Download presentation
Presentation is loading. Please wait.
1
Chapter 12 Sample Surveys
Copyright © 2007 Pearson Education, Inc. Publishing as Pearson Addison-Wesley
2
Background Population: An entire group of individuals
Sample: A selected group of the population There are 3 ideas that will allow us to stretch beyond the data we have.
3
Idea 1: Examine a Part of the Whole
Draw a sample. Examining an entire population is almost always impossible A sample survey is a study that asks questions of a sample drawn from some population Samples should be representative of the population
4
Bias Samples that don’t represent every individual in the population fairly are said to be biased. Bias must be avoided There is usually no way to fix a biased sample and no way to obtain any useful information
5
Idea 2: Randomize Randomization takes away bias
Allows us to make inference about the population
6
Idea 3: Sample Size It’s the size of the sample(not the population) that makes the difference. Exception: Sampling more than 10% of a very small population The sample doesn’t need to be a certain percentage of the population, it just needs to be representative,
7
What about a census? Census: Including everyone in the population
Census problems: Pretty much impossible to do this – hard to locate homeless people, babies are being born constantly.
8
Populations and Parameters
Parameter: A numeric value used in a model. A parameter that is part of a model for a population is called a population parameter. Any summary found from the data is a statistic. The statistics that estimate population parameters are called sample statistics.
9
Sampling Methods Voluntary response Convenience Simple Random
Stratified Cluster Systematic Multistage
10
Voluntary Response Sampling
Examples: Newspaper ad asking to answer a survey Mail survey Online survey This type of sampling with almost always include bias. Only those with strong opinions about the topic will care to answer the survey.
11
Convenience Sampling Survey the people most convenient for you.
People on your bus People at your lunch table People in your class This type of sampling does not always give a representative sample.
12
Simple Random Sampling (SRS)
Most common method Assign each individual a number and then randomly select. The sampling frame is a list of individuals from which the sample is drawn. Sampling variability: Each SRS will be different. This is usually the best method. Samples will be unbiased and representative.
13
Stratified Sampling The population is first sliced into homogeneous groups, called strata Then SRS is used within each strata. Can also be called stratified random sampling.
14
Cluster Sampling Split the population into similar clusters.
Select one or more clusters at random and perform a census within each.
15
Systematic Samples Sample every nth person, starting with the kth person. For example, you might survey every 10th person on an alphabetical list of students starting with the 3rd person. To make it random, you must start from a randomly selected individual.
16
Multistage Sampling Combining several methods are called multistage samples. Most surveys use some combination of methods.
17
Types of Bias Voluntary response bias: Individuals choose if they want to participate Not representative. Only people with strong opinions respond. Nonresponse bias: Individuals choose not to respond Don’t want to respond Schedules don’t allow them to respond
18
Response bias: Anything in the survey design that influence the respondents.
Question wording Interviewer appearance Undercoverage Part of the population is not fairly represented There is no way to recover from a biased sample!
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.