Scribd
Upload
56 views

Hypothesis Testing

1. Hypothesis testing is a statistical procedure used to assess whether data from a sample is compatible with a null hypothesis. It involves specifying a null hypothesis and alternative hypothesis, choosing a significance level, computing a test statistic and p-value, and making a decision to accept or reject the null hypothesis. 2. The null hypothesis states that there is no difference or relationship between variables, while the alternative hypothesis states there is a difference. Type I and Type II errors can occur when deciding whether to accept or reject the null hypothesis. 3. A p-value less than the predetermined significance level, typically 5%, provides sufficient evidence to reject the null hypothesis in favor of the alternative hypothesis.

Uploaded by

prabhjot singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
56 views

Hypothesis Testing

1. Hypothesis testing is a statistical procedure used to assess whether data from a sample is compatible with a null hypothesis. It involves specifying a null hypothesis and alternative hypothesis, choosing a significance level, computing a test statistic and p-value, and making a decision to accept or reject the null hypothesis. 2. The null hypothesis states that there is no difference or relationship between variables, while the alternative hypothesis states there is a difference. Type I and Type II errors can occur when deciding whether to accept or reject the null hypothesis. 3. A p-value less than the predetermined significance level, typically 5%, provides sufficient evidence to reject the null hypothesis in favor of the alternative hypothesis.

Uploaded by

prabhjot singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

Hypothesis Testing

Epidemiology and Biostatistics


BPH III
Department of Community Medicine and School of Public Health
PGIMER, Chandigarh
Hypothesis testing
• A statistical procedure to assess the compatibility of data in a
sample with the null hypothesis
• A hypothesis is a tentative idea that is postulated to be tested.
• Cannot be accepted until scientifically proven true.
• Generates p-value, which gives the probability of the observed
data, or more extreme data if the null hypothesis were true
For example
• It can be hypothesised that there is an increased incidence of
colorectal cancer among diabetics. (Researcher 95% confident)
• This idea cannot accepted as is unless evidence from a scientific
research justifies the claim.
• A hypothesis is a statement about the population from which samples
are drawn.
• And generate sufficient evidence
Types of Hypothesis

Hypothesis

Alternative
Null Hypothesis
Hypothesis
Alternative Hypothesis
• Denoted by H1 or HA
• Significant difference and difference is caused by non random chance.
• Alternate always contradicts the Null Hypothesis
• Also know as the “Research Hypothesis”.
• Out of Null and Alternate hypothesis only one can be true.

Example- In discussed example the alternative hypothesis can be stated


as “the proportion of cases developing colorectal cancers is higher
among diabetics in comparison to those of non- diabetics.
Null Hypothesis
• Denoted by Ho
• States there is no significant difference
• Absence of difference, association or effects.
• Statement which the researcher wishes to get rejected
• Null Hypothesis when testing for a single mean:“ There is no
statistically significant difference between sample mean and
population mean, thus observed differences have been obtained just
by chance”

Example- In discussed example the null hypothesis can be stated as,


“there is no difference between the proportion of cases developing
colorectal cancer in diabetic and non diabetic groups”.
Types of Error
In process of Hypothesis testing we encounter certain error that are;
Example
New Drug A Old Drug B
Ho - No difference between two drugs in treatment of disease.
HA- New drug is better than old drug
Null Hypothesis

True False
Result of How much error to be
Reject Type I Error No Error
Clinical allowed that is denoted by p
Trial Accept No Error Type II Error value.

Type I error – False Positive


Type II error- False Negative
• P value is probability of type I error
• Normally taken 5% it is significance level also called alpha
• 5% or 1% is set as cut off for rejecting null hypothesis.
• At 5% level of significance the probability of type I error is 5 out 100
times or chance of making correct decision are 95 times out of 100.
• Smaller is p value more is the evidence for rejecting null hypothesis
• If p value <0.05 means results are statistically significant
Steps
I. Specifying the null and alternative hypothesis
II. Choosing the level of significance
III.Compute the test statistic value
IV. Perform the statistical test, i.e., calculate probability of obtaining
data you got if null hypothesis were true.
V. Make a decision about the hypothesis- accept or reject the null
hypothesis
VI. Conclusion
1. State H0 and HA – Smoking & Lung
Cancer Study

• H0: there is no association between smoking and lung cancer (RR=1)


• HA: there is a association between smoking and lung cancer (RR ≠ 1)
2. Choosing a Statistical Test
Choice depends on:
• study design
• measurement scale of the variables
• study size

• Test for comparison of 2 means: Student t-test


• Test for comparison of 2 proportions: Chi-square test
3. Specify a Level of Significance
• Level of significance = an arbitrary cut-off, a small probability, for
deciding whether to declare the null hypothesis untenable

• Also called alpha level

• Commonly, alpha set at 0.05 (5%) or 0.01 (1%)


4. Perform the Statistical Test, Compute p-value

• Chi-square tests provide chi-square test statistic, which must be


converted to P-value (use computer or look-up table)
• P-value = probability of observing a difference as great or greater than
the observed difference, if the null hypothesis were true
• P-value influenced by:
– size of difference / strength of association
– size of the sample
5. Make Decision about Hypothesis

• If computed P-value < alpha, reject H0, i.e., conclude that there is a
difference/effect/relationship which is unlikely to be due to chance*

• If computed P-value > alpha, do not reject H0, i.e., conclude that
difference/effect/relationship could be due to chance*

* You could be right or you could be wrong!


6. Make Decision about Hypothesis
A decision rule is used to accept or reject the null hypothesis
• P – value
P<α
Reject the null hypothesis
Statistically significant
• Test statistic
Test statistic (calculated value) < Table value of α
Accept the null hypothesis
Statistically insignificant
Thank You

You might also like