Wellbeing Survey - Sample Size to identify minimum 2% effect size

[Jimbobalot]

Hi

I am currently trying to determine a suitable sample size to detect AT LEAST a 2% increase in high levels of life satisfaction. The survey is on wellbeing and is made up of 10 scale questions (0-10) and 1 open-ended question.

The current sample size I have determined is around 8571 (80% power and 5% significance). But unsure how to consider that this covers the 'at least' part of the question in regard to margin of error / CIs and also how it should consider 'non-response'?

Any help from anyone on how to ensure an effective sample size would be much appreciated. It will also use stratified randomization to try and ensure it is reflective of the population.

Thanks!

 

[hlsmith]

What instrument are you using?

How is it scored: x out of 10?

How were you defining a 2% increase (e.g., based on average score)?

Are you going to try and survey a group of people at two time point in order to find the increase in satisfaction? If so, are you able to match their two scores?

What is the current assumed satisfaction score that you are trying to increase, so what is the baseline value?

Note: likely given the scoring structure and small value of improvement you would like to measure, many, many, people will be needed, as shown by you initial calculation.

 

[Jimbobalot]

10/11 questions are a seven-point scale and 1 question is open-ended.

I need to find out how many individuals should be sampled so that the measure can discriminate interventions that increase the proportion of people reporting very high life satisfaction by at least 2%. Previously it was 30% so it needs to increase to at least 32%.

I also need to consider non-response rates / incompleteness. The addition of an open ended questions will also influence this.

 

[hlsmith]

In order to get closer to discussing the sample size calculation you need to answer more of the questions I posited.

Thanks.

 

[Jimbobalot]

What do you mean by instrument? The 2% increase is just based on a wish to show the efficacy of the new measure. I don't understand the use of a 'two time point'?

 

[hlsmith]

Instrument: the tool you are using. Is this just something you made in-house?

How do you show a 2% increase unless you administer the same tool twice to the exact same people?

Please provide more details, I cannot fully follow what you are trying to do.