Back to Tutorials HomepageMarketing Research - Tutorial 4
Overview
Sampling - Random sampling; sampling distribution
Data Entry & Coding
First 1/3 of class (If not covered in previous Tutorial) - Go to the menu under random sample generator. Also read and discuss Ch. 8.2 (pages 104-107) in the exercise book). The purpose of this exercise is to learn how to pick a random sample from a population. Let us say you want to pick a sample of 100 students from the student directory of the U. The directory has 244 pages of student listings (pages 105-348). There 4 columns in each page and about 38 entries per column making a total of 152 listings per page. If you want to pick 100 students randomly, you have to pick a page number randomly and then pick a person in that page randomly. So you need to generate two sets of random numbers. The first set corresponds to page numbers ranging from 105-348 and you need a total of 100 random numbers (i.e., because you want a total of 100 students in you sample). The second set of random numbers would range from 1 to 152 (i.e., numbers within a page). Use the program to generate these numbers.
(Points to ponder - What is a random sample ? Why do we need a random number generator ? Why can't we generate numbers ourselves ? Are we leaving out anyone from our sample ? If we need 100 valid addresses, should we aim for more than 100 ?)
Learning points
Steps in sampling
Non-probability versus probability sampling
Whether each element in the sample has a known chance of being selected or not
Probability sampling
Simple random sampling - Using sampling procedures so that each element has equal
chance of being selected in the sample
Second part of class - Go to Speedway Bus Lines in the menu (Ch. 8.1 in the Exercise book). How large should a sample be for any variable ? That is the purpose of this exercise. There are several parts to this program. Go to Part A and try out several sample sizes. Try sample sizes of 1, 5, 10, 30, 50, and 100. Write down what you observe as sample size increases. Repeat the procedure for Part B. Spend a lot of time on Part B and write down your observations of what happens as sample size increases. Repeat the procedures for the parts c, d, and e.
(Point to ponder: Is there certain sample size above which we get good samples ?)
Learning points
Probability sampling
Simple random sampling - Using sampling procedures so that each element has equal chance of being selected in the sample
Sampling distributions - Distribution of means of samples drawn from a population
The Central Limit Theorem -When sample size is large, sampling distribution is normal
Estimating confidence intervals
Third part of class - Discussion
of data entry/data analyses based on Case 6-2.
Back to Marketing Research Homepage