Ricerche di mercato e numerosità campionaria: due procedure a confronto

Sample size reduction in marketing research is one of the most important aims of researchers and experts. The smaller the sample, the lower the costs and the lesser the realization time of surveys. The purpose of this study, in the context of simple random sampling, is to compare two different procedures to determine the sample size. The first procedure is based on a previous empirical or theoretical determination of estimators and involves a higher risk of error. The second one is not based on this risk assumption because it requires prior population knowledge. The simulation carried out using empirical data shows that the first is more efficient in practical applications when population sizes are restricted. Furthermore, both procedures are adequate in relation to the accuracy level assumed for the estimators.