; Measures of Dispersion Measures of Dispersion: Measures of central tendency give us an idea of the concentration of the observations about the central value of the distribution. Let us consider two groups, each of 6 students with their scores ina particular examination. Group-A: 48, 50, 52, 51, 50, 49 Group-B: 1,2, 100,99, 98,0 The average score for each batch is 50. tis evident that the distribution A and B have the same they differ from AM. Such variation is usually called dispersion, Definition. Measures of dispersion give the degree of scatterness about the centr siving measure of variability or lack of homogeneity ofthe data. Purpose of Dispersion. A measure of Dispersion serves two purposes 1. It provides one ofthe most important characteristics ofa frequency distribution. 2. helps us to compare two oF more frequency distributions. (Characteristics of an ideal Measure of Dispersior: Itshould be rigidly defined. [should be easy to understand and easy tocaleuate It should be based om all the observations ofthe dataset. It should be so defined thatthe measure is amenable to mathematical treatment in further analysis. should be affected as small as possible by sampling fluctuation. should not be affected by extreme observations. “Measures of Dispersion: Following are the measures of dispersions: a) Absolute measures: 1 Range fi Quartile deviation fii, Mean deviation jv. Standard deviation b) Relative measures: AM, but fal location and thus 1 Co-eficient of variation i Covetticient of ringe fii, Co-eficient of mean deviation iv. Co-eficient of quart a) Absolute measures of dispersion, These measures are absolute in the sense that they are expressed, in the same statistical units in which the original data are presented. i, Range: The range of a set of values isthe difference between the highest and the lowest value in the e deviation set. Thus if xjand x, denote the smallest and the highest value respectively in a set then the range is given by Range (R) =, In grouped frequency distribution, the range is taken either as the difference between the highest and. the lowest mid-values or as the difference between the lower boundary of the first class and the last class. For a set of observations 90, 110, 20,51, 210 and 190, the range is 210.20~190. Mer 1. Itis the simplest measure of dispersion, 2. Itis easy to understand and easy to calculate 3. Itis based on the extreme observations only and no detail information is required, 4, It gives us a quick idea of the variability of the observations involving least amount of time and calculations. Demerits, . It is a crude measure of dispersion as the two extreme observations may subject to sampling fluctuations, 2. It would be misleading if any of the extreme value is very high. 3. Iteannot be calculated ifthe extremve classes ofthe frequency distribution are open. Paget of 7Measutes of Dispersion Uses 1 tis used in meastirement of share marke fluctuation. 2. Risalso used in statistical quality control work 4, Quartile deviation: If Q,and Q, denote the first and third quartile respectively, then the inter- quartile range ts Q=Q)-Qs. The inter-quartle range is frequently reduced to the measure of Semi- {interquartile range, also knovn asthe Quartile Deviation (QD), by dividing it by 2. Thus op = 222 Quartile Deviation is definitely a better measure than the range as it makes use of 50% of the data. But since it ignores the other 50% of the data, it cannot be regarded as a reliable measure. Ex Find out quartile deviation from the following frequency distribution Frequency 3 y= 1683, Q=225, Que 2758 1 Therefore, Q.D = 2=9. - 2758-1683 =5375 3540 iii, Mean deviation: The mean deviation is an average of absolute deviations of individual observations from the central value of a series. Let 2, ‘mean is defined by 1%, ben observations of a variable with mean € then mean deviation from arithmetic ) And MD(M ES Ale a, where w= Sy Mer 1,_ Itiseasy to understand 2, Itis relatively easy to calculate 3. It takes all the observations into account, 4. It is Jess affected by extreme values. Demerits, 1. Itis not amenable to further algebraic treatment 2. Iteannot be calculated if the extreme classes ofthe frequency distribution are open. 3, Itis ess stable than standard deviation, Since MD is based om all the observations, it isa better measure of dispersion than range or quartile deviation, But the step of ignoring the signs of the deviations creates artificiality and renders it ‘useless for further mathematical treatment. jv, Standard deviation: The arithmetic mean of the squares of the deviations of the given observations: from their arithmetic mean is known as variance. The positive square root of variance is the Standard deviation. Page 2 of 7Measures of Dispersion Let ast tion is defined as 1», ben observations ofa variable then standard des LSC, -3 In case of frequency distribution or grouped data, In case of frequency distribution, oO fnt 22RD fa tN =D hae Ne 2 Fes 3 =D he 248 )= Dh Bx, Caleulate the SD from the following frequency distribution [ei] st-values ) [Frequency () [tm [Txt fesy 3s z 3 [125 5-10 | 75 5 375 | 281.25 ais] 125 7 75 | 109875 | 175 EE 25 Br | a7 5 | 106s1 25 ad 16 | H0_|_ 12100 325 3 260_| 8150 5 3 [aias | a8 75 754 [W625 | 4076875 V543.58— 479.61 99 1. Itisrigidly defined 12. Tetakes all the observations into account, 3, Iis amenable to algebraic treatment, 4. his less affected by sampling fluctuations. 5. TheSD of combined series can be obtained if the means, S1Ys and number of observations in each series i given. Demerits 1, It is not readily comprehensible; the calculation requires a good deal of time and knowledge of arithmetic. 2, is affected by the extreme values. 3. Iteannot be calculated ifthe extreme classes of the frequency distribution are open. Page 3 of 7Measures of Dispersion tis the most useful measure of dispersion and has got inimense use tn such as sampling, core vanced statistical work” bility and unifornlly of lation analysis, normal eutve errors, comparing ¥ wo sets of da atc ¥ When the deviations of the observations are taken from any standaral deviation reduces to Root Mean Square Dev = fala ubitrary value A other than <, on, which Is clefined by. Where =f, and A isany aritary value Empirical Relation among QD, SD, and MD: For moderately asytnmetric distribution, 1 Quarite deviation» Standard deviation Mean deviation from mesi= Standard deviation 1 Relative measures of dispersion: Whenever we want to compare the variability of two setie, which differ wide in their measures of location or which mcasited in different units, the absolute serve our purpose properly. In this situation, we usually calculate the relative measures of dispersion, which are pure numbers, independ i. Co-efficient of variatian: This isthe most important measute of re Pearson, The coefficient of variation is defined as the ratio of the to the arithmetic mean of t measures of dispersion do not of units of measurement lative variation developed by Karl standatd deviation of the distribution same distribution and usually expressed as percentage, jc. 82s 100=" 100 7 ¥ Wheres and ¥ are the standard deviation ane me ofthe distribution respectively Ex Let us suppose that we wish to obtain some insight into whether height is more variable than the ‘weight in the same population. For this purpose, for instance, we have the following data obtained from 150 childzen in a community: Tks 2kg. 030} Examination of the respective standard deviation does not tell us in any meaningful way avhich characteristics has more variability than the other, because they ate measured in different units, Thus the CV for weight is greater than that of the height, we would conchide that weight has more variability than height in the population, il, Coefficient of range: The co-efficient of range isa relative measure corresponding to range and is obtained by the following formula Covefficiont of rang Where L and § are respectively the largest and the smallest observations in the data set Corefficient of Mean Deviation: As the mean deviation ean be compated from mean, median, mode or from any arbitrary value, a general formula for computing coefficient of mean deviation follows: Co-elfcient of mean deviation= 100 MDA) 4 Where, Als the mean, medan mod or any ker siary vale, Covent of quote deviation The ocficent of quire deviaon sti by 2-2 con=2=2.,100 OP O40, iore and, denote the fist and thie quae pete age 4 of 7Measures of Dispersion ‘Theorem-l: The variance ofthe frst a natural number is 2 Proof: Consider (y,¥,.4).04,) denote respectively the frst n natural umbors ate (1230-08) ‘Then x ans EY Dew ea ey 2 aed) 42434. er e243 tenon? = MPD OneD 6 n+ )2ne1 Nene) Hendin ey? 6 nin) 2m 2125 _nnst) (=D 4m natural number i “Therefore, the variance ofthe firs “Theorem-2: The standard deviation is independent of origin but depends on scale nr) be n values of a variable X. Let us define the new values corresponding t0 i Proof: Let (51,235. the original observed values by the following relation: w, = Daou (origin) and his the wih ofthe class interval (eale) Thergire,x,~A= hu, ar, = Ahold) nC) Where A is any provisional value where, x cand Subtracting i from (9, we have : x -¥=h(u,-) on. S(6,- ori AS 3 =P LS? vi? or.S, = HS? : oS, =|, It is clear from the above relation that S, does not depend on the origin (A) but depends on scale (b) Page $ of 7Measures of Dispersion Theorems: Fortwo unogaal observatins, MD = SD =. where MD SSD = Standard deviation and R = range Proof: Let the two observations be x and x: (*:> %:) BY deviation of x; and x9 ate respectively = mean deviation, definition, the mean, range and standard Now. MD page 6 of 7 cicMeasures of Dispersion “Theorem-t: SD is the least possible root mean square deviation. (0, Root mean square deviation isthe least when the deviations are taken from the arithmetie mean. Proof: Let (x, “-x,) be the values ofk observations with corresponding frequencies fi fief Also let ¥ be the mean of the observations and A beany arbitrary value. 1 a yes ay = EE Alls -7)+@-aF fe rhe aS fie -nrba-a Eh a = LY p(x, 7) +2 positive quantity, yee positive quantity 57 = 5? +a postive quantity SeS" “Theorem-5: If ¥ and be th ve AM and SD respectively of non-negative observations, then Rn-125. roof: Let us consider (8s, the~-v8s) bem nonnegative abservations: We Krewe Since x's are non-negative, Sa, 20 eave dn > (= ‘oa! zns* SP Or-)2s? > Wa-l25. Page 7 of 7