Home / HOME / application of statistics in educational research Testing Hypothesis

application of statistics in educational research Testing Hypothesis

Testing Hypothesis

Hypothesis testing begins with an assumption or the changes of getting a the loss of some things which are always negative in properties i.e. two hypothesis created one accepted then the other should be rejected and vice-versa. There are two types of hypothesis:  (i) Null hypothesis (Ha) or alternative hypothesis (Ha), Hypothesis may be true or false accepted or rejected.

It is a prediction about the activities of our daily life which help to evaluate the program regularly. So, hypothesis most be tested to obtain the correct result which can be calculated by the following methods:

  1. Error of measurement
  2. T-test
  3. Chi-square test
  • Error of measurement

There may be some error in any kind of measurement on the basis of principle of measurement there may be some error certainly. The process of measurement itself is not correct process. Error may be corrected by calculating the average between two score and error score. More  the error less the reliable and less the error more the reliable. So each and every  measurement certain error. This is universal theory of measurement. However the measurement process should be corrected to get reliable result to decrease the error in the measurement max-mincon theory should be applied.

Max-mincon means to control the error in the measurement experimental variance should increase. Minimizes the error variance and controls the external variance.

 

Calculation

  • Marks obtained from Janasahayog Higher School, Itahari-24, and grade-VIII subject Social Subject mark of 20 students.

Table :-24

Subject Social No. of Students- 14. Calculation of correlation of obtained marks in odd and even number of question

sn Odd(X) Even(Y) X2 Y2 xy
1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

9

10

8

9

10

8

9

7

10

7

7

9

8

7

6

7

7

5

5

5

10

9

11

9

8

9

8

10

6

9

8

6

7

7

7

6

5

7

6

5

81

100

64

81

100

64

81

49

100

49

49

81

64

49

36

49

49

25

25

25

100

81

121

81

64

81

64

100

36

81

64

36

49

49

49

36

25

49

36

25

90

90

88

81

80

72

72

70

60

63

56

54

56

49

42

42

35

35

30

25

C=153 SY=153 S X2=1221 S Y2=1227 SXY=1190

Source:-spot survey

 

  Table :-25

Calculation of standard deviation from mark of social studies of class-8

X f d fd Fd2
10

11

12

13

14

15

16

17

18

19

1

1

2

2

1

3

2

3

2

3

-5

-4

-3

-2

-1

0

1

2

3

4

-5

-4

-6

-4

-1

0

2

6

6

12

25

16

18

8

1

0

2

12

18

48

N=20 Sfd=6 Sfd2=148

Assume mean (A)=15

6.2 T-test

Generally, to study the two groups  significance difference we use T-test. Two mean are calculated and it is a hypothesis testing process. Again we can say that the test of significance of the difference between the two mean is called T-test. In it, test between the hypothesis of statistical may be accepted her or may be rejected total by the T-test. At the significance levels at 5 % or 1 % of the observersation.

Degree of freedom (N-1)

In the small sample group (N < 30) to find the reliability between two mean df = N1 + N2 – 2

  1. ii) Null Hypothesis

Null hypothesis is the first stage of any research. It is also said negative hypothesis. Generally, it is prepared to proposed of reject. If null hypothesis is rejected than our alternative hypothesis is correct according to L.R. Gay, “Null hypothesis states that there will be relationship (or difference) between variables and that any relationship found will be a change relationship.

Accept or reject condition of null hypothesis.

  1. If the calculated value is less than tabulated value the hypothesis is retained.
  2. If the calculated value is more than tabulated value the hypothesis is rejected

we had taken 14 students randomly from class six and divided in two groups by giving 20 objective question of science subject. The students were farmers’ child and labour child. We have presented obtain mark by two group of T-test.

There is no different between the farmers’ child and lobour childs in average achievement.

 

 

Table :-26

Obtained mark of farmers’ child of class six in Science.

Obtain mark of labours child of class six in Science.

 

Here, the degree of freedom is 12 by finding the value of t. At 5 % level and 1 % level calculated value is 0.37 tabulated value at 5 % level is 2.179 and 1 % level of significance less than 3.055. So the Null hypothesis is retains. So there is no relationship between farmers child and labour child and there is no differnces between the achievement of farmer and labour child.

6.3 Chi-square test ( X2)

It is one of the simple test and it is widely used non parametric test in statistical work. It is denoted by Greak letter ( X2) (Chi-square). The( X2)test is depend on interval and ratio scale. Generally it opposite to t-test. ( X2) test had been started by Helmert at first. Karl Pearson had developed and used in statistical work in 1990. It is known as test of goodness of fit. It shows the degree of relationship between two variables but it does our show the quantity of degree of relationship.

 

  1. Pass Value of boys =
  2. Fail value of girls =
  • Pass value of girls =
  1. Fail value of girls =

For calculating 2.

Table: 29

fo fe fo-fe (fo-fe)2 (fo-fe)2/fe
16 15.54 0.46 0.21 0.01
5 5.46 -0.54 0.29 0.05
21 21.46 -0.54 0.29 0.01
8 7.54 0.46 0.21 0.03
0.10

 

Now, degree of freedom

df = (r -1) (C – 1)

= ( 2 – 1) ( 2 -1)

= 1 × 1

= 1

The above calculation shows 2 value 0.1 is less than tabulated value 3.84 if table value at 5 % level and 6.64 at 1 % level. So our calculated value cannot be rejected. There is no effect of gender and obtaining mark of Science of achievement level of the students.

 

 

Conclusion

 

From the above statistical analysis the following conclusion are found.

  1. Shree Janasahayog Higher Secondary School, Itahari, final examination 2071 of grade ix students obtained marks in Science and English’s mean values are 36 and 35.33 respectfully. This mean value shows the result of class ix is not satisfactory.
  2. In this way, this schools final examinational 2071 of grade xi, obtain mark in Nepali and English mean values are 45.47 and 26.33 respectfully and it seems low graded. In fact the result is not so satisfactory.
  3. Shree Janasahayog Higher Secondary School, Itahari. In the final examination 2071 of grade ix obtained marks in Science and English median values are 37 and 34 final examination 2071 of grade xi English and 44 respectfully.
  4. Shree Janasahayog Higher Secondary School, Itahari, the final examination 2071 of grade ix obtained marks in Science and English range values 7.6 and 55 and co-efficient of range values are 0.84 and 0.66 respectively.
  5. Standard deviation of class xi of compulsory Nepali is 9.83 and co-efficient of variance is 22.55 %.
  6. Shree Janasahoyog Higher Secondary School, Itahari- 27, five years students enrollment on the basis of sex, the enrollment shows the number of students are decreasing. It has also shown is pi-chart and histogram. Pass and fail percentage from class are to ten has also shown in pie-chart.
  7. The polygon from grade IX shows the normal distribution and polygons increase and decrease respectfully in graphs. Average pass students number is higher and result is satisfactory.
  8. Percentile rank of grade IX students in Science result us Anis Basnet, Prajita Basnet and Dhan Kumari Newar got the first second and 3rd position respectfully.
  9. Quartile rank of XI students in the English shows 3.95 the higher value and 0.07 the least value.
  10. Standard error measure in English XI shows the value 21.376 and it is not better and reliable.
  11. T-test result the positive variance between two test held on different time interval of class-8 students of in social shows not in students achievement level.

Similarly T-test measures no significant difference and hypothesis calculated can not be rejected.

  1. By testing x2 square test. The mark obtained by boys and girls above 32 and obtained mark. The calculation shows that the hypothesis is can not be rejected. So there is no relation between the sex different in achievement level of gender.

About admin

Leave a Reply

Your email address will not be published. Required fields are marked *

PHP Hit Counter Example Code

Page content...

This page has hits