Here is some background info and responses to your questions:
What you got from OpenEpi could be sufficient for your study depending on the effect size you are hoping to detect and the variability in the population and other variables. However, it's important to ensure that your sample is representative of the population you are studying and that your study is adequately powered to detect any significant differences or associations you are interested in.
Also, I think the population size that you input to the OpenEpi software is too low. I think it should be the population of healthcare workers you are targeting. This could be the whole population of healthcare workers in the US or in the region you are doing the survey and trying to find the variability in.
The equation used in OpenEpi for sample size calculation is a commonly used formula and can be suitable for many types of studies, including cross-sectional studies. However, it's important to consider whether the assumptions underlying the formula are appropriate for your study and to input correct numbers/information.
There are many online sample size calculators that you can use. But you need to privide more information.
For example, are you sure about the size effect?
The effect size is the magnitude of the difference or association you want to detect. It depends on the research question and the type of statistical test you plan to use. For example, if you want to compare the proportion of healthcare workers who follow a specific feeding practice to the proportion who do not, the effect size could be the difference in proportions (i.e., the absolute difference between the two proportions). Alternatively, you may use a standardized effect size such as Cohen's d.
In general, calculating the sample size requires some information about your study design and the effect size you want to detect. For example:
- What is the level of precision you desire (i.e., what is the margin of error you can tolerate)?
- What is the level of confidence you desire (e.g., 95%, 99%)?
- What is the expected proportion of healthcare workers who will respond to your survey?
- What is the effect size you are interested in detecting (e.g., the difference between two proportions, the odds ratio, etc.)?
- What is the alpha level (i.e., level of significance) you are using for hypothesis testing?
Also, please note that you need to adjust for potential nonresponse too! You may want to adjust your sample size upward to account for potential nonresponse or missing data. A common rule of thumb is to increase the sample size by 10% to 20% to account for potential nonresponse. The dropout rate you mentioned usually comes from previous studies and surveys and is not a very accurate measure.
Your point about the high sample size is a valid point. While having a larger sample size can improve the precision of estimates and increase statistical power, there are practical limitations to consider, such as response rate and availability of participants. In some cases, it may be more appropriate to use a smaller sample size to ensure a higher response rate and adequate representation of the population. Ultimately, the sample size should be determined based on a balance between statistical power, practical limitations, and the research question being investigated.
Very low bounty!
Hi! How much do you think it is worth?
Hi. I can give you general guidance on your questions. Some of them are not very clear. You can tip me after. Sounds good?
Yeah sounds good to me :)