X2 table

1
 Chapter six
Chi-Square Distributions (x2)
1.1. Areas of application
In previous chapters, we have based our decisions on the assumption that the population was
normally distributed. When this assumption about the population cannot be made, then it
becomes necessary to use other procedures. One of the tests used in such situations is known as
Chi square (X2) test. This test is good for nominal or ordinal scale of measurement.

X2 distribution has the following characteristics

1. It involves squared observations and hence it is always positive.
2. The distribution is skewed to the right so that its skewness is positive.
3. Similar to t distribution, there is a family of chi-square distributions. There is a particular
distribution for each degree of freedom.
The degree of freedom (df) are calculated as follows;

a. df=(k-1), where K is the number of categories in one sample test

b. df=(K-1)(r-1),where K is the number of columns and r is the number of rows in a cross
classification table(contingency table) for various categories of two or more independent
samples.

1.2. Tests for independence between two variables

The chi-square independence test can be used to test the independence of two variables.
Steps:
1. State the null and alternative hypothesis
H0: the column variable is independent of the row variable
H1: the column variable is not independent of the row variable

1
2. Select a random sample and record the observed frequencies
3. Compute expected frequencies
4. Compute the value of the test statistic
5. Decision rule: Reject H0 if X2≥ X2α

Illustration: Let us illustrate the test of independence by considering the study conducted by the
MOHA soft drink company. MOHA manufactures and distributes three types of Soft drink:
Mirinda, 7 UP, and Pepsi. In an analysis of the market segments for the three Soft drinks, the
firm’s market research group raised the question of whether preferences for the three Soft drinks
differ among male and female drinkers. If soft drinks preference is independent of the gender of
the drinker, one advertising campaign will be initiated for all of MOHA’s soft drink. However, if
soft drink preference depends on the gender of the drinker, the firm will tailor its promotions to
different target markets.

Step 1:
H0: Soft drink preference is independent of the gender of the drinker.
H1: Soft drink preference is not independent of the gender of the drinker.
Step 2: After identification of the population as all male and female drinkers, a sample selected
and each individual asked to state his or her preference for the three soft drinks.
Contingency table* for soft drink preference and gender
GENDER SOFT DRINK PREFERENCE

Mirinda 7 UP Pepsi Total

Male 20 40 20 80
Female 30 30 10 70
Total 50 70 30 150
*When data are arranged in table form for the chi-square test, the table is called a contingency table.

Step 3: Expected Frequencies

GENDE SOFT DRINK PREFERENCE
R
Mirin 7 UP Pep Total
da si
Male 26.67 37.33 16 80
Female 23.33 32.67 14 70
Total 50 70 30 150

Expected frequency = (Row total)(Column total)

2
 Sample size

Step 4:

Computation of the chi square test statistic

Gender Preference Observed Expected fij - eij (fij - eij)2 (fij - eij)2
frequency(fij) frequency(eij) eij
Male Mirinda 20 26.67 -6.67 44.44 1.67
Male 7 UP 40 37.33 2.67 7.11 0.19
Male Pepsi 20 16 4.00 16.00 1.00
Female Mirinda 30 23.33 6.67 44.44 1.90
Female 7 UP 30 32.67 -2.67 7.11 0.22
Female Pepsi 10 14 4.00 16 1.14
Total 150 X2=6.12
Step 5: Degree of freedom= (2-1)(3-1)= 2

X20.05 at degree of freedom=2 is 5.991, X2 > X2α , therefore Reject H0. The preference is not
independent of the gender of the drinker.

1.3. Tests for the equality of several proportions

In this situation, samples are selected from several different populations, and the researcher is
interested in determining whether the proportions of elements that have a common characteristic
are the same for each population.

Illustration: A psychologist selected 100 people from each of four income groups and asked
them if they were “very happy.” The percent for each group who responded yes and the number
from the survey are shown in the table. At a 0.05 level of significance test the claim that there is
no difference in the proportions.

Household Less than 30,000 – 75,000 – 100,000 or Total

income 30,000 (24%) 74,999 (33%) 99,999 (38%) more (49%)
Yes 24 33 38 49 144

3
 No 76 67 62 51 256
Total 100 100 100 100 400
Solution
Step 1: State the hypotheses and identify the claim.
H0: p1 = p2 = p3 = p4
H1: At least one proportion differs from the others.
Step 2: Find the critical value. (R - 1)(C - 1) = (2 - 1)(4 - 1) = 3. The critical value is 7.815.
Step 3: Compute the test value.
EXPECTED FREQUENCIES
Household Less than 30,000 – 75,000 – 100,000 or Total
income 30,000 (24%) 74,999 (33%) 99,999 (38%) more (49%)
Yes 36 36 36 36 144
No 64 64 64 64 256
Total 100 100 100 100 400

Calculation of X2

Step 4: Make the decision reject the null hypothesis since 14.149 > 7.815.

1.4. Goodness- of fit tests

The X2 test is used to test whether there is a significance difference between the observed number
of responses in each category and the expected number of responses for such category under the
assumptions of null hypothesis. In other words, the objective is to find out how well the
distribution of observed frequencies (fo) fit the distribution of expected frequencies (fe).

Illustration: Assume that a shoe manufacturer company sold 30 pairs of shoes. There is six size
of shoes and the company expects the size of shoes are sold in equal proportion, that is each size
is sold 5 times. In the survey conducted, the actual number of frequency each size sold in
sequence is as follows; 4, 7, 3, 6, 8, 2. A comparison of observed frequency and expected
frequency is tabulated below;

4
 Size
A B C D E F
Observed frequency 4 7 3 6 8 2
Expected frequency 5 5 5 5 5 5

Steps involved in X2 test

Step 1: H0: All sizes are equally likely to occur.
P1=P2=P3=P4=P5=P6
H1: All probabilities are not equally likely to occur.
Step 2: Assume α =0.05 .
Step 3: Expected frequency (fe) = 5.
Step 4:
X2 = (4-5)2/5 + (7-5)2/5 + (3-5)2/5 + (6-5)2/5 + (8-5)2/5 +(2-5)2/5
=1/5 + 4/5 + 4/5 + 1/5 + 9/5 +9/5
= 28/5 = 5.6
Step 5: Critical value of X20.05 and df = (K-1) = (6-1) = 5 is given as 11.07.
Since our computed value of X2 = 5.6 is less than the critical value of X2=11.07, we can
accept the null hypothesis.