Page 78 - ASBIRES-2017_Preceedings
P. 78

Ganegoda, Wickramasinghe & Appuhamy


           Table 3: Model fitting information            the  outcome  categories.  Table  4  illustrates
                                                         overall goodness of fit test.
                  -2 Log    Chi-
         Model                         df  Sig
                  likehood  square                                 Table 4: Goodness of fit

         Interce                                                    Chi-Square  Df         Sig.
         pt only   293.614                               Pearson    62.786       70        .717


         Final    157.053   136.561    8    .000         Deviance   67.792       70        .553


              The Table 3 illustrates comparison of      According to goodness of fit indices the data
       intercept only model vs final model. As can       and  the  model  predictions  are  similar  and
       be  seen  in  Table  3,  the  final  model  gives   that the model fit is good.
       significant  improvement  over  the  baseline
       intercept only model. This indicates that the            Following    Table    5    illustrates
       model gives better predictions than guessed       parameter  estimate  coefficients  for  the
       value based on the marginal probabilities for     model.

                                       Table 5: Parameter estimates


        Threshold                               Estimate         Wald           Sig.

        [Mathematics Marks = A]                 -24.333          1403.2         0.000

        [Mathematics Marks = B]                 -20.747          1371.85        0.000
        [Mathematics Marks = C]                 -19.035          1234.46        0.000

        Location [Attitude_Like=1.00]           -19.304          1796.02        0.000

        [Attitude_Like =2.00]                   -19.224          3086.81        0.000
        [Attitude_Like =3.00]                   -18.261          3086.81

        [Attitude_Like =4.00]                   0                3086.81
        [Attitude_Value=1.00]                   0.347            3086.81          0.475

        [Attitude_Value=2.00]                   -0.41            3086.81        0.336
        [Attitude_Value=3.00]                   0                3086.81

        [Attitude_Confident=1.00]               -2.398           3086.81          0.000

        [Attitude_Confident=2.00]               -1.985           3086.81        0.000

        [Attitude_Confident=3.00]               0.039            3086.81        0.935

        [Attitude_Confident=4.00]               0                3086.81


              Table  5  shows,  attitudes  (Students’           That is students achieved high marks
       like in mathematics, Students’ confidence in      for  Mathematics,  due  to  having  high  range
       mathematics)  have  a  significant  effect  for   of likeness and confidence in Mathematics.
       the  students’  mathematics  achievement          Following Table 6 represents significance of
       except  the  attitude  students’  value  of
       mathematics at 5% level of significance.          the proportional odds assumption.



                                                       68
   73   74   75   76   77   78   79   80   81   82   83