Scribd
Upload
14 views

Power Analysis Notes-1

The document discusses power analysis in statistical testing, focusing on Type I and Type II errors, and the importance of minimizing these errors through proper experimental design. It introduces G*Power software for calculating power and sample size, illustrated with an example of a clinical dietician comparing two diets for diabetic patients. Key assumptions for conducting power analysis include expected differences, standard deviations, alpha levels, and desired power levels.

Uploaded by

troyautin
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
14 views

Power Analysis Notes-1

The document discusses power analysis in statistical testing, focusing on Type I and Type II errors, and the importance of minimizing these errors through proper experimental design. It introduces G*Power software for calculating power and sample size, illustrated with an example of a clinical dietician comparing two diets for diabetic patients. Key assumptions for conducting power analysis include expected differences, standard deviations, alpha levels, and desired power levels.

Uploaded by

troyautin
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
You are on page 1/ 10

POWER ANALYSIS

Dr. Elder
TYPE I AND TYPE II ERRORS

Reality
H0 is True H0 is False
Decision Made

Type I Error Correct


Reject H0
(α) Decision

Do Not Correct Type II Error


Reject H0 Decision (β)

1
POWER

Reality
H0 is True H0 is False
Power is the probability of
Decision Made

Type I Error detecting a significant


Reject H0 POWER
(α) difference/relationship when one
actually exists in the population
Do Not Correct Type II Error
Reject H0 Decision (β)

2
MINIMIZING ERRORS

• Type I error is minimized a priori (before the experiment). The researcher sets the
probability they are willing to tolerate.
ü Convention is 5% (or .05)

• Type II error is minimized by calculating the power of an experiment needed to minimize


beta.
ü Power is 1-β; convention is 80% (or .80) which gives Type II error of .20
ü Sample size, effect size, alpha, and test type all affect power

3
G*POWER

• G*Power is a free software that can be used to calculate information related


to power.

• Examples:
ü Observed power in a study (this would be a post hoc power analysis)
ü Sample size needed to obtain a certain power (this would be a priori power
analysis)

4
EXAMPLE: POWER ANALYSIS FOR
INDEPENDENT SAMPLES T-TEST

A clinical dietician wants to compare two different diets, A and B, for diabetic
patients. She hypothesizes that diet A (Group 1) will be better than diet B (Group 2),
in terms of lower blood glucose. She plans to get a random sample of diabetic patients
and randomly assign them to one of the two diets. At the end of the experiment,
which lasts 6 weeks, a fasting blood glucose test will be conducted on each
patient. She also expects that the average difference in blood glucose
measure between the two groups will be about 10 mg/dl. Furthermore, she
also assumes the standard deviation of blood glucose distribution for diet A to be
15 and the standard deviation for diet B to be 17 (this comes from the literature
of similar studies, we’re “best guessing” here). The dietician wants to know the
number of subjects needed in each group assuming equal sized groups.
5
ASSUMPTIONS

• Notice the assumptions that the dietician has made in order to perform the power
analysis. Here is the information we need to know or have to assume in order to
perform the power analysis:
ü The expected difference in the average blood glucose; in this case, it is set to 10.
ü The standard deviations of blood glucose for Group 1 and Group 2; in this case, they are set to 15 and
17, respectively.
ü The alpha level, or the Type I error rate, which is the probability of rejecting the null hypothesis when it
is actually true. Common practice is to set it at the .05 level.
ü The pre-specified level of statistical power for calculating the sample size; this will be set to .80.
• Notice that the dietician didn’t specify the mean for each group, instead she only
specified the difference of the two means. This is because that she is only interested in
the difference, and it does not matter what the means are as long as the difference is
the same. 6
3) Fill in information and
click calculate and transfer
to main window.

2) Click

1) Select how many tails the test is. Fill in alpha,


desired power, and group sample size ratio. Equal 4)Then click the main
sized sample groups are assumed, meaning the calculate
allocation ratio of N1 to N2 is 1. 7
We can see that in order to achieve 80%
power we will need two groups of at least
N = 33!

8
USEFUL RESOURCES

• Examples of tests from this course (the previous example came from this site):
https://stats.oarc.ucla.edu/other/gpower/

• The G*Power manual gives specifics for (almost) all the different tests available
in G*Power:
https://www.psychologie.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-
Naturwissenschaftliche_Fakultaet/Psychologie/AAP/gpower/GPowerManual.pdf

You might also like