Configuring Sampling Parameters
Sprig offers the ability to customize the response options for the Web and Mobile platforms.
Sprig, by default, will determine the total number of responses that need to be collected in order to meet statistical significance (yes we do all the math, so you don't have to!).
When adjusting response sampling, there are two things you have to consider:
- Response Options
- Results Confidence (Adjust Confidence)
There are two response options to choose from, Until a specified number or Ongoing.
Until a specified number
Sprig will collect a fixed number of responses, after which the study will be set to complete.
Sprig will deliver a fixed number of responses per day and will continue doing so until paused (or marked as completed).
Results Confidence (Adjust Confidence)
By default, we recommend that all studies run at the 95% confidence interval (High confidence, slower results). However, you do have the option to adjust to accommodate your own user insights needs.
- Lower confidence, fastest results (80% confidence interval)
- Medium confidence, faster results (90% confidence interval)
- High confidence, slower results (95% confidence interval) - Default
- Highest confidence, slowest results (98% confidence interval)
How does Sprig calculate these numbers? We use the sample size calculation for a binary outcome. This calculation is applicable to all of the question types offered by Sprig. Click here to read more.
Custom Response Sampling
Sprig incorporates best in class practices to limit sampling bias. How does this sampling work? We use what's known as the leaky bucket.
What you should know is that under standard guidelines, not everyone that qualifies for a study will receive one. Instead, we throttle sends to ensure a study runs over a period of time (days or in some cases a week, depending on response rate)
How to Override Sampling in Sprig
Want to have more control over your study delivery? By setting a Sprig study to Deliver Continuously AND selecting Custom from the dropdown you can effectively nullify our pre-sets for sampling
Please enter a value between 1 - 100% for custom sampling. Setting this value to 100% will ensure that all qualifying users will receive a study.
Another option is to tell Sprig exactly how many responses to collect before the study should be set to complete
Updated 3 months ago