I was recently asked the following question in an interview.

I had to identify what is the size of the market which is interested in buying Organic Milk in the City. I was given a survey and had to find out a sample to get the results.

I came up with the profile of people who should be targeted for the survey (People who live in costly localities, have their own house, etc.). Based on the profile I estimated that somewhere around 600,000 lived in the city.

Then I was asked the following questions 1. What should be the number of people who should be surveyed? 2. How can I confirm whether the sample I just surveyed is a representative of the population? 3. What would be the level of significance of the same? And Why?

Can anyone help me out here, please?


  • $\begingroup$ It helps to have some demographic questions on the survey: gender, age, race, income, education, and so on. Then results of those can be matched against census data. Trouble is people hate answering those questions, so you might use questions that give some demographic clues. (Nr of bedrooms in house, airline travel, where shop for groceries, zip code.) More difficult to match in your case because of the specialized target population. $\endgroup$ – BruceET Nov 8 '17 at 16:06
  • $\begingroup$ I would reccomend looking into stratum sampling $\endgroup$ – I'mAnAccountantIKnowAlotOfMath Nov 8 '17 at 16:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.