27-2. SAMPLING DISTRIBUTION Consider all possible samples of size n which can be drawn from a given population at random. For each sample, we can compute the mean. The means of the samples will not be identical. If we group these different means according to their frequencies, the frequency distribution so formed is known as sampling distribution of the mean. Similatly we can have sampling distriaution of the standard deviation ete. AWhile drawing each sample, we put back the previous sample so that the parent population Femains the same, This is called sampling with replacement and all the subsequent formulae will pertain to sampling with replacement. 982SAMUNG ANOREARENCK eens a a M an is called the af means te called und error i called prectstan (2 Standard error. The standard devintion of the sampling disteibu (SE. Thus the standard error ofthe sampling distribu Mere irene thonentnslereelpioei poomeneamies and observed values. The reeipeocal of the sti Ten > 80, sample is called ange ot fs assumed to be normal wise small The sompling distribution of large sample 27.3. (1) TESTING A HYPOTHESIS* To reach decisions bout populations onthe basis of sample information, we make 2 assumptions about the populations involved. Such gesumptuns, which nay o fray ave called statistical hypothesis By tating aiypothesi ts meant a proces fr decting t acceptor eet the hypthesin The method consists in aovuming the hypothesis andthen computing the probability f getting the abserved sample Ith probability i a certain preasigned value the hypathesis erected. 2) Errors Is bpsthens ejected whl shuld have been rtd ey AE has been commutted. On thesthet hand, fa hypothesis is accepted while tebe ren Fejeced we say that a has been made. The statistical testing of aims at limiting the Type I error tea reassigned value (sy 9% or 1=) and to muinsniss ‘Type Il error. The only way to reduce both types of errs isto inceate the tam pore (3) Null hypothesis. The hypothesis formulate assumption tht it eta scaled the one procedure is beter than another procedures. Similarly to tet whether hip betreen eo variates, we {at chere is no relationship. By accepting a nl hypothesis, we mean that on the bas of the ‘Matis calculated from the sample, we donot reject the hypothesis It however, des at sale fhat the hypothesis is proved tobe true, Nor its ejection ilies that i i disproved for the sake of rejecting it, und and is denoted by Hy. To test whet 27-4. (1) LEVEL OF SIGNIFICANCE 4 agbactaity level talon which we rej tbeiyootheris is known as Cs dala ‘The region in which a sample value falling ts rejected, is known as the 2 fe generally take two critical regions which cover 5% and I~ areas ofthe normal curve ‘The shaded portion in the figure corresponds to 5 level of significance. Thus the probability of the value of the variate falling in the critical region is the level of significance. Depending on the nature of the problem, we use a singlesail test or double-tail test to estimate the significance of a result. In a ouble-tail test, the areas ofboth the tails ofthe curve representing the sampling distribution are taken into account whereas in the single tail test, only the area on the right of an ordinate is taken into consideration, For instance, to test whether a coin is biased or not, double-tail test should be used, since a biased coin gives either Fig. 271,more number of heads t Is (which corresponds to right tail), or more number of tails heads (which corresponds to left tail only) (2) Tests of significance. The procedu reject the hypothesis is called the’ nables us to decide whether to ac lere we test whether the diffs the values given by two samples or these differences are small as to