Statistic Hypothesis Testing

What is a hypothesis?
How to write down the null and alternative hypothesis
Introduction to the commonly used terms in testing hypothesis.
Steps on Testing Hypothesis

What is a hypothesis?
• A statistical hypothesis is an assumption about a
population parameter .
• This assumption may or may not be true.
• A premise or claim that can be tested.
Example
“The average height of males in the Philippines is 1.7 m.”
A researcher might conduct a statistical experiment to test the

validity of this hypothesis.
What is hypothesis Testing?
• It is the key procedure in inferential statistics
• Hypothesis testing was introduced by Ronald Fisher, Jerzy

Neyman, Karl Pearson and Pearson’s son, Egon Pearson.
• Hypothesis testing is a statistical method that is used in making

statistical decisions using experimental data.
Few Ways on Hypothesis testing
• Traditional method
• P-value method
• Use excel or other statatistics software(easiest)
Steps in Hypothesis Testing (Traditional approach)
• Formulate the hypotheses
• Null Hypothesis
• Alternative Hypothesis
• Set desired significance level - (5% or 0.05) “commonly used”
• Type I error (α) Alpha
• Type II error (β) Beta
• Two tail or one tail Test
• Calculate the degrees of freedom ()
• Find the Critical value (table)
• Choosing the appropriate Test Statistics then calculate the test value(z,t,
anova, chi square, etc.)
• Compare the test value from critical value
• Decide ()
Steps in Hypothesis Testing (P-value approach)
• Formulate the hypotheses
• Null Hypothesis
• Alternative Hypothesis
• Set desired significance level - (5% or 0.05) “commonly used”
• Type I error (α) Alpha
• Type II error (β) Beta
• Identify which test should use (Two tail or one tail Test)
• Choosing the appropriate Test Statistics then calculate the test value (z,t, anova, chi
square, etc.)
• Calculate the degrees of freedom (formula varies)
• Look for the T-test table to get the P-value (estimate), using the df and test value
• Compare the p-value from significance level
• Decide
• if p < α, the test is significant (reject H0)
• If p > α, the test is not significant (failed to reject H0)
Two Hypotheses
Null Hypothesis Alternative Hypothesis
• Denoted as H0 • Denoted as HA or H1
• currently believed to be true • Also called “Research hypothesis”
• Include equality (= or < or > ) • Include inequality ( ≠ or < or >)
• Opposite of what you’re testing • The claim you’re testing

• This is the hypothesis that challenges the null hypothesis
• It is a statement which states that there is no • Can’t be proven true (only to reject the null)
significant relationship • It is a statement which states that there is a significant
or no significant difference between two or difference between the variables
more variables,
or one variable does not affect another or there is a significant difference between two or more
variable variables
or one variable has an effect on the other variable
How to write down null and alternative hypothesis?
Example 1:
“The average age of Youtube users is 25.” A researcher believe that the
average age of Youtube user is not equal to 25.
H0 : μ = 25 null hypothesis “youtube users average age is 25”

H1 : μ ≠ 25 alternative hypothesis “youtube users average age is
not 25”
Note: Note:
We use μ mean Equality and Inequlaity will determine if you’ll be
We use p for proportion using two tail or one tail test. That will be further
for Null H0 we use equality (= or < or > ) explain later.
for Null H1 we use inequality (≠ or < or > )
Example 2
“The researcher wants to test if grade of the students that plays mobile
games has significant difference on the grades of students that don’t play
mobile games.
H0 : μgrade mobile games = μgrade no mobile games - null hypothesis

H1 : μ grade mobile games ≠ μ grade no mobile games - alternative hypothesis
Example 3
“The average grades of students playing mobile games is lower than those
students that don’t play mobile games.”
H0 : μmobile games > μno mobile games - mean of grades of students that plays mobile games is greater than or equal to
the mean grades of students that don’t play mobile games
H1 : μmobile games < μno mobile games – mean grades of that plays mobile games is less than to the mean grades of
students that don’t play mobile games
One-Tailed test Two-Tailed Test
One tailed test ( < or > ) Two tailed test ( ≠ )

greater than,bigger, improved,
more than, increased
Not equal, no difference,
less than, less, lower, decreased

Significance Level
To test the null hypothesis, the researcher must set the level of
significance.
• The level of significance is the probability of making Type I error and

it is denoted by the symbol α “alpha”
• Usually 5% or 0.05 or 5% probability of incorrectly rejecting the null
hypothesis is acceptable
• Selected Cut off value (how strict)
• Critical Value(traditional)
Type I error
Degrees of freedom
• The degrees of freedom depicts the number of pieces of independent
information available for computinf variability. Fo rany statistical used
in testing hypothesis, required degrees of freedom varies depending
on the size of the distribution.
For single group of population
df = N - 1
For two groups
df = (N1 + N2 )- 2
P-value
• Probability value; indicates how likely it is that result occurred by chance
alone.
• If P value is small; it indicates the result was unlikely to have occurred by
chance alone. These results are known as being statistically significant.
• Small p value = greater chance alone(somthinng happened) test is

significant
• Large p value = result is within chance or normal sampling error (nothing
happened) test is not significant
• Range is 0-1
T Test for Depedent
T-Test for Dependent

Statistic Hypothesis Testing

Uploaded by

Copyright:

Available Formats

Statistic Hypothesis Testing

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Statistic Hypothesis Testing

Uploaded by

Copyright:

Available Formats

What is a hypothesis?

How to write down the null and alternative hypothesis

Introduction to the commonly used terms in testing hypothesis.

Steps on Testing Hypothesis

A researcher might conduct a statistical experiment to test the

• Hypothesis testing was introduced by Ronald Fisher, Jerzy

• Hypothesis testing is a statistical method that is used in making

• Opposite of what you’re testing • The claim you’re testing

H0 : μ = 25 null hypothesis “youtube users average age is 25”

H0 : μgrade mobile games = μgrade no mobile games - null hypothesis

One tailed test ( < or > ) Two tailed test ( ≠ )

less than, less, lower, decreased

• The level of significance is the probability of making Type I error and

• Selected Cut off value (how strict)

• Small p value = greater chance alone(somthinng happened) test is

You might also like