Assignment 3 Notes

Assignment 3 Notes Goal of a sample survey? - Learn something about a large population Population Parameter - Is some summary quantity calculated over the entire population Sample Statistic - Is a summary quantity calculated from the sample data Population mean - μ Population Standard Deviation - σ population proportion - Sample Mean - Y bar Sample Standard Deviation - S Population Proportion - ^P Simple Random Sample of size N - Every possible subset of size N from the population of the same size has the same chance of being chosen for the sample Stratified Random Sample - First the population is subdivided (Stratified) into sub-populations called strata, and then simple random samples are drawn from each stratum. Cluster Sample - First the population is subdivided into sub-populations called clusters. Then a random sample of a few clusters is selected with every element (Person) in the clustered surveyed. Multistage Sample - A multistage sample is any sampling scheme that uses a combination of simple random, stratified and cluster sampling. Systematic Sample - A systematic sample is one that is drawn systematically from some ist of population elements. Census - A census samples every element of the population Define Bias - Any sort of systematic - Intetional or not misrepresentation of the population by the sample Sampling Error - sample to sample variation caused by the random component of a sampling scheme. Bias from selecting or voluntary samples - open invitations, internet, radio, publications Bias from convenience samples - At-hand subjects. Customers entering a store, patients admitted to the hospital. non response bias - Sample member who cant or wont respond Response bias - Poorly worded questions or poor interview technique can dramatically bias the response of sample members. Under Coverage BIas - A bias is potentially introduces in the sampling frame does not include all members of the target population. What is an observational Study ? - in which the researcher simply records information of existing groups and does not assign the subjects to the treatment groups. What is a experiment - A experiment is a study in which the researcher randomly assigns the subjects to the treatment groups Placebo - Inactive treatment ( sugar pill ) Control - Aspects of the experiment that we know may have an effect on the response variables. Randomize - The assignment of the subjects (experimental units) to the treatment groups to even out effects that we cannot control Replicate - Over as many subjects as possible Block - the subjects into similar subgroups before assigning the treatments to reduce variation in the response variables. Matched pair design - special type of blocked design where blocks of matched pairs are first set up and then the treatment levels are randomly assigned within each matched pair.