Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Statistical
Methods
• There are experiments that yield
qualitative responses which can
either be ranked (ordered) or
categorized (classified). These
measurements like competence,
palatability, smell, texture,
appeal and acceptability of a
product.
• When an evaluator or a judge is
assigned to evaluate and rank the
taste and acceptability of four brands
of coffee, it will be impossible to give
an exact measure of the taste and
acceptability of the coffee products.
But it will be easy for the evaluator to
rank them as the best, second best
and so on.
1. The Chi-Square (X2) One
Variable Case
• The Chi-square test is also called the
“goodness of fit test.” It answers the
question. How well does our observed
distribution fit the theoretical
distribution?” The data used in this test
are the frequencies obtained for the
different categories.
a. The null hypothesis is
There is no significant difference
between the observed frequencies
and the expected frequencies.
b. The alternative hypothesis is
The observed frequencies are
different from the expected
frequencies
c. Test the null hypothesis using the formula:
Where
fo = the observed frequency in a given
category
fe = the expected frequencies in a given
category
• To test Ho, find the X2
at degrees of freedom
=k-1 and alpha at 5 %
or 1%.
• Decide whether to reject or
accept the null hypothesis.
Reject Ho if the computed value
of chi-square is equal or greater
than the critical value.
Otherwise, accept it
Example
• A marketing researcher would like to know
whether the color of packaging is a determinant
of consumer selection among housewives. They
were made to select among soaps packaged in
three different colors, red, white, and brown. The
following data were obtained:
Color Red White Brown
Number of 200 300 100
housewives
(fo) selecting a
colored
package
• There are three 3 gates of the University of the
East. The building maintenance supervisor
would like to know if the gates are equally
utilized. As an experiment, 600 students are
observed as they enter the school.
• The number of students using each gate is
reported below. At .01 significance level, can we
conclude that there is a difference in the use of
three gates?
Gate Number of Students
Recto 245
Lepanto 205
Gastambide 150
• Smilingbee, a fast growing food corporation,
would like to know whether there is uniformity in
the sales of their fried chicken from Monday to
Sunday. They listed the number of fried chicken
sold from Monday to Sunday in one of their
outlets in Manila. Use the Chi-square goodness
of fit test. For the level of significance, use alpha
1%.
Day Mon Tue Wed Thu Fri Sat Sun
Responses Percentage
Excellent 30
Good 40
Fair 20
Poor 10
The manager of the canteen would like to
know whether the data gathered by the
student apply to the professor of the
University. He asked randomy selected 60
professors.The following table shows the
data he collected.
Responses F
Excellent 5
Good 15
Fair 20
Poor 10
Evaluation:
A TV station classifies its programs into five
broad categories and based on past surveys,
the ratings in audience viewers are 30% soap
operas and sitcoms; 20% news and current
affairs; 10% sports events; 30% variety shows;
10% talk shows. A sample of 300 viewers is
taken with the following results on viewership:
122, 45, 28, 80 and 25 respectively. At 1% LOS,
test to determine if there was a significant
change in audience viewership.
2. The Chi-Square test of
independence
• They deal with two or more categories
• Data consist of frequency counts tabulated
and placed in appropriate cells in
contingency table
• It is not very obvious to see immediately
how to assign expected frequency to each
category
• The null hypothesis states that each
cell probability will equal the product
of its respective row and column
probabilities that implies
independence of the two
classification.
• The alternative hypothesis states that
the two classifications are dependent
on one another
TABLE OF OBSERVED FREQUENCIES
a1 a2 a3 G
b1 b2 b3 H
c1 c2 c3 I
D E F T
TABLE OF EXPECTED FREQUENCIES