Discussion Response 1Y

 Respond to the discussion post below with YOUR educated opinion in 3 sentences WITH scholarly source backing it up 

In describing confidence intervals on a mean, z and t intervals differ by the population mean and the sample size. If the sample size is small and the population mean is unknown the t-interval is used by making an estimate from the sample (Andale, 2013). As sample size grow the t-distribution can still be used. This is typically occurs when the sample size is 30 or above (Sullivan).

At my company, as we migrate customer to online ordering, we talk about creating a frictionless experience. Meaning taking as much friction out of the transactional process as possible. Basically making it easier to complete the transaction. But how do we measure this. One way is to use the time it takes to complete the transaction. We could set an acceptable time, survey a random group of customers and get the mean time to complete the task and standard deviation and then calculate a t-interval for our level of confidence, whatever we decide. This would tell us that we have an X% (say 95%) confidence that our mean would fall between T1 and T1. If the interval is small enough we may be satisfied. If we wanted more confidence we could increase the sample size(Jarrette, 2011) .