Sample size calculation
Even for the most experienced researchers determine the sample size to investigate can be difficult. For this reason, we have prepared this web application that allows you to calculate the sample size and the relative confidence interval to be able to have a representative sample.
Here you can calculate the sample size or the confidence interval obtained
To obtain a representative sample of the target it is necessary to define a sample structure (through the stratification variables) and the sample size.
Defining the sample structure and the sample size is a very complex operation that depends on numerous factors.
For example, under some very restrictive conditions, it is possible to calculate the sample size with a simple program.
This program is used to calculate:
- How many interviews to carry out in order to have a representative sample (under some constraints).
- The level of precision achieved with your sample.
To use this program you must have clear two concepts:
- Confidence interval
- Confidence level
Example: in your sample the 38% answered a certain question "Yes". If you have an 5 confidence interval (ie of 5%) it will mean that the percentage of people who would answer "Yes", in the event that you interviewed the whole population, would be between 33% (38% -5 %) and 43% (38% + 5%).
It represents how much you want to be sure of the result.
Example: continuing with the previous example, putting the level of confidence at 95% will mean that with a 95% probability the percentage of people who would answer "Yes", in the event that you interviewed the whole population, would be between 33% (38 % -5%) and 43% (38% + 5%).
The number of a sample is calculated assuming a certain confidence interval (typically 5%) and a certain level of confidence (typically 95%).
In order to calculate the sample size, you have to establish these two values first.