Has anyone calculated the sample size required to make CSAT survey results statistically significant in AWS QuickSight

Question & Answer
, ,
1

Hello everyone,

I’m currently working on a Customer Satisfaction (CSAT) dashboard in AWS QuickSight, and I want to calculate the required sample size to ensure that the survey results are statistically valid. Specifically, I’m looking to calculate the sample size needed for a 95% confidence level with a 0.5% margin of error for a population of around 1,000 active clients.

I understand the standard sample size formula involves:

  • The Z-score (for 95% confidence, this is 1.96),

  • The population proportion (p) (which I’m assuming to be 0.5 for the most conservative estimate),

  • The margin of error (set to 0.5%).

  • ZZZ = 1.96 (for 95% confidence),

  • p=0.5p = 0.5p=0.5,

  • e=0.005e = 0.005e=0.005 (margin of error),

  • N=1,000N = 1,000N=1,000 (active clients).

Has anyone successfully implemented such a calculation in AWS QuickSight? Are there any alternative methods or workarounds for calculating sample sizes within QuickSight, or would you recommend calculating it externally and then importing the result?

Any guidance or examples would be greatly appreciated!

2

Hello @Asfak !

If this number is going to remain static, I would recommend doing this outside of Quicksight and then applying that number as the default value of a Parameter that you can apply to calculated field.

3

Thanks for your response. I have managed to calculate this dynamically.

To calculate the sample size dynamically for 90% confidence level and a 10% margin of error.

ActiveClientPopulation= countIf(clientid, {clientstatus_isactive}=1)

InitialSampleSize= (1.645 * 1.645 * 0.5 * (1 - 0.5)) / (0.10 * 0.10)

FinitePopulationCorrection= ceil(max(InitialSampleSize) / (1 + ((max(InitialSampleSize) - 1) / ActiveClientPopulation)))

1 Like