Scribd
Upload
46 views

Lecture 4 - Hypothesis Testing

- The document discusses hypothesis testing, including defining common terminology, performing hypothesis tests using distribution analysis and t-tests, types of errors, how sample size and effect size influence results, and comparing the alpha level to the p-value. - It provides examples comparing coin flips to hypothesis testing and illustrates key concepts like the t-test, confidence intervals, and how hypothesis testing is performed.

Uploaded by

Ayodele Ogundipe
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
46 views

Lecture 4 - Hypothesis Testing

- The document discusses hypothesis testing, including defining common terminology, performing hypothesis tests using distribution analysis and t-tests, types of errors, how sample size and effect size influence results, and comparing the alpha level to the p-value. - It provides examples comparing coin flips to hypothesis testing and illustrates key concepts like the t-test, confidence intervals, and how hypothesis testing is performed.

Uploaded by

Ayodele Ogundipe
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 23

Hypothesis Testing

Dr Kamran Mahroof
Associate Professor
School of Management
Faculty of Management, Law and Social Sciences
University of Bradford, United Kingdom

1 25/12/22
Objectives
– Define some common terminology related
to hypothesis testing.
– Perform hypothesis testing using the Distribution Analysis and t
Test tasks.
Judicial Analogy
Coin Example

T T
H
H
2.04 Poll
• If you have a fair coin and flip it 100 times, is it possible for it to
land on heads 100 times?
•  Yes
•  No
2.04 Poll – Correct Answer
• If you have a fair coin and flip it 100 times, is it possible for it to
land on heads 100 times?
•  Yes
•  No
Coin Analogy
• You used a decision rule to make a decision, but
was the decision correct?

Types of Errors

ACTUAL H0 Is True H0 Is False


DECISION
Fail to Reject Null Correct Type II Error
Reject Null Type I Error Correct
Coin Experiment – Effect Size Influence
• Flip a coin 100 times and decide whether it is fair.

55 Heads 40 Heads
45 Tails 60 Tails
p-value=.3682 p-value=.0569

37 Heads 15 Heads
63 Tails 85 Tails
p-value=.0120 p-value<.0001
Coin Experiment – Sample Size Influence
• Flip a coin and get 40% heads and decide whether it is fair.
Statistical Hypothesis Test
Comparing  and the p-Value
• In general, you
– reject the null hypothesis if p-value < 
– fail to reject the null hypothesis if p-value  .
2.05 Multiple Choice Poll
• Which of the following affects alpha?
a. The p-value of the test
b. The sample size
c. The number of Type I errors
d. All of the above
e. Answers a and b only
f. None of the above
2.05 Multiple Choice Poll – Correct Answer
• Which of the following affects alpha?
a. The p-value of the test
b. The sample size
c. The number of Type I errors
d. All of the above
e. Answers a and b only
f. None of the above
Performing a Hypothesis Test
• To test the null hypothesis H0:  = 0, SAS software calculates the Student’s t
statistic value:
( x   0)
t
sx
• For the test score example:

t  (1190.6251200) -0.5702
16.4416
• The null hypothesis is rejected when the calculated value
is more extreme (either positive or negative) than would be expected by
chance if H0 were true.
Performing a Hypothesis Test

The t statistic can be positive or negative.


The t Test Task
Confidence Interval Plots

95% CL Std
Mean 95% CL Mean Std Dev Dev
1190.6 1157.9 1223.4 147.1 127.3 174.2
Hypothesis Testing

This demonstration illustrates the concepts


discussed previously.
Exercise
This exercise reinforces the concepts
discussed previously.
2.06 Multiple Choice Poll
• A 95% confidence interval for SAT scores is (1157.90, 1223.35).
From this, what can you conclude, at alpha=0.05?
a.The true average SAT score is significantly different from 1200.
b.The true average SAT score is not significantly different from
1200.
c. The true average SAT score is less than 1200.
d.None of the above – You cannot determine statistical significance
from confidence intervals.
2.06 Multiple Choice Poll – Correct Answer
• A 95% confidence interval for SAT scores is (1157.90, 1223.35).
From this, what can you conclude, at alpha=0.05?
a.The true average SAT score is significantly different from 1200.
b.The true average SAT score is not significantly different from
1200.
c. The true average SAT score is less than 1200.
d.None of the above – You cannot determine statistical significance
from confidence intervals.

You might also like