Chi-Square Test Explained: Steps & Examples
Uploaded by
ScribdTranslationsChi-Square Test Explained: Steps & Examples
Uploaded by
ScribdTranslationsCHI-SQUARE TEST
2
Chi-Squared ( is the name of a test of
hypothesis that determines whether two variables are
related or not.
Steps:
Make a conjecture.
2)Escribir la hipótesis nula y la alternativa.
2
.
3) Calculate the value ofcalc
4) Determine the value of p and the degree of freedom.
5) Obtain the critical value.
6) Make a comparison between the chi-square
calculated and the critical value.
7) Interpret the comparison.
1
CONTINGENCY TABLE
It is the table that contains the counted obtained data and
organized.
Example:
USE OF SEAT BELT
SECURITY
GENRE YES NO
FEMININE 50 25
MALE 40 45
2
FORMULATION OF HYPOTHESES
NULA (H0It is the one in which it is ensured that both
analyzed parameters are independent of each other
another.
ALTERNATIVE (H1It is the one in which it is assured
that the two analyzed parameters are indeed
dependent.
3
EXAMPLE
Melissa conjectures that the use of seat belts,
In drivers, it is related to gender.
H0The use of the seatbelt is independent
of the genre.
H1The use of the seatbelt is not
regardless of gender.
4
FREQUENCY TABLE
EXPECTED
To calculate each and every value in the table
expected frequencies are performed:
TotalColumn Paradichacelda Row Paradichacelda
Total
SumTotal
5
CREATE A TABLE WITH THE VALUES OF THE CONTINGENCY TABLE AND
ADD A ROW AT THE BOTTOM AND A COLUMN AT THE SIDE
RIGHT.
50 25
40 45
6
Perform the sums by rows, by columns, and the total sum
50 25 75
40 45 85
SUM OF ROWS
FREQUENCIES OF
VALUES
OBSERVED
90 70 160
SUM OF COLUMNS TOTAL SUM
7
Use the formula to obtain the expected frequencies.
90 75 70 75
42.1875 32.8125
160 160
90 85 70 85
47.8125 37.1875
160 160
EXPECTED VALUE FREQUENCIES
8
CALCULATED CHI-SQUARE
To obtain the value of the Calculated Chi-Squared, one ...
it has the formula
2
2 f0e
f
calc
fe
f:0 Frequency of the observed value.
f:e Frequency of the expected value.
9
EXAMPLE
TABLE OF OBSERVED VALUES TABLE OF EXPECTED VALUES
50 25 42.1875 32.8125
40 45 47.8125 37.1875
2 f0e
f 2
calc
fe
2 50 42.1875 2 25
32.8125
2
40 47.8125 2
45 37.1875 2
calc
42.1875 32.8125 47.8125 37.1875
2 1.4468 1.8601 1.2766 1.6413 6.2248
calc
10
DEGREE OF FREEDOM v
To calculate the degree of freedom (v), the following is done:
v NumberofrowsNumber of columns1
11
EXAMPLE
TABLE OF OBSERVED VALUES
50 25
40 45
v 2 1 2 1
v 1 1 1
12
LEVEL OF SIGNIFICANCE
It is the error that can be made when rejecting the
null hypothesis being true.
Generally, a significance level is used.
0.05, which indicates that there is a probability of 0.95
that the null hypothesis is true.
13
EXAMPLE
Melissa conjectures that the use of seat belts,
in drivers, it is related to gender. The
data is shown in the table below. Melissa performs the
test of your conjecture using chi-square with a
significance level of 1%.
USE OF SEAT BELT
SECURITY
GENRE YES NO
Feminine 50 25
MALE 40 45
So, there is a significance level of 0.01.
14
VALUE OF THE PARAMETER p
To calculate the value of p, the following is done:
p 1 Level of significance
Example:
p 1 0.01 0.99
15
TABLE FOR VALUES OF
CRITICAL CHI-SQUARE
16
EXAMPLE
17
COMPARISON BETWEEN THE
CHI-SQUARE VALUES
CALCULATED AND THE CRITIC
If the calculated chi-square value is less than or equal to
that the critical chi-square is then accepted
null hypothesis, otherwise it is not accepted.
Example:
2 Critical value
calc
6.2248 6.635
Then the null hypothesis is accepted, which is 'The use
the seatbelt is independent of gender.
18
Example 1
[Link]
[Link]
19
EXERCISE 1
20
21
22
EXERCISE 2
23
24