~ per quantity of salt. There are so many techniques to draw a sample from a population, Here we shall discus only the techniques of Simple Random Sampling. Before introducing they20simple random sampling, some of the important terms and the basic concepts are defined first, Population A complete collection of individuals, objects or measurements under consideration in a statistical study will be called a population. OR A population may be defined as the large collection of similar units, The number of units in the population (i.e. size of a population) is always denoted by N. For example, a population may be of heights of all students of a college, a Population may be of all books in a library, a Population may be of all Persons using a particular medicine, a population may be of prices of a commodity over a period of time, etc. 361362 bs Pr : . lem \ population may be finite or infinite depending upon Wheth Ne ae er countable number of units or an uncountable number of units, oon Sample \ small part of a population is called sample. The number of wi (Le. size of a sample) is always denoted by n. its jn, the Parameter Any quantity calculated from a population is called a population mean, population variance, mean is denoted by i parameter. For ex, ete, are therefore parameters, The : oP lat {cand the population variance is denoted by o?, a Statistic Any quantity culated from a sample is called a statistic, For example, sam ; » ele, are therefore statistics. The sample mean is denoted 4 \.and the sample variance is denoted by s?, mean,sample variance Sampling with and without replacement Sampling is said to be with replacement when we draw a unit from a finitt population and return it to the population before the next unit is drawn. In this case each unit can be drawn more than once and the probability of drawing of each unit remait constant throughout the sampling procedure, Sampling is said to be without replacement, if we do not return the selected v the population and draw the next unit. In this case each unit cannot be drawn mo: once and the probability of drawing of each unit changes throughout the i procedure Census and Sample Survey a Census or complete enumeration means to get the information abouteach« unit in the population, a technique of getting information about the che g only a part (i.e by studying a sample) of the \ sample survey is the population by studyin, >Statistics Probioms 5 ” random sample is drawn unit by unit by means of « f “Random Numbers Table”or by the use of “Co ty ers, The detail is not needed to be Siscusseay * In practice a simple or by the use 0! System” Random numb Programs” that provide as it is supposed to be beyond o fthe scope of the book. Selection of a Simple Random Sample population of 2000 students in a college and we want to ray dents from this college. According to this techniques, ° sr ail we allot serial numbers from 1 to 1000 to all students i.e. 0001, oom 0003,....2000 and then prepare 2000 chits of these numbers. After shuffling the chi, > draw any 100 chits one by one. The students bearing the numbers on the slips ville included in our required sample. This procedure of making chits and then drawing js called lottery method. This method is not convenient when we havea large Population, 5 y apply computer programs. Suppose there is simple random sample of 100 stu we pling Distributions Consitier all possible samples of size n which can be drawn from a given population (cither with or without replacement). For each sample we can compute a Statistic, suchas the mean, variance, etc. Which will vary from sample to sample. In this way we obtains distribution of the statistic which is called its sampling distribution. Therefore, the sampling distributions may be of mean, variance, etc. . Sax Sampling Distribution of the mean ( x ) if we draw all possible samples each of size n from a finite population of N units with mean and variance c? . Then we compute mean of every sample i.e. X. Therefore, the statistic (X ) is now a random variable and form a probability distribution. This distribution is called sampling distribution of mean. The sampling distribution of mean has the following properties in case of sampling without replacement and in case of sampling with replacement. -Properties in case of Sampling without replacement 1 Mean of all sample means is equal to the population mean i.e E(X) = wet X is said to be an unbiased estimator of w. where o? is the population variance . Gesyo gy Semen properties case of Sampling with replacement 368 1 Mean of all sample means is equal t ‘0 the populati a x is said to be an unbiased estimatoy oe of p. ie E(X) = por a a a Ce Where 6” is the population variance . gandard Error The standard deviation of sampling distribution of & is called standard error of ¥ : N-n Voy it oase of without replacement in case of with replacement Thus S.E. of (X)=0, = vn and S.E.of(X)=o, = © Example 1 | Apopulation consists of five numbers 0, 2, a) 4, 16, 8 List all possible samples of size 2 that can be drawn from this population without replacement b) Find mean of each sample c) Construct sampling distribution of x. d) Verify that mean of all sample means is equal to population mean. Solution: Since N = 5 and n = 2 and the sampling is done without replacement , then all 5! i am a possible samples = °Cy = 73 —py = 10as shown below;SS ae ee Statistics Propy, ‘ ane 366 ——_____—_——_—____ ey, All Possible | Sample Mean oS Sample No. Samples 920% 1 (0, 2) 1 " 2 (0, 4) 2 3 (0, 6) 3 4 (0, 8) 4 5 (2,4) 3 6 (2, 6) 4 7 (2,8) 5 8 (4, 6) 5 9 (4, 8) 6 10 (6, 8) 7 Total 40 ° Sampling Distribution of X is Nauawnes | |” Total The Mean of X = E(X) 0+2+4+6+8 5 ae 7 , ‘ . Therefore, it is verified that , mean of all sample means is equal to population meal | i a = 4 and since, population mean =gy ooo ——-—— 367 f eo opulation consists of five numbers 0, 3, 6: 9 . AP ; List all possible samples of size 3 that can be draw from this population u without replacement . ») verify that, Mean of ¥ = E(x) = ee sinee N= 5 and n= 3 and sampling is done without replacement. Then all eee -| 31(5 -3)! 0 cgsible samples paulation mean and variance are computed ag Xx _ 30 Themean and variance of ¥'are computed as an All Possible Pea nage] (0, 3, 9) | (0,3,12) | (0, 6, 9 | (6,12). | | @9, 12) 1 G30,9) | @,6,12) | | (6, 9, 12) be -4(359,12) 4 | | cTotaletas 60 wWOIAIAUNUEYMatintes 368 : » 3 00 wee , Mean of = BCX) = 0 M © M= NO.OF samy, Therefore, itis verified that &CX) in ye yey Now Fv) fot hon whore m © numbers of sampley m m 390 (a vax) 7 10 (io v so? N=n 18'$-3 Also F(X) 4 n N=) 3° 5-1 o Neon Hence it is also be verified that V(X) n'N-=1 Example 3 Draw all possible samples each of size 2 from the population 2, 4, 6 and & using sampling with replacement, Find mean of each sample and verify that i () Mean ofR-= wand (i)'7(R) = F Solution: Firstly, we compute Population mean and variance xx _ 2 _ 120 (2) mer 4 Since N= 4 andn=2 then, possible samples = N" = 4? = 16 ‘The possible samples and their means are computed as a_i—— 369 | G4) 4 | 1 G6 25 ; G&S | 6 | (6, 2) j | 5 = p. Therefore Meanof (xX) = p Now V(#) = ) where m = numbers of samples 2 : V(k) = & fe = 25 16 16 * 2 . Also V(x) = & OF ms s n 2 Hence itis also be verified that V(X) = Rrample 4 Draw all possible samples each of size n = 3 from the population 2, U1 by using “pling with replacement. Find standard error of sampling distribution of_. Statistics Problems « p 370 7 Solution : since N= 2.and n=3 then possible simples = N* = Possible Samples. Sample Mean X x (2,2, 2) . (2,2, 11) @ 112) (11, 2,2 (2,11, 11) (11,2, 11) (11, 11,2) (11, 11, 11) Total ; se S.E. of (®) =0, = 992_(32Y 6 8 8 Example 5 If the size of a sample is 64 and standard error of mean is 1.5 . What should be the sample if standard error reduced to 0.6 Solution: Since Since then Now and