I am trying to calculate sample size for my descreptive survey study aiming mesurement of nursing students knowledge and attitude of evedence based practice in nursing.
Plz suggest me prevelence rate and metod of calculating sample size for above said study.
Regards.
The prevalence you want is an estimated prevalence of your outcome on your population, and you are the one who know your population. That said, sometimes you don't have any idea of what the prevalence of your outcome is, and in those cases there is usually 3 ways out.
The first is to conduct a literature review. Most likely something about your outcome has been studied, and prevalence of your outcome on somewhat similar populations might have been published already. Look for that and use those as an estimate of your prevalence.
The second is to conduct a pilot study - interview a small number of people that fit your study criteria, and check what is the prevalence of the outcome in this population. This estimate is probably not very precise and might be inaccurate, but it gives you somewhere to start.
The third is to side with caution. If you play around with any sample size calculator for determining the prevalence in a population, keeping all variables constant but the anticipated prevalence, you can see that the maximum sample size is obtained when the anticipated prevalence is 50%. If you don't have any idea of the prevalence of your outcome on your population, you can set this prevalence to 50%, since this will return to you the maximum sample size required in all prevalence scenarios.
Other than the estimated prevalence, you also need to determine your precision limits ("error margin") and confidence level, which also affect the sample size. This calculators also assume a probabilistic sampling, which is almost never true, but if you have a clustered sampling (e.g., you are going to measure your outcome in nurses in 3 hospitals that deal with different kind of patients), then you need to correct for such clustered design.
The website www.openepi.com conducts calculation of sample sizes for proportions, while still displaying you the formula used in the calculation.
Please try G*Power (http://www.gpower.hhu.de/)
It works on Mac and Windows computers.
You will need to specify certain requirements such as expected effect size and alpha level.
I hope this helps.
The recommendation is based on the fact that choosing a value nearer to 0.5 leads to the largest n, within certain limitations. e.g. if the range of prevalence by previous studies is 30–40 %, then P should be set at 0.4, whereas if the range is 70–80 %, it is better to take 0.7, i.e., the value nearer to 0.5 or 50 %. If you don't know the P or doubt about the value of P, it is best to err towards 50% as it would lead to a larger sample size.
Here you the reference：
Naing L, Winn T, Rusli BN: Practical issues in calculating the sample size for prevalence studies. Arch Orofac Sci 2006;1:9–14.
Arya R, Antonisamy B, Kumar S. Sample size estimation in prevalence studies. Indian J Pediatr. 2012;79:1482–1488.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size