Question

Choose a population that you would plan to sample or survey. Your discussion board thread title should be "Sampling from ______". Some ideas for populations to take a sample from: 

Mesa students

San Diego community college students

All San Diego college students

Adults in San Diego

Starbucks customers

Your choice!

Describe an perfect scenario sampling method: Describe in your own words one of the sampling methods learned in class and how it could be applied to your population - in this case, you can assume you will have access to things like a list of everyone living in San Diego.

Describe a realistic sampling method. Being that you don't actually have a list of all San Diego residents (or similar for your population), how would YOU go about trying to get a representative sample? No need to use any fancy terms or definitions here, just describe how you'd collect data from 100 people for your sample.

What are some limitations that will arise with your realistic scenario? Are there groups that might be left out? 

Answer all of this in Approximately 150-200 words in length and well-written.

Answer 1

"Sampling from Starbucks Customer"

One of the realistic sampling method is stratified sampling. In stratified sampling we first divide the population of N elements in H groups (say). These groups are known as strata. From each strata , sometimes called a stratum, we draw a random sample by simple random sampling method. Groups are divided in such a way that data in a group is homogeneous and two groups are heterogeneous. Then we study only the subset obtained by srs (representative sample).

In Starbucks data of size 100, suppose we have to know which is the most popular Coffee. First we will divide our data on the basis of gender, age, visiting hours, occupation(or their working hours). Draw sample of size 10 from each group . We have a subset of 40, consisting all kinds of customers. Data can be in the form of questionnaire.

Limitations

Overlapping will be an issue. It is possible that a customer falls into multiple groups. In that case we have to choose in which group that customer belongs

If we are collecting the data on a particular day (say weekday) there might be the possibility we leave out those customers who visit only on the weekends. This might change our results.