Study Guide Mar4613 Ch 9 10 11 12 14 Chapter 9 Developing the Sample Plan Census Type of sampling plan in which data are collected from or about each member of a population o Attempt a census when the overall population is limited in size even if you can t get information from everyone Sampling plan process of selecting the people or objects to be surveyed interviews or observed Collect information from a sample Goal is to determine what is likely to be true for a population based on data obtained from only a subset of the population Sample Subset of elements from a larger group of objects An easier and cheaper alternative to a census Six steps for drawing a sample 1 Define the target population o Population all cases that meet designated specifications for membership in the group o Stay clear when defining the population people households organizations over the age of 30 etc o The simpler your definition the less costly it will be However don t let cost dictate the answer you re trying to find o Parameter characteristic or measure of a population Example studying FSU students we can describe this population on a number of parameters such as proportion of job holders study time etc o Statistics measures of a sample allowing us to make assumptions about the population o Sampling error the difference in results from the sample and results obtained from the population as a whole 2 Identify the sampling frame o Sampling Frame list of population elements from which a sample will be drawn the list could consist of individuals units institutions Sampling frames can be tough to find or use For example a phonebook could be used as a sampling frame however what about all the people who just use cell phones will you get an adequate sample 3 Select a sampling procedure o Nonprobability Samples sample that relies on personal judgment in the element selection process o Types of Nonprobability Samples Convenience Samples Population elements are included in the sample due to their convenience People or other criteria are used in theses samples because they were in the right place at the right time Going to the mall and asking surveys doesn t cover the part of the population that doesn t visit the mall though Judgment Samples sample elements are handpicked because they are expected to serve research purpose Example Toy company chooses 100 kids to test their toys but not randomly they hold interviews to see what children are the smartest this will not accurately predict the rest of the population Quota Sample chosen so that the proportion of sample elements with certain characteristics is about the same as the proportion of the elements with the characteristics in the target population Example choosing 1000 high school students But dividing it evenly between gender grade and race This may not accurately reflect the population of the entire school o Types of Probability Samples o Probability Sample each member of the target population has a known random chance of being included in the sample Simple Random Sample each unit included in the sample has a known and equal chance of being selected for study and every combination of population elements is a sample possibility Systematic sample plan in which every kth element in the population is selected for the sample pool after a random start Example 250 students need to be interviewed there are 5000 students in the school s0 5000 250 equals 20 so go down the list and ever 20 names pick a name continue till you have 250 names that are random Total Sampling Elements number of population elements units or people that need to be drawn from the population to end up with desired sample size Therefore if we want 250 students to interview we better choose 500 because we can expect half to blow the interview off Stratified Sample The population is divided into mutually exclusive and exhaustive subgroups and samples are chose from each subgroup Note that samples are chosen from each subgroup not one Cluster Samples similar to stratified sample except samples are chosen from a random subgroup Area Sample special from of cluster sample in which geographic areas are chosen as the clusters then randomly selected to use for samples Example a farmer sections of 100 different areas of his farm Plot 99 63 48 50 and 7 were chosen at random to study and will hopefully represent all plots 4 Determine the sampling size Three factors affect the size of the sample needed o Amount of diversity or variation of the parameter in question within the population is beyond your control In other words the larger the variation in the study the larger the sample size should be to increase accuracy The smaller the variation the less the sample size is fine we just never know the variation until we do the numbers o Secondly how precise the estimate must be If the study needs to be very precise the larger the sample size will need to be Example you are studying average diners Do you include diners within 100 of the mean income or 500 within mean income Depends on what you consider average Precision degree of error in an estimate of a population parameter o Last item affecting sample size would be the degree of confidence you would like in your estimate Confidence degree of certainty that the study reflects the true value The larger the sample the more confidence we can have Example average diner income is 40000 We are 85 confident that the populations diners will fall between 38000 and 42000 for their average income 2000 o NOTE The size of the population has no direct effect on the sample size If you need 1 sample to reflect 1000 people than that is ok However the degree of confidence precision ad variation desired is what drives larger samples A few approaches to determining sample size other than the statistical approaches listed above include Marketing research the problem with this is data collection is variable therefore every person you study will cost money When the funds run out that is usually when your sample is done Another approach is to simply use what other similar studies used If you are confident that they were historically accurate just use the same amount of samples 5 Select the sample elements 6 Collect the data form the designated elements Other vocab Snowball Sample Judgment sample that relies on the researcher s ability to locate an initial set of respondents with the desired characteristics Sampling Interval number of population elements to count K when selecting
View Full Document
Unlocking...