Вы находитесь на странице: 1из 6

NON-PARAMETRIC TESTS

Advantages:
1. Most nonparametric tests involve easy computation, fewer mathematical details, and easier to
understand and to apply;
2. they may be used to test data that are measured on a nominal or ordinal data;
3. many nonparametric test s make it possible to work with small sample size, and very helpful to
the experimenter involved in a pilot studies or dealing with rare types of characteristics;
4. In multi-sample cases, there are available nonparametric tests for testing hypothesis
concerning observations drawn from several different populations without checking the
assumptions about the population;
5. Probability statements associated with most nonparametric are exact probabilities, which are
independent of the shape of the population
Disadvantages:
The major disadvantage of nonparametric tests is that if in fact all parametric assumptions
were satisfied in the data the values were of the required level of measurement, nonparametric
tests would disregard much of the information contained in the data. It is also though the
information conveyed by the figures used in the parametric test were ignored by its nonparametric
counterpart. In general, nonparametric tests when applied to data which are nearly normal or
nearly normal are not so powerful as the equivalent parametric method.

A. CHI-SQUARE TEST FOR INDEPENDENCE


The test for independence is used to determine whether two variables are related or not. For
example we might test whether a person `s music preference is related to his intelligence as
measured by IQ. We then take a random sample and for each subject determine his music
preference and classify his IQ into different categories (high, medium, low). The observed
frequencies are presented in what is known as a contingency table shown below:
Music
Preference

IQ
High

Medium

Low

Total

Classical

40

26

17

83

Pop

47

59

25

131

Rock

83

104

79

266

Total

170

189

121

480

contingency table containing r rows and c columns is referred to as an rxc table. The row and
column totals are called marginal frequencies. Note that in a test for independence, these marginal
frequencies are not fixed in advance but depend instead on the way the sample distributed itself
across the various cells in the table.
Procedure:
1 State the null hypothesis.
Ho: The two variables are independent or Ho: There is no significant relationship between
music preference and IQ.
2 State the alternative hypothesis
Ha: The two variables are not independent or Ha: There is a sig. relationship between music
preference and IQ.
3 Choose the level of significance.
4 Test Statistic: Chi-Square Test of Independence
5 Decision Rule: Rejected Ho if x2 > x2 ,(r-1)(c-1)

Compute the test statistic, given by


r

x 2

(Oij Eij ) 2
Eij

i 1 j 1

where Oij= observed number of cases in the i th row of the jth column
Eij= expected number of cases under Ho

(column to tal) x(row total)


Grand total

Decision

8. Conclusion

Remarks:
1 The test is valid if at least 80% of the cell has expected frequencies of at least 5 and no cell has

an expected frequency
1.
2 If many expected frequencies are very small, researchers commonly combine categories of
variables to obtain a table having larger cell frequencies. Generally, one should not pool
categories unless there is a natural way to combine them.
3 For a 2x2 contingency table, a correction called Yates` correction for continuity is applied. The
formula then becomes
r

x
2

( Oij Eij - 0.5) 2


Eij

i 1 j 1

Example:
Using the table above:
1 Ho: Music preference and intelligence are independent
2 Ha: Music preference and intelligence are not independent

3
4

= 0.05
Test Statistic: Chi-Square Test for Independence

Decision Rule: Reject Ho if

Computation:
Music
Preference

x 2computed > x 2tabulated

x 2c >9 . 488

IQ
High

Medium

Low

Total

Classical

40 (29.4)

26 (32.7)

17 (20.9)

83

Pop

47 (46.4)

59 (51.6)

25 (33.0)

131

Rock

83 (94.2)

104 (104.7

79 (67.1)

266

Total

170

189

121

480

x
2

(Oij Eij ) 2

i 1 j 1

= 12.38

Eij

7
8

Decision:
Since 12.38 > 9.488, rejected Ho.
There is sufficient evidence at 0.05 level of significance that music preference and
intelligence are not independent.

B. CHI-SQUARE GOODNESS OF FIT TEST

a test to determine if a population has a specified theoretical distribution.


It is based upon how good a fit we have between the frequency of the occurrence
of observations in an observed sample and the expected frequencies obtained
from the hypothesized distribution.
The Chi Square Test for Goodness of Fit is used to determine if there is a
significant difference bet the observed distribution and the expected distribution.
A goodness-of-fit test between observed and expected frequencies is based on the
quantity

where X is a value of a random variable X whose sampling distribution is


approximated very closely by the chi-square distribution
= observed

and

= expected

If the computed
is small, it indicates a good fit, leading to the acceptance of
H o.
Degrees of Freedom. The number of degrees of freedom in a chi-square goodnessof-fit test is equal to the number of cells minus the number of quantities obtained
from the observed data, which are used in the calculations of the expected
frequencies.

1 Sample Problem
Test the null hypothesis that observed sales do not differ significantly with
expected sales at 0.05 level of significance.
Table 1
Expected and Observed Sales in IBG Food Chain
Fried
Chicken

Expected
30

Observed
30

French Fries
Hamburger
Spaghetti
Palabok

30
30
30
30

34
28
38
20

To solve the problem, we will follow the steps in hypothesis testing:


2 Sample Problem
In 100 tosses of coin, 63 heads and 37 tails are observed. Is this a balance coin?
Use a 0.01 and 0.05 level of significance.

QUIZ:

1 Test the hypothesis whether IQ level is independent of educational attainment among


0.05.
350 respondents. Use the data provided below at
Educational Attainment

2 Test

IQ Level

Secondary

Tertiary

Post Graduate

Low

58

25

12

Average

46

48

26

High

62

41

32

the null

hypothesis that observed sales do not differ significantly with expected sales at
0.05 level of significance.
Expected and Observed Sales in IBG Food Chain
Fried
Chicken
French Fries
Hamburger
Spaghetti
Palabok

Expected
30

Observed
30

30
30
30
30

34
28
38
20

0.01
3 At
, use the following data to test the hypothesis that success in examination
taking is independent of fraternity membership.

Frat Member
Non-Frat Member

Pass
10
417

Fail
14
145

4 A random sample of 200 married men, all retired, were classified according to
education and number of children.
Educati
on
element
ary
seconda
ry
college
Total

Number of children
0 to
2 to
1
3
over 3

Total

14

37

32

83

19
12
45

42
17
96

17
10
59

78
39
200

5 Compute and interpret the correlation coefficient for the following grades of 6
students selected at random:

student

Math grade
(x)

Engish
Grade (y)

1
2
3
4
5
6

70
92
80
74
65
83

74
84
63
87
78
90

Вам также может понравиться