Research II

Chi-square test for independence

Chi-square test
 Its objective is to compare the differences of the sample frequencies with the expected
frequencies.
 The test is applied when you have two categorical variables from a single population. It is
used to determine whether there is a significant association between two variables.
 It is used to (a) goodness-of-fit test, (b) test of independence, and (c) test on difference
among sample proportions.
Chi-square Test for Independence
 Involves two variables and to know whether this variables follows a certain distribution
or not.
 What is tested is the assumption that the two variables are statistically independent.
 Independence implies that knowledge of the category in which an observation is
classified with respect to another variable has no effect on the probability of being in one
of the several categories of the others variables.
 Since two variables are involved, the frequencies are entered in a bivariate table or
contingency table.

Dimension of the table;

r x c where
r = number of rows
c = number of columns

When to use Chi-square test for independence?

Chi-square is appropriate when the following conditions are met:
1. The sampling method simple random sampling
 Refers to any sampling that has the following properties:
 The population consist on N objects.
 The sample consist of n objects.
 If all possible sample of n objects are equally likely to occur, the
sampling method is called simple random sampling.
2. The variables under study are each categorical
  Categorical Variables or Qualitative takes on value that are names or labels.
The color of the ball (e.g., red, green, blue) or the breed of dog (e.g., collie,
shepherd, terrier).
3. If sample data are displayed in a contingency table is a useful tool for examining
relationships between categorical variables. The entries in the cell can be frequency
counts or relative frequencies (just like one-way table).

This approach consists of four steps:

(1) state the hypotheses (HO and HA)
(2) formulate an analysis plan, (use a=.01 - .10)
(3) analyze sample data, and (solve df. fe, X² , p)
(4) interpret results (compare computed and tabular value of p).

Two possibilities in making a decision:

1. If the computed value is greater than the tabular value then reject the null hypothesis and
accept the alternative hypothesis. (This implies that the expected and the frequencies are
not equal).
2. If the computed value is less than the tabular value then accept the null hypothesis and
reject the alternative hypothesis. (this implies that the expected and the observed
frequencies are equal).

If X² computed > X² tabular then reject Ho, accept Ha

If X² computed < X² tabular then accept Ho, reject Ha