where C is a constant. G2) SD(aN )=JalSD(.x,), Where @ is a constant. (ii) SDUN, #6)=SD(X) (iv) SDUX, +¥) = (Var(V)+Var(¥) () SD(X,-¥) = ¥ar(X)+Far(y) ; Provided Vand ¥, are independent variables, 4.2.3.8 Coefficient of variation Coefficient of variation (CV) is defined as ratio of SD of a variable to its mean (.Y), 2 x100 expressed in percentage. i.e. CV = x 4.2.3.9 Coefficient of standard deviation . Coefficient of standard deviation is defined as ratio of SD of a variable to its mean (F) ie. Co-efficient SD =< 4.2.3.10 Uses of Dispersion It tells about the reliability of a measure of central value. It makes possible to compare two sets of data about their variability,. Measures of dispersion 4.3 Z-scores / Standard scores ' The value of a normally distributed random variable that has been standardized to have a mean of zero and a SD of one by the transformation z= X-# is called Z score, for o sample data z =X.=* | A °2" score is the number of standard deviations an s : observation is above or below the mean. It gives measure of dispersion of an individual observation in standard deviation unit. Zscores indicate the direction (+/-) and time of standard deviations away from the mear that a particular datum lies assuming % is normally distributed. It helps to indicate outlier. A value with Zscore greater than 3 or less than -3 is considered as an outlier/extreme value. A value with Zscore greater than 2 and less than 3 or less than -2 and greater than -3 is considered as influential value. Example 4.7: Mean number of times a child speak is 12 and the standard deviation is 4, find Z score of a child who spoke to other children 8 times in an hour. Solution: ,=12, o=4and X = 8 then ~1.0 It means child is normal Example 4.8: Student's Score, Class Means, Class Standard Deviations, and z Scores on four different courses are given as under Subject Raw Score Mean SD_ 2Z-Score English 85 70 14 +107 . Statistics 57 63 12 -0.50 Mathematics 65 72 16 -0.44 Economics 80 50 15 +2.00 As above table shows that Z-Score in economics is highest i.e. +2.00, it is concluded that level of the knowledge in economics of this particular student is high related to his other class fellows. Z-Score in statistics is lowest i.e. -0:50,"it is concluded that level of the knowledge in statistics of this particular student is low related to his other class fellows. 115Measures of dispersion fixample 4,9: Calculate variance, SD and CV from the following data: 5, 10, 20, 20, 25, 29 and 1 Solution; . x, x? 5 25 10 100 20 400 20 400 25 625 29 841 31 961 “yO 1 SY) = 3352 j LY, M0 | YEN? = 3352 | \ wd Oy, Yuya 40)! yo [~ ‘| 4442 (‘) 7a.857 7 7 . ” ” sp; 3 Vis NST RN Coefficient of variation: Cl’ r x 100 ee 100 = 44.4% Example 4,10: Calculate AM, Variance, SD using direct and coding method and CV. 10, 20, 30, 40 and 50 Solution: y 2 Where u, = xre2u 10 | \ \ | ' 50 2500 = leas Viv tsa) SX) = $500 | yMeasures of dispersion Direct Method: YX. _ 130 =30 n 2 (EN) _ $500 (150 5 XE {2A) st. Ge 2 ) <0 900 = 200 lon 5 n \ 5 Coding method: yu 5 + Sx h= 42x10 =30 a ils a | -(3) \uoy- =(3-1)100 = 200 s) SD: 5220-144 Coefficient of variation: 100 = a x100 = 47.13% Example 4.11: Calculate AM, variance, standard deviation and CV from the following data of Weights of animals in Kg. [ Weights | 50 -150 | 150-250 | 250-350 | f| 35 70 Solution: Weights LUX AA, 50-150 | 35 | 100|3500__| 350000 150-250.| 70 | 200 | 14000_| 2800000 250-350 | 105 | 300 | 31500 | 9450000 350-450 | 70 | 400 | 28000 | 11200000 _| 450-550 | 35 | 500 | 17500_| 8750000 _ = 315 94500 | 32550000 = 300 94500 3 117Measures of dispersion =1ISATKg Short cut method: D = X -A = X - 300 weights | [x [D. [| st ee} 50-150 | 35 | 100 | -200 | -7000 | 1400000 150-250 | 70 | 200 | -100 | -7000 | 700000 250-350 | 105 |300/0 {0 0 350-450 | 70 | 400 | 100 | 7000 | 700000 450-550 | 35 | 500 | 200 | 7000 | 1400000 = 315 Qo 4200000 = 3004 2 = 300 315 LAP, ome. 0 yh 3s (315 §° =13333.33Kg” ; § = /13333.33Kg? =115.47Kg X-A X-200 Coding method: u = S eieeriog Weights xu iu: fie 50-150 | 35 | 100] -1 | -35. | 35 150-250 | 70 | 200}0 250-350 | 105 | 300] 1 | 105 | 105 350-450 | 70 | 400} 2 | 140 | 280 450-550 | 35 | 500} 3 | 105 | 315 = 315 315 | 735 118Measures of dispersion belch = 200 378 100 00 Lh y= isan MISATKe Coefficient of variation: 5.47 00 ST a0 - to.d9% wo. cr 4.4 Moments Moments are defined as the AM of various powers of the ) Na. CP the observations: from any value. These are denoted by My/hy My fl otc. fo pupilation data. and MMs MyM, etc. tor sample data. The basic peiets sat Gh observe the symmetry, skewness and kurtosis of a frequency won, 4.4.1 Moments about mean (central The 1" sample moment about mean is denoted by M, Is given as Dexa) u,=+——, For ungrouped data. ” DA (x,-X) m, Sp fi For grouped data, 2s r=1,2,3,4, These moments are also called the mean moments. 119Measures of dispersion Thus ™, .m, = cm, = S* (variance) 4.4.2 Moments about any value “a” (Non central moments): The +" sample moment about arbitrary value “a” is denoted by mis given as Yr -ay YD n = = +— , For ungrouped data, n ” cia 2 an = =——, For grouped data ay Yi groupe r=). These moments are also called the raw moments, thus m)=hm =X -a> XN =a+m 4.4.3 Moments about origin or zero: The ” sample moment about origin or zero is denoted by is given as m' ->———, For ungrouped data , For grouped data; rela 4.4.4 Relations between moments The following relations exist between moments about mean (central) and moments about any arbitrary value “a” / “zero” (non-central), m= my~ ne m, ment + 2m! m, =m, 4mm, + 6mm? 3m"Measures of dispersion 4.4.5 Moment ratios The ratios (h, and ,). are called moment ratios when both numerator and denominator mi mM, are moments. They are defined as }, =. and by =—+ my my Example 4.12: From the given data x, = 5, 6, 9, 8, 4, 10 (i) Calculate first four moments about origin and convert them into moments about mean (ii) Calculate first four moments X = 6 and convert them into moments about mean (iii) Find by and br , Solution: (i) Moments about origin [x, [x [x Xe 5 [25 [125 | 625 7 16 |36 | 216 1296 L9 [81 | 729 | 6561 8 164 [512 | 4096 4 (16 | 64 256 10 | 100 | 1000 | 10000 42 | 322 | 2646 | 22834 2 2646 _ agi; mi = 6 Converting into moments about mean 1m, =0 and my =m —m{2 = 53.6667 -(7)" = 4.6667 121Moasures of dispersion iy mi, ~ Amin + Gmhnt2 ~ 34 = 3805,667 ~4(441)7 + 6(53.6667)(7)? ~3(7)* = 32.667 (ji) Moments about X = 6 mS ms my: Sond X,-6 m Leics! 9 310 61,6667 Converting into pone about mean m=0 my =m mi? = $,6667-(1)° = 4,667 my = mi ~ 3mm +n}? = 15 ~3(5.6667)(1)+2(1) =0 ig = ni, ~ Amn} + Gmnsanj? — 3maft = 61.6667 —4(15)1 + 6(5.6667)(1)? -3(1)* = 32.667 122Measures of dispersion (iil) by and bay , , my i} hens “( nd ang? my Wan? b=} = 109 my henna? Example 4.13: The following distribution gives the weights of 45 cotton bales, Find first four moments about (i) mean and (ti) 40 | Weights(ka) | A847) 4719 | ta ALAN | Abed | ah d7 l Frequency 2 \ Q Ww i WwW Solution: (i) moments about mean Weights (ka) | fi | | 4 hay 2 To} 2 wo Vy} ia weal 9} a0] 400 Abad W]e] avo als Wy aa} ana ane 10 |e} Aeo x ay 1910 1\ v 110 dh vrs » , 4 AM ADT AON NYT OX WT OY 12,8889] "84,0005 HM 2700 | $449,5402 13,4332 | 59.2501 2044000 | 1170,5004 “21,9990 [54,7750 Hhaddo | IZ 1d 1b5. sald | 19749 0.776 0,900 T1Ho | 26,010 AL AON? oA AIA6. ’ 35,550 Pode | daa.no9d | What 2754 lo ltt | 440,052 | 6604,492 yay, vy m =0; meas = MET gon 123(ii) Moments about 40: D.=X.-4 Weights f. XD: fD D Dp Db 35-37 2 136 4 ~8 32__ -128 512 37-39 338-2) 6 12 -24 48 ___3s4t $400 0 © O 0 41-43 10 42.2 20 40 80 160 43-45 11/44 4 44 | 176 704 2816 45-47 10/46 6 60 360 2160 12960 eee 45 110 620 2792 16496 ye ye . Lire 1h Se Example 4.14: Calculate first four moments about origin and convert them into central moments. Classes 1-5 6-10 11-15 | 16-20 21-25 Frequency 7 11 6 4 1 Solution: Classes | fi X | fx | fx | fXxt 15° [713/21 1 63 | 189 | 567 __610 | 11] 8 | 88 | 704 | 5632 | 45056 (iris | 6 [43/78 | 1014 113182 | 171366 ;16-20 | 4 {18 | 72 | 1296 | 23328 | 419904 121-25 1 | 23| 23 | 529 | 12167 279841 { __|29| | 282 | 3606 | 54498 | 916734 124Measures of dispersion Moments about origin: yx wt = ye iw m= fete 3606 _ 4.345 LS 29 roe a i RE SS igo) m= ZA 6734 srg 517 df 29 LS Central moments m,=0 my =m, =m? = 124.345 -(9.724)' = 29.79 mt, =I = 3mnym, + 2m} = 1879.241 ~3(124.345)(9.724) +2(9.724) = 90.78 im, =m, —4mnunt + 6mm? ~ 3m4 = 3101 1.517 -4(1879.241)( 9.724) + 6(124. 345)(9.724)* -3(9 nay’ = 2239.47 4.5 Symmetrical distribution A distribution is said to be symmetrical if the left tail and right tail of the frequency curve are equal. In a symmetrical distribution mean, median and mode are identical in other words its smaller values and larger values dre equal in numbers. 4.6 Skewness A lack of symmetry in a frequency distribution is called skewness. 4.6.1 Positively skewed distribution A distribution is called positively skewed distribution if right tail of the frequency curve is longer than left tail. In this distribution mean > median > mode. 1254.6.2 Negatively skewed distribution A distribution is called negatively skewed distribution if right tail of the frequency 4 is smaller than left tail. In this distribution mean < median < mode. Negatively Symmetrical Negatively Skewed Skewed X¥XoX 4.6.3 Coefficient of skewness (i) Karl Pearson coefficient of skewness: X-8 3(X-X) Sp = = and Sy = ——— -3<5,53 k SD k SD J a (ii) Bowley's coefficient of skewness: . +Qi1-2median Sk = Gst~zmedion -I 0 then distribution is positively skewed. Graphically skewness may be present as follows. ea 126. Measures of dispersion 4.7 Kurtosis Kurtosis is the degree of peakedness of a distribution usually relative to a normal distribution. A distribution having a relatively high peak is called leptokurtic. A distribution which is flat-topped is called platykurtic. A normal distribution which is neither very peaked nor very flat-topped is also called mesokurtic. Following figure illustrate these types of kurtosis, / ne Leptokurtic / Mesokurtic Platykurtic f=" (Population) and 2,="% (Sample) 2 me If Be = 3 then distribution is mesokurtic. If & > 3 then distribution is leptokurtic. If Be < 3 then distribution is platykurtic. Example 4.15: Find skewness for the following data. (i) X =31,59;X =33.17;5 =10.98 (ii) Q, = 5.06; ¥ =9.15;Q, =16.57 (iii) m, =39.91;m, =120.75 127Moasures of dispersion Solution; WN ON) ap so 4447) wo s 109% Data is negatively skewed OQ 2N 16.57 + $.06- 279.15 9 ong OAI2D s OU, 16.57 5.06 Data 15 positively skewed () 5, < fh = ew MAUS 0.a7K9 Data is positively skewed dn, Swor Example 4,16: Find kurtosis for the following data. m, OSX NOM, ATK 29 Solution: m, 47 me (6 11.22, data is Lepto-kurtic ) Grouping error: The difference between numerical values calculated from the original data and the data in the form of a frequency distribution, is called the grouping error, Sheppard's Correction: Grouping error arises in the calculation of moments fron the frequency distribution by the assumption that all values are equal to the mid-g of each class, W. F. Sheppard introduced Sheppard's correction which are given a . follows: ne m,(corrected),= m,(uncorrected) Tz my(corrected) = my(uncorrected) 2 ant h m,(corrected) = m4(uncorrected) — 2 m,(uncorrected) + 240 Sheppard's correction is applied only when (i) Frequency distribution is continuous (ii) Class interval is uniform 128Measures of dispersion Multiple Choice Questions 1. Variance of a variable is always: (a)>1 (b) <1 (c)>0 (d) <0 2. Variance of a constant is (a) constant (b) 1 (c)0 (d) not possible 3. S.D is calculated from H.M: _” (a) Always (b) Never (c) Often (d) None of these 4, Range based on: (a) Upper and lower quartiles (b) Squared deviations (c) Minimum and Maximum observations —_(d) Absolute deviations 5, AM of the squared deviations of values from their mean: (a) Mean deviation (b) Standard deviation (c) Variance (d) Quartile deviation 6. If S.D (X) = 5, then S.D (2X+5) is: (a)5_ (b) 10 = (c) 15 (d) 20 7. A measure of dispersion is always: (a) Positive (b) Zero (c) Small (d) One 8: If X and Y are independent, then S.D(X - Y) = (a) S.D (x) ~ S.D(Y) (b) S.D(X) + S.D (1) (c) JVar(X) + Var(¥) (d) o{Var(X)- Var(¥) 9. For independent variables X and Y, if S.D(X) = 8, S.D(Y) = 6, then S,D(X-Y) (a) 10 (b) 2 (c) 14 (d) 28 10. If mode is less than the mean, distribution is: (a) Symmetrical (b) Normal —(c) Positively skewed (d) negatively skewed 129Measures of dispersion 11, Right tail is longer than the left tail, distribution is: (a) Negatively skewed (b) Positively skewed (c) Symmetrical (d) None 12. If mean = 40, mode= 42, distribution.is: (a) Negatively skewed (b) Positively skewed — (c) Symmetrical (d) None 13. If Y = 5X + 10, then mean deviation of Y is: (a) MD(X) (b) SMD(X) c) SMD(X)+10 d) MD(X)+10 14, Variance (or S.D) remains unchanged by change of: (a) Origin (b) Scale. -—(¢) Both (a) and (b) (d) unit 15. The lowest value of variance is: (a) 1 (b) 0 (c) -2 (d) -1 16. A normal distribution has 68.26% of the observations: (b) ¥+2s (c) ¥+3s (d) ¥+4S 17. Distribution is symmetrical, if Vo: (a) Negative (b) Positive (c) Zero (d)3 18. For normal (mesokurtic) distribution, between x-2S and X+2S: (a) 95.44% — (b) 50% (c) 68% (d) 99.73% 19. Observations lying within limits (X35) in the normal distribution is: (a) 68.27% (b) 95.44% (c) 70% (d) 99.73% 20. To compare the variations of two or more than two series: (a) Mean (b) Standard deviation (c) Variance (d) CV 21, Quartiledeviation: (@)2c H) 3020 (a) Ser 22. Mean deviation: (a) 26 (o) 40 © Se (a) Se 3 5 6 5 130Measures of dispersion 23. Bowely's co-efficient of skewness lies between: (a) Oand 1 (b) -1 and 0 (c) -1and +1 (d) -00 to 24. If co-efficient of skewness is -0.58, distribution is: (a) Positively skewed (b) Symmetrical . (c) Negatively skewed (d) asymmetrical 25. The types of dispersion are: (a) 2 (b) 3 (c):4 (d)5 26. The square root of second central moment: (a) Variance (b) Standard deviation (c) Quartile deviation (d) Mean deviation 27. For positively skewed distribution: (a) Mean < Median < Mode (b) Mean < Mode < Median (c) Mean > Median > Mode (d) Mean = Median = Mode 28. For negatively skewed distribution; Mean......Median.......Mode: (a)= (b) < (c) > (d) >< 29. In grouped data, the range is the difference between: (a) Two extremes class frequencies (b) Two extremes mid points (c) Two extremes class boundaries (d) Both b &c 30. Which is poor measure of dispersion in open-end distribution: a) Range b) Quartile deviation c) Semi-inter quartile range d) AM 31. The range of constant “A”: a) Zero b)A ct dn 32. For relative dispersion, unit of measurement: a) Changed b) Vanishes cc) Does not vanish = d) unit 33. The range of a series of -2, -3, -5 and -10 is: a) -12 b) 8 ¢)-8 d) 12 131Measures of dispersion 34, The variance of 5, 5,5, 5 and Sis: a)5 b) 25 c) 125 ay 35, If A.M=25 and S? = 25, then co-efficient of variation (C.V) is: a) 100% b) 25% c) 20% d) 1% . 36. Mean deviation is always: a) Less than S,D b) Equal thanS.D —_c) More than S.D d) Negative 37. In symmetrical distribution, the co-efficient of skewness is: a)-1 b) +1 c)0 d) 0.5 38, First moment about mean is always: a) One b) Zero c) mean d) SD 39, First moment about origin is equal to: a) One b) Zero c) mean d) SD 40. In a skewed distribution mean, median and mode are always: a) Identical b) Different c) Zero d) Same 41. Mean deviation is associated with: a) A.M b) HM ¢) QD d) GM 42. Third moment about mean (ms) is zero, distribution is: a) Positively skewed b) negatively skewed c) Symmetrical d) asymmetric 43, Sum of absolute deviations of values are least if measured from: a) Mean b) Mode c) Median d) GM 44. Sum of squares of deviations is least from: ,a) Mean b) Median c) Mode d) HMMeasures of dispersion 45. The second moment about mean is: a) Variance —_b) Mean c) SD d) Zero 46. The variance (S.D) of constant is: a) Constant —b) Unity c) Zero d) © 47. Standard deviation of 2, 4, 6, 8 and 10 is 2.83, then standard deviation of 102, 104, 106, 108 and 110 is: a) 283 b) 102.83 c) 2.83 d) 28.3 48. Standard deviation changes by the change of: a) Origin b) Scale ¢) Algebraic d) Both (a) and (b) 49. Y = X+3, then range of Y is: a)3 b) Range(X+3) c) Range (X) d) Range (X-3) 50. If Y=3X+5, then S.D of Y is: a) 9SD(X) b) 3S.D/") c) 3S.D(X)+5 d) 3S.D(X) £5 51. If b2(B2) =3, distribution is: a) Leptokurtic b) Platykurtic c) Mesokurtic d) Skewed 52. Var(2X+3) is: a) 4Var (X) b) 2Var (X) ) 4Var (X)+3 d) 2Var (X)+3 53. Variance (standard deviation) is calculated from: a) Mean b) Median c) Mode d) GM 54, If X and Y are independent than Var (X-Y) is equal to: a) Var(X).- Var(Y) b) Var(X) + Var(Y) — ¢) [var(X¥)+ var(¥) d) fvar(X)-var(¥) 55. bpmeasures: a) Symmetry b) Dispersion c) kurtosis * d) Skewness 133Measures of dispersion 56. The standard deviation of 3, 3, 3,.-.+, 3 is: a)3 b)8 c),Zero d) 16 57. Symmetrical distribution is: a) t-shaped b) J-shaped c) Bell-shaped d) long tailed 58. A.M=136.75, Median=148.37 and Mode=152.80: a) Positive skewed b) Negatively skewed c) Symmetrical d) asymmetrical 59. Addition of extreme value in a data set affects more a)Q.D b) Mode c) Median d) Variance 60. For nominal data we use a) Mean b) median c) mode d) GM 61. For ordinal data set we use a) Mean b) median - ¢) mode d) GM 62. For a data set n = 20, X = 20 and S? = 16 then )'(x - x)= a)1 b).320 c) 2000 d)o 63. Z score for X = 25 is 1.9 find SD if mean of X is 18 a)7 b) 3.68 ©) 1.39 d)o 134of key Ans | Sr. Ans | Ans Sr. Ans Sr. Ans Sr. Ans} Sr. St. a ¢ a c c d S 4 | 30 | 135Measures of dispersion Exercise QNo. 4.1: (a) Define measure of dispersion and its types. (b) Write different methods for measuring the dispersion. (c) Define the Range, Quartile deviation, mean deviation, variance, standard deviation, moments, skewness and kurtosis. (d) Describe the properties of variance and standard deviation. Q No. 4.2 (a): Find variance and Standard deviation from the following calculations, (i) n=65;(X- XY =8424 (ii) n= 80; 0X = 6824; )X? = 584595 (iit) f= 200; fir=9; fi? =150; and w= Sa (iv) n =10; 3° D = 52; 9° D? =952; and A=570 (v) 2 =10; 3D =-128; ° D* = 3472; and D=X-110 (b) (i) Find CV if S.D. = 10 and ¥ =25. (ii) Find CV if variance = 16 and ¥ =50. (ili) Find SD if CV = 30% and X =15. (iv) Find variance if CV = 40% and ¥ =25. (v) Find mean if CV =»25% and SD =15. (vi) Find mean if CV = 5% and variance = 9. Qo. 4,3: Find Variance, SD and CV. Weight (kg) | 0-9 | 10 - 19 | 20-29 | 30-39 | 40-49] 50-59 t 3 17 36 58 27 6 Q No. 4.4: Find QD, SD and MD about median. Income (000) in Rs. | O- 4.9 | 5- 9.9 | 10- 14.9] 15- 19.9 | 20 - 24.9 ‘No of employees 8 “28 60 30 5) 136Measures of dispersion Q No. 4.5: Find Moments about mean by calculating moments about zero. {Classes | 0- 9.9 | 10 - 19.9 | 20 - 29.9 | 30 - 39.9 | 40 - 49.9 | \f 5 7 15 12 [6 Q No.4.6: The marks X scored by a sample of 56 students in an examination are summarized by 1=56, YX =1026, yx? = 20889 Calculate the mean and standard Geviation of the marks. Q No. 4,7: Find variance by (i) Direct (ii) Indirect and (iii) Coding method. Classes [ 1- 10 | 11-20 | 21-30 | 31-40 [41-50] ft 5 10 18 | 6 2 | Also find coefficient of variation. Q No. 4.8: Find mean deviation about mean, median and mode also find their coefficients. Marks | 1 = 20 | 21 - 40 | 41 - 60 | 61 - 80 | 81 - 100 | t 5 | 2 | 50 18 8 Q.No. 4.9: Find mean deviation about median, Classes | 1 - 15 | 16 - 30 | 31- 45 | 46 - 60 | 61-75 | 76-90 | tf 6 10 2 | 20 | 8 | L Qo. 4.10: The mean stress rate of a group of students for a particular course is 6.40 and standard deviation is 2.5. Suppose a student’s stress raw score is 12, find his Z score. . 137Measures of dispersion Solution QNo. 4.2: : Sr. | Variance Standard deviation @ se 5S =3.60 2 7 5 =5.60 if 2 84595 6824)? = 5 (ii) | gtk 7 _ 58459) - (aes ) ae L4 n 80 80 i 2 =17. (iii) eo Efe Ef ) ‘= -{5-( 9 a Joe S=17 Bs 200 (a5 = 299.19 iv 2 ae @ S = 8.26 oo gtk a ) =% 3) = 68.16 | n n | a aaa 128)? S= 13.54 ) Sta Le @) -# = =) = 183.36 (b) owt 5% 100 | Sr._| Computation Sr. | Computation Sr._| Computation ro} vax |! cvetxrioo |] 30-5 x10045 25 50 15 __ | = 40% = 8% =45 Hiv) | az (vy) 5 (vi) _3 40 = 5; 100 25 = > x 100 5 = 3 * 100 5=10 55S? = 100 X=60 X=60 QNo. 4.3: Weights(Kg)|_f [ x [ul fu | fut 0-9 3 [45 |[-2[-6 [12 10-19 17_[14.5[-1[-17 [17 20-29 [36 [245/0 lo [0 30-39 58 | 34.5|4 40-49 [27 |44.5|2 [54 | 108 50-59 [6 [54.5|3 Total 147 107 [249 138fA 107, - “7 - (i) j= Meee Se sash “fu 07 R= fa RASS E th= 145430 10 = 4178 ky oy wan io0 33.9% The 00 = 43994 [er Kx Tul ful fut fix %) [161 32 7792 ale Z 1) -) BT 9% 11245101010 124 126 | 17.451 1 | 30 | 29 | 151.2 131 22.45) 2110 | 20 50.2; | +4 (110 422.36 393 5 AP 9025 = Oy = 14.95 + 7 (98.25 — 96) Scie qg -9.37 My _ 1533-937 _ a yecoonyks Jy f(xy eta (&) S= 458 (000) n 131 a +65 ah= 995+ (655 - 36) = 12.41 (000)Rs |= = 20,97(090) Rs? efir- Gr) rsa ae 3.22 (O00)Rs uf 139Measures of dispersion a) 808 | 3001.86 2 | 349573.57 2 | 5812640.44 | 5 17904820.29 | 5 (2449458212) 1314195.21 | 48564718.28 29204.34 1079215.96 6.5) = -365.95 — 3(26.5)* = 43936.3525 = 6.11 marks QNo. 4.7: Classes f x fx D fd fp? [ul fu | fuel] 1-10 [5 [5.5 | 275 | 151.25 -20_-100_ 2000 -2 | -10 | 20 | 11—20/10 15.5155 | 2402.5 10-100 | 1000 [-1/-10 110 | (21-3018 255/459 11785 0 20 10 01/0 107) 731-40/6 135.5) 21375615 10 60 1600 1116 16 "41-50 2 45.5. 91 4140.5 20 [40 [800 (214 18] Total. 41 945.5 25960.25 -100 4400. | -10 44 |Measures of dispersion rye eyye Zz ear Ci) tar) fie = ae YK Ws tlh : Wp cay 2806 f 0,07 Se 1007, 6V ¥ 4AN00 & - df LNOO = M467% QNo. 4.8; , - maks | oo | X ful yu fen | ef [peal peal fe 4| 1-20 [5 10.5} 2) 10] 05° 15 | 204 201 11985 | ja 40 | 20 | 30.5] 1) 20] 20,5" | 25 | 416 404 {394} a-60 |50 | 5405/0 Jo |40,5-] 75 | 40 10/15 61-80 | 18 | 70.5] 1 | 18 | 60.5-[93_ 356.4 | 365.4 ai-100/8 | 90.5] 2] 16 | 80.5-| 101. 318.4 | 322.4 Total | 101 . 14 131 (8 | 17 Mean deviation about mean: X = A YMA 1319.2 uy 101 0 Mean deviation about median: X 148 (5 c) = 405 + 220505 ~ 25) = rh = 505 +% 7 20 = 51.3 marks Mb = 13.06 marks 50.7 marks Y/R] 1289.8 mp = PEN gp = 1277 marks Mean deviation about mean: In~ hh Renae 50-20 20 = 50.2 mark: Be Tp oft 95 tsp 204 50— 18 “20 = 904 marks enh ee mp — LL AI _ 12953 99 99 marks L 101 144Measures of dispersion \Q No. 4.9: Marks | f |X [CB | ef | f[x-X] 1-15 [6 [8 [0.5- [6 | 205.2 16 - 30 | 10 | 23 | 15.5- | 16 | 192 31-45 | 25 | 38 | 30.5- | 41 | 105 46 - 60 | 20 | 53 | 45.5- | 61 | 216 61-75|8 | 68| 60.5- | 69 | 206.4 76-90/2 | 83| 75.5-| 71 | 81.6 Total [71 1006.2 Mean deviation about median: X = | + 37 ) = 305+ £55 — 16) = 422 MD = ———— = —— = 14.17 QNo. 4.10: 142