Product Tips: Sampling
When you see a chart in your dashboard analysing the responses received from one of your Clicktools surveys (i.e., Net Promoter), how do you know if the views are really reflective of those in your target audience?
The term ‘sampling‘ is used to represent the number of people who must complete your survey in order for you to get a representative view of your target audience.
The term ‘confidence level’ represents how often the true percentage of the population would pick an answer that lies within the confidence interval. A confidence level of 95% means you can be 95% certain; the 99% confidence level means you can be 99% certain etc.
Combining the confidence level and interval in our example, we can be 95% sure that between 85% and 95% of the population would choose answer “A”. Theconfidence interval is the plus-or-minus figure reported alongside survey results. For example, if 90% percent of your sample pick answer “A” and you are using a confidence interval of 5, it is fair to say that if the entire target audience had answered the question, between 85% (90-5) and 95% (90+5) would have picked answer “A” also.
The tables below outline how many completed responses you need to treat the results as reflective of the whole of your population for different combinations of confidence interval, confidence level and target population.
Most research is carried out using 95% confidence level with a 3% or 5% confidence interval (highlighted in green and orange).
Now that you have confidence that your responses are truly reflective, be sure to give all the responses you receive equal billing. If using NPS, use your detractor’s views to make improvements and identify your promoters to spread the word.So, for example, if you have 1,000 customers, you need results from 278 customers to be 95% confident in your results to a level of + or – 5%.