MARK 302 Lecture Notes - Lecture 12: Central Limit Theorem, Sampling Distribution, Standard Error
Document Summary
Available budget: sample size is often limited to the available budget of the survey study. Rule of thumb: a client may insist on a specific sample size, but if it is not adequate or excessive (and unnecessarily expensive) the researcher has a professional responsibility to present arguments for a more appropriate sample size. Number of subgroups to be analyzed: the more subgroups there are, the larger the required total sample size: 100+ in major subgroups, 20-50 in lesser subgroups. Application of traditional statistical methods: there is information required in order to extrapolate sample data to a population, including the standard deviation, the acceptable level of sampling error, the desired level of confidence, etc. Normal distribution: a continuous, symmetrical, bell-shaped distribution around the mean (mean = median = mode) Central limit theorem: the distribution of a large number of sample means or sample proportions will approximate a normal distribution, regardless of the.