Page 371 - DMTH404_STATISTICS
P. 371

2
                                                                                          Unit 26:   - Test Hypothesis



            Also we note that O  + O  = E  + E  = n.                                              Notes
                            1   2  1   2
            The above result can be generalised for a manifold classification. If a population is divided into
            k mutually exclusive classes with observed and expected frequencies as O , O , ...... O  and E , E ,
                                                                      1  2     k     1  2
                                     k  (O -  E  ) 2
                                  2
            ...... E  respectively, then   = å  i  i   is a c  - variate with (k - 1) d.f. Here again we have
                                                   2
                k                        E
                                    i=  1  i
             k     k
            å O =   E =  N  (total frequency).
                i å
                     i
            i= 1  i= 1
                   Example 40: 300 digits were chosen from a table of numbers and the following frequency
            distribution was obtained :

                     Digit     :   0   1    2    3    4    5    6    7    8    9
                   Frequency   :  26   28   33   32   28   37   33   30   30   23

            Test the hypothesis that the digits are uniformly distributed over the table.
            Solution.
            When H  is true, the expected frequency of each digit would be 30.
                   0
                     1    2
                 2
                    =  å O -  N
                          i
                    30
                    1    2   2   2   2    2   2   2   2    2   2
                   =  (26 +  28 +  33 +  32 +  28 +  37 +  33 +  30 +  30 +  23  ) 300-  =  4.8
                    30
                       2
            The value of   from table for 5% level of significance and 9 d.f. is 16.92. Since the calculated value
            is less than tabulated, there is no evidence against H . Thus, the distribution of numbers over the
                                                     0
            table may be treated as uniform.

                   Example 41: A sample analysis of examination results of 200 M.B.A.'s was made. It was
            found that 46 students had failed, 68 secured a third division, 62 secured a second division and
            the rest were placed in the first division. Are these  figures commensurate with the general
            examination result which is in the ratio of 2 : 3 : 3 : 2 for the various categories, respectively?
            (Given : Table value of chi-square for 3 d.f. at 5% level of significance is 7.81.)

            Solution.
            H : The students in various categories are distributed in the ratio 2 : 3 : 3 : 2.
             0
            The expected number of students, under the assumption that H  is true, are :
                                                               0
                                                     2
                       expected number of failures  =      ´ 200 =  40 ,
                                                (2 3 3 2+ + +  )

                                                        3
                       expected number of third divisioners  =  ´  200 =  60 ,
                                                        10
                                                          3
                       expected number of second divisioners  =  ´  200 =  60  and
                                                         10

                                                       2
                       expected number of first divisioners  =  ´  200 =  40 .
                                                       10


                                             LOVELY PROFESSIONAL UNIVERSITY                                  363
   366   367   368   369   370   371   372   373   374   375   376