It's interesting that these rather straight-forward terms and definitions are so frequently misunderstood. Let's put this to the test. Imagine that we want to assess how much our customers like the customer service we provide. We have a short questionnaire with 5 questions that they can rate 1 (poor) to 5 (very good). We have two samples where customers rated their customer service, each completed by 10 different customers but across similar times of day and staff. Each sample has a mean of 3.5, however one has a low standard deviation and won has a high one. Which group is more pleased? It's interesting that these rather straight-forward terms and definitions are so frequently misunderstood. Let's put this to the test. Imagine that we want to assess how much our customers like the customer service we provide. We have a short questionnaire with 5 questions that they can rate 1 (poor) to 5 (very good). We have two samples where customers rated their customer service, each completed by 10 different customers but across similar times of day and staff. Each sample has a mean of 3.5, however one has a low standard deviation and won has a high one. Which group is more pleased?