Question

6)
you are the owner of a restaurant. you need to find how well your cooks doing. you decide to sample their food. You have five cooks on a regular schedule and are open from 8am to 10pm. You decide you will got there twice a week for the next 5 weeks to sample 10 meals as this will be enough. You teel the staff you will be there at 6:30 PM on Tuesdays and 10AM on Fridays.
A) tell me two things wrong (bias), statistically speaking, with this sampling procedure. TEN MEALS IS ENOUGH.
B) Describe a "better" way to sample (remember what you just told me was bad)

Answer 1

a)

In statistics, sampling bias is a bias in which a sampleis collected in such a way that some members of the intended population are less likely to be included than others.

• Undercoverage. Undercoverage occurs when some members of the population are inadequately represented in the sample. As when he reaches at that time, some chefs might not be cooking their dishes. Also, all 10 meals might not include all chefs' dishes.

b) Stratfied sampling. We separate each chef's dishes in different strata