Surveys How to create one Surveys Must Be
Surveys How to create one
Surveys Must Be Random What does it mean to be random? 1. Outcome can not be guessed 2. Must be fair 3. Underlying outcomes are equal as likely û
Pick a Number at Random 1 2 3 4 û Which number did you pick? û On average 70% pick 3, therefore is this truly a random sample? û
Random intervals on the calculator Go to 5: randint(0, 6) - this allows the calculator to randomly choose a number between 0 and 6 û Randint(0, 6, 3)- produces three random integers together between the numbers of 0 to 6. û
Population vs. Sample Keep in mind that a population is hard to survey û Therefore most surveys are completed with samples û The larger the sample the more realistic the answer is û
Bias Keep in mind that a sample is more bias than a population û Must ensure you have a widespread sample. û
Types of Sampling Simple Random Sample û Stratified Random Sampling û Cluster Sampling û Systematic Sampling û
Simple Random Sample SRS û The most common type of sample utilized for a survey û Each combination of people have an equal chance of being selected û
Stratified Random Sampling Population is first divided up into homogeneous groups before being selected û These groups are called strata û The stratas are selected through simple random sampling û
Cluster Sampling Dividing the population into multiple smaller clusters û Select one or two clusters at random and perform survey û Clusters are heterogeneous and strata are homogeneous û
Systematic Sampling When the order will not create a bias answer then systematic sampling may work û Incorporate a system to your selection (every 10 th student) û
- Slides: 11