Chi Square Test of Proportion

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

GUIMARAS STATE COLLEGE

GRADUATE SCHOOL

CHI SQUARE
The Chi – Square is the most commonly used nonparametric test, it is used with the ordinal
data in the form of frequencies or proportions. The observed frequencies are compared to the
expected frequencies and the differences are tested at a desired level of significance. It can be used
to determine whether the frequencies we observed previously differ significantly from an even
distribution (or any other distribution we might hypothesize).
Basic requirements of the chi-square test

❖ Assumption #1: You have one dependent variable that is measured at


the dichotomous level (i.e., the variable has two categorical, independent
groups). Dichotomous variables can be nominal or ordinal, meaning that the independent
groups are either unordered or ordered. Examples of dichotomous variables that
are unordered (i.e., nominal variables with just two independent groups) include gender
(two groups: "males" and "females"), employment status (two groups: "employed" and
"unemployed"), transport type (two groups: "bus" and "car"), smoker (two groups: "yes"
and "no") and experimental treatment (two groups: "intervention" and "control").
Examples of dichotomous variables that are ordered (i.e., ordinal variables with just two
independent groups) include risk (two groups: "low" and "high"), IQ (two groups: "above
average" and "below average"), age group (two groups: "under 18 years old" and "18
years or older") and exam performance (two groups: "passed" and "failed").

❖ Assumption #2: You have one independent variable that has three or more
categorical, independent groups (i.e., known as a polytomous variable). Polytomous
variables can be also be either nominal or ordinal. The word polytomous simply means
that there are three or more independent groups. Examples of polytomous variables that
are unordered (i.e., nominal variables with three or more independent groups) include
ethnicity (three groups: "African American", "Caucasian" and "Hispanic"), transport type
(four groups: "cycle", "bus", "car" and "train") and profession (five groups: "consultant",
"doctor", "engineer", "pilot" and "scientist"). Examples of polytomous variables that
are ordered (i.e., ordinal variables with three or more independent groups) include risk
(three groups: "low", "medium" and "high"), physical activity level (four groups:
"sedentary", "low", "moderate" and "high") and exam performance (five groups: "under 40
marks", "40-49 marks", "50-59 marks", "60-69 marks" and "70 marks and over").

❖ Assumption #3: You should have independence of observations, which means that there is
no relationship between the observations in each group of the independent variable or
between the groups themselves. Indeed, an important distinction is made in statistics
when comparing values from either different individuals or from the same individuals.
Independent groups (in a chi-square test of homogeneity) are groups where there is no
relationship between the participants in either of the groups. Most often, this occurs
simply by having different participants in each group.

1|P age
For example, if you split a group of individuals into two groups based on their gender
(i.e., a male group and a female group), no one in the female group can be in the male
group and vice versa. As another example, you might randomly assign participants to
either a control trial or an intervention trial.

Example:
Suppose your instructor returns the exam and hands out the answer key. You construct a
frequency distribution of the correct responses to the 50 – item test as follows:
A-12, B-14, C-9, D-5, E-10
Thus, 12 of the 50 items had a correct answer of A, 14 had a correct answer of B, and so on.
These are called the observed frequencies (f0). Observed frequencies obtained in an actual
frequency. Test if the instructor has a tendency to assign any particular correct response from A to E.

Step 1:
Ho: The instructor shows no tendency to assign any particular correct response from A to E.
Ha: The instructor shows a tendency to assign particular correct responses from A to E.

Step 2:
Level of Significance = .05, two-tailed

Step 3:
Chi-square test:
Tabular value = 9.49
Note: This value was based on the df = k - 1, where k is the number of categories in the observed
frequency distribution (df=5-1).

Step 4:
Do not reject the null hypothesis if the computed value is -9.488<X2<9.488

Step 5:
Compute
Formula:

Where fo = observed frequency in any category


fe = expected frequency in any category

A 12 10 2 4 .4
B 14 10 4 16 1.6
C 9 10 -1 1 .1
D 5 10 -5 25 2.5
E 10 10 0 0 0

2|P age
X2 = 4.6

Step 6:
Conclusion
The computed X2 value 4.6 is less than 9.49. Thus, we do not reject the null hypothesis. This
means that the instructor shows no tendency to assign any particular correct response from A to E.

Problem set 8:
Suppose a die is thrown 180 times by a Casino manager to see if there was a difference in the
outcomes of the six faces of a die. The outcomes was recorded as follows; 1 appears 25 times, 2-40
times, 3-37 times, 4-22 times, 5-24 times, 6-32 times. Test at .01 level of significance. Follow the
steps in testing hypothesis.

3|P age

You might also like