In: Statistics and Probability
Question 2) A marketing research manager intends to survey a simple random sample of Swiss consumers and ask questions about level of beverage consumption. How can she estimate the population standard deviation for this variable before selecting the sample? Why would she need to do that? What would happen if her initial estimate were not reflected in the sample results (i.e., if the estimated standard deviation in the sample turned out to be larger than expected)?
a) Population standard deviation
can be estimated by studying the analysis of similar country like
Switzerland. She needs to keep in mind that the demographic factors
of the selected country are similar to Switzerland. Switzerland
being a tourist destination, she needs to account the tourist
population as certain percentage of the total population. Studying
analysis of neighboring European countries will definitely be the
first and most viable option to proceed further
b) She may need to estimate the
population standard deviation before selecting the sample to ensure
that the samples are selected correctly, i.e., the sample selection
follows the principles of sample being heterogeneous in nature,
stratified selection, and should be a good representative of the
complete population
c) If the initial estimate were not reflected in the sample results (i.e., if the estimated standard deviation in the sample turned out to be larger than expected), then she may go for stratified random sampling. In which the sample selected will be divided into smaller groups known as strata. The strata are formed based on shared attributes or characteristics of the respondents. These stratas are homogenous within and heterogenous amongst each other.