Scribd
Upload
23 views29 pages

Sample Size Determination

Uploaded by

wan nur adnin wan ghazali
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
23 views29 pages

Sample Size Determination

Uploaded by

wan nur adnin wan ghazali
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/ 29

SAMPLE SIZE

CALCULATION
SITI NOR ISMALINA ISA, PHD
DEPARTMENT OF BASIC SCIENCES

FACULTY OF HEALTH SCIENCES

UITM PUNCAK ALAM


DETERMINING SAMPLE SIZE

• An important step in planning a statistical study.


• Part of designing a high-quality study.
• A study must be of adequate size, relatives to the goal of the study.

“How many samples would I need to produce a result with a


certain level of confidence?”
WHAT HAPPENS IF A STUDY HAVING SMALLER SAMPLE
SIZE THAN IT SHOULD?

• Inadequate power to detect the hypothesized effect.


• Small sample size leads to P value being higher & Confidence
interval wider.
➢ Thereby concluding no difference between groups.
• An under-sized study can be a waste to resources because it is incapable
of producing useful study.
➢“ Absence of evidence is not evidence of absence”
WHAT HAPPENS IF A STUDY HAVING LARGER SAMPLE
SIZE THAN NECESSARY?

• The larger the sample the more accurate is the study but there
must be a balance between sample size and the cost.
• Adequate sample size protected some subjects from
unnecessarily having to endure treatments, especially those that
might have unwanted or grievous outcomes.
• Larger than necessary sample size put people in jeopardy
unnecessarily.
KEY CONSIDERATIONS

• What is the primary outcome?


• What's the scale of measurement for the outcome?
• The variability of the outcome?
• What is the desired level of significance and power?
• What defines a clinically/bilogically relevant effect (effect size)?
This is perhaps the most challenging question.
TYPES OF SAMPLE SIZE DETERMINATION

1. Sample size for estimation


• Sample size to estimate proportion/s in a population
• Sample size to estimate mean/s in a population
2. Sample size for hypothesis testing
• Sample size to compare means between two groups of populations
• Sample size to compare proportions between two groups of
populations
SAMPLE SIZE FOR ESTIMATION
(MEAN AND PROPORTION)
ESTIMATING A MEAN (SINGLE MEAN FORMULA)
• The objective of this kind of sample size calculation is to determine the number of sample
size to obtain a specific confidence interval of a specified width.

• Let say, you planned to do a study to estimate the duration of exercise among students in
Faculty of Health Sciences (FSK)
• The result should be reported as "mean duration of exercise (DEx.) and its 95% CI".

e.g. mean DEx. 16.5 mins/day (95% CI: 15.5, 17.5)

The value of a study can be judged by the width


of Confidence Interval.

Wide CI means .. a poor study.


ESTIMATING A MEAN (SINGLE MEAN FORMULA)

This will determine the width of CI

We call this part as “precision” ∆ .

2 ∆ = Width of CI

Note: You have to know the SD value from


literature or simulation, and the precision (Δ)
from your own need.
To estimate duration of exercise

If we plan for 95% confidence (5% error), so Z = 1.96;


And SD ( ) is estimated as 4.3 (Duration of Exercise)

Impossible to check for


normality assumption

Δ = margin of error between sample mean and population mean


Now, it is the researcher decision to select which sample size will be appropriate for the study.
ESTIMATING A PROPORTION
(SINGLE PROPORTION FORMULA)
• Let say, a study is planned to estimate the prevalence of obesity in
Kampong X.
• The result should be reported as Prevalence (Proportion) of obesity and
its 95% CI".

37% (95% CI: 27%, 47%).

Usually reported
like this!!!
If we plan for 95% confidence (5% error), Z = 1.96, and P (prevalence of obesity)
is estimated as 40%.

Now, it is the
researcher
decision to
select which
sample size will
be appropriate
for the study.
HOW TO DECIDE THE PRECISION?

(1) Generally, smaller precision is better.


(2) However, commonly, researchers are limited with the availability of
resources .
(3) It may depend on previous studies:
- In case of the first study, a relatively wide CI is still considered
valuable.
- Previous studies have reported with a certain width of CIs in their
studies. Somehow, if we want to repeat the study, we should come
out with a better width of CI (added value).
SAMPLE SIZE FOR HYPOTHESIS TESTING
SAMPLE SIZE FOR HYPOTHESIS TESTING

• Sample size calculation for hypothesis testing must consider two types of
errors. The main objective of sample size calculation is to obtain adequate
number of samples when testing a null hypothesis considering specific
type 1 and type 2 errors.
• Consider effect size
THE EFFECT SIZE

• It is the magnitude of the difference to be detected. The effect size is determined by finding
a clinically important difference between groups in the frequency or value of the
phenomenon. A larger sample is required to detect smaller differences.
• Do you want to be able to detect a 5% or a 20% difference in the proportion of people
who are in a bad mood?
• Do you want to be able to detect a 12-point or a 25-point difference in mood score
between the two groups?
• If you opted for the 5% or the 12-point options, you will require a larger sample than if you
had opted for the 20% or 25 point options.
COMPARING TWO GROUP MEANS
(INDEPENDENT GROUPS)

(Snedecor & Cochran, 1989)


COMPARING TWO GROUP MEANS
(DEPENDENT GROUPS)

(Snedecor & Cochran, 1989)


SAMPLE SIZE CALCULATION FOR LINEAR
REGRESSION

Tabachnick & Fidell (2001)

N > 50 + 8m

Where m = number of predictors


CALCULATING SAMPLE SIZE FOR
KNOWN POPULATION
Krejcie R V & Morgan D W (1970)

n = required sample size


ꭓ2 = the table value of chi-square for 1 df at the desired confidence level (3.841)
N = the population size
P = the population proportion (assumed to be 0.50 since this would provide maximum sample size)
∆ = the degree of accuracy expressed as a proportion (0.05)
Table 1 is applicable to any defined
population

NOTE:
There is no need of using the formula
since the table of determining sample size
has all the provisions you require to arrive
at your sample size.
ONLINE SOFTWARE
SOFTWARES (LICENSED SOFTWARE – FREE)

Many more………
WITHDRAWALS, MISSING DATA AND LOSSES TO
FOLLOW UP…..
• Sample size values obtained from formulas or software will generally need
to be inflated to allow for expected dropout or loss to follow-up of study
subjects or other sources of missing data.

• It may be necessary to calculate the number of subjects that need to be


approached in order to achieve the final desired sample size.
DROP-OUT RATE

• Suppose it is expected that 10% would drop out, and we have previously
determine the sample size needed is 52.
• The formula to make up for the losses of samples
REFERENCES

• Daniel W.W. (2006). Biostatistics. A Foundation Analysis in the Health Sciences. 8th Edition.
John Wiley: New York
• Krejcie R.V. & Morgan D. W. (1970). Determining sample size for research activities.
Educational And Psychological Measurement. 30:607-610.
• Snedecor, G.W & Cochran, W.G. (1989). Statistical Methods, 8th Edition. Iowa State
University Press.
• Tabachnick, B.G. & Fidell, L.S. (2001). Using Multivariate Statistics. Fourth Edition. Needham
Heights, MA: Allyn & Bacon.
SAMPLE SIZE DOES MATTER !

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy