Sunteți pe pagina 1din 15

HYPOTHESIS

TESTING
ERVIN C. REYES
HYPOTHESIS TESTING

Hypothesis Testing is a procedure


which is used to substantiate or
invalidate the null hypothesis It deals
with verifying a claim or conjecture
about a parameter or distribution of
the population.
LEVEL OF SIGNIFICANT
OF A TEST
It is the probability of committing
the type 1 error in deciding whether
you are going to accept or reject the
null hypothesis. It is also known as
the alpha level of a statistical test.
STEPS IN HYPOTHESIS TESTING

1. State the null hypothesis.


2. State the corresponding alternative hypothesis.
3. Set the level of significance.
4. Compute for the value of statistics.
5. Determine the critical value.
6. Compare the computed and the critical value.
Decision Rule: If the absolute computed
value is equal to or greater than the
absolute critical value, REJECT the null
hypothesis. Otherwise accept it.

7. State the implication of your decision


TEST OF SIGNIFICANT
DIFFERENCE
1. Z – test is used to find out if there is a
significant difference between: (n ≥ 30)
2. T – test is used to find out if there is a
significant difference between:(n < 30)
3. F –test is used to find out if there is a
significant difference among three or more
groups. ANOVA
TEST OF SIGNIFICANT
RELATIONSHIP
1. Chi – square – used in testing the significance of the
relationship between two nominal variables.
2. Spearman’s Rank Order Correlation – used in
testing significance of the relationship between two
ordinal variables.
3. Pearson’s Product Moment Correlation – used in
testing the significance of the relationship between two
variables in the interval/ratio scales.
EXAMPLE
Ho: The mean weight before of overweight individuals is the
same as the mean weight after attending the diet program. Weight Weight
before after
(kg) (kg)
Ha: The mean weight before of overweight individuals is
higher than the mean weight after attending the diet
program. 1 100 93
Test Procedure: t-test for related samples
2 89 85
Decision Rule:
3 83 74
Reject Ho when p-value ≤ α
4 98 94
5 108 105
By: C. Vista 6 95 88
THREE MAIN TYPES OF T-TEST:

• An Independent Samples t-test compares the


means for two groups.
• A Paired sample t-test compares means from the
same group at different times (say, one year apart).
• A One sample t-test tests the mean of a single
group against a known mean.
1. Choose Stat>Basic Statistics>Paired t...
2. Click in the First Sample text box and specify before.
3. Click in the Second Sample text box and specify After.
4. Click in the Options button.
5. Make sure the Test Mean text box says 0.
6. Click the arrow button at the right of the Alternative drop-
down list box and select not equal.
7. Click OK.
8. Click OK
N Mean StDev SE Mean
Before 6 95.5000 8.7350 3.5660
After 6 89.8333 10.3425 4.2223
Difference 6 5.66667 2.33809 0.95452

95% CI for mean difference: (3.21299,


8.12034)
T-Test of mean difference = 0 (vs not = 0):
T-Value = 5.94 P-Value = 0.002
“I couldn't claim that I was
smarter than sixty-five other
guys--but the average of sixty-
five other guys, certainly!”
― Richard P. Feynman,
THANK YOU!!!

S-ar putea să vă placă și