Scribd
Upload
8 views5 pages

Statistical Hypothesis Testing Examples

The document presents three statistical hypothesis tests conducted by a trainee under Professor Jha Jayant. The first test examines if the average blood pressure of patients differs from 120 mmHg, concluding no significant difference; the second compares exercise times between two groups, also finding no significant difference; the third test evaluates a diet program's effect on weight, revealing significant weight loss. Overall, the findings indicate that while blood pressure and exercise times are statistically similar, the diet program has a significant impact on weight reduction.

Uploaded by

Robinson Morey
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views5 pages

Statistical Hypothesis Testing Examples

The document presents three statistical hypothesis tests conducted by a trainee under Professor Jha Jayant. The first test examines if the average blood pressure of patients differs from 120 mmHg, concluding no significant difference; the second compares exercise times between two groups, also finding no significant difference; the third test evaluates a diet program's effect on weight, revealing significant weight loss. Overall, the findings indicate that while blood pressure and exercise times are statistically similar, the diet program has a significant impact on weight reduction.

Uploaded by

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

FINAL ASSIGNMENT

Statistical Methods

Professor: Jha Jayant Trainee: Robinson Morey

February, 2025

Exercise 1: One-Sample Hypothesis Test

Problem Statement: A health researcher wants to test whether the average


blood pressure of a group of patients differs from the standard population mean
of 120 mmHg. A random sample of 35 patients is taken, and their mean blood
pressure is found to be 118 mmHg with a population standard deviation of 10
mmHg.

Hypotheses:

 Null Hypothesis (Ho): μ=120 (The average blood pressure is 120 mmHg)

 Alternative Hypothesis (Ha): μ≠120 (The average blood pressure is different)

Test Statistic: Using the z-score formula for a single sample:

 =118 (Sample mean)


 μ =120 (Population mean)
 σ = (Population standard deviation)
 n =35 (Sample size)
Decision Rule: For a 95% confidence level, the critical values for a two-tailed
test are z=±1.96. Since -1.18 is within the range (−1.96,1.96), we fail to reject
Ho. There is no significant evidence to conclude that the mean blood pressure
differs from 120 mmHg.

Exercise 2: Two-Sample Hypothesis Test

Problem Statement: A fitness coach compares the weekly exercise time of two
groups: Group 1 (40 people) follows a new exercise routine, and Group 2 (50
people) follows a traditional exercise plan. The mean weekly exercise time for
Group 1 is 5.5 hours with a standard deviation of 1.5 hours, while for Group 2,
it is 6 hours with the same standard deviation.

Hypotheses:

 Ho: μ1= μ2 (Both groups have the same exercise time)

 Ha: μ1≠ μ2 (The exercise times are different)


Test Statistic: For two independent samples, the z-score formula is:

Decision Rule: For α=0.05, the critical values are z=±1.96. Since -1.62 is within
the acceptance range (−1.96,1.96), we fail to reject Ho. There is no significant
evidence that the new exercise routine results in a different exercise time.
Exercise 3: Paired-Sample Hypothesis Test

Problem Statement: A dietitian wants to analyze whether a new diet program


significantly reduces weight. A group of 250 individuals had their weight
recorded before and after following the diet for 8 weeks. The weight differences
(before - after) for each individual were measured, and the summary statistics
are:

 Sample mean of weight differences: =3 kg

 Sample standard deviation of weight differences: sd= 4 kg

 Sample size: n= 250

The dietitian wants to test whether the diet results in a significant weight
change.

Hypotheses

We set up the hypotheses for a paired t-test:

 Null Hypothesis (Ho): μd= 0 (The diet does not cause a significant weight
change)

 Alternative Hypothesis (Ha): μd≠ 0 (The diet does cause a significant


weight change)

Since this is a two-tailed test, we check for significant changes in both


directions.

Test Statistic (t-Score Calculation)

The formula for the t-score in a paired-sample test is:

 =3 kg (Sample mean of differences)


 μd= 00 kg (Null hypothesis assumes no change)
 sd=4 kg (Sample standard deviation)
 n= 250 (Sample size)
Since t=11.86 is very large, it indicates that the weight difference is statistically
significant. We would reject the null hypothesis Ho (that the diet had no effect).

This means the diet program significantly changed the participants' weight.

Using Z-Score:

Test Statistic: For paired samples, the z-score formula is:

Substituting the values:

Decision Rule: For α=0.05, the critical values are z=±1.96. Since the computed
z-score = 11.86 is far outside the acceptance region (−1.96,1.96), we reject the
null hypothesis Ho. There is strong statistical evidence that the diet program
significantly changes weight.

You might also like