edwards CONCEPTUAL DRAFT OF CHAPTER 1 INSTRUCTIONS

Maxfield_RM_9e_PPT_Ch_08.pdf

Home >Education homework help >edwards CONCEPTUAL DRAFT OF CHAPTER 1 INSTRUCTIONS

Research Methods for Criminal Justice and Criminology, 9e Chapter 8: Sampling

Chapter Objectives (1 of 2) By the end of this chapter, you should be able to

1. Describe how probability sampling makes it possible to represent large populations with small subsets of those populations.

2. Explain why the chief criterion of a sample’s quality is the degree to which it represents the population from which it was selected.

3. Explain the chief principle of probability sampling: Every member of the population has a known, nonzero probability of being selected into the sample.

4. Describe how probability sampling methods make it possible to select samples that will be representative.

5. Explain how our ability to estimate population parameters with sample statistics is rooted in the sampling distribution and probability theory.

Chapter Objectives (2 of 2) 6. Explain how simple random sampling is logically the most fundamental

technique in probability sampling.

7. Describe the variety of probability sampling designs that can be used and combined to suit different populations and research purposes: systematic sampling, stratified sampling, and multistage cluster sampling.

8. Explain the basic features of the National Crime Survey and the Crime Survey for England and Wales, two national crime surveys based on multistage cluster samples.

9. Describe how nonprobability sampling methods are less statistically representative than probability sampling methods and offer appropriate examples for nonprobability sampling applications.

10. Describe the variety of nonprobability sampling types, including purposive sampling, quota sampling, and snowball sampling. Provide examples of each.

Sampling and Election Polls (1 of 2) • Preelection polls are often used in an attempt to predict final results.

• After failing to accurately predict the 2016 presidential election, the American Association for Public Opinion Research (AAPOR) assembled a committee to study the failure of preelection polling.

• They found little evidence of systematic bias.

• Inaccuracy of support for Trump was due to the impact of late- deciding voters and overrepresentation of college-educated participants (who were not typically Trump supporters) in polling.

• In 2020, pollsters accurately predicted the results, but overestimated Democratic candidate support.

Sampling and Election Polls (2 of 2)

• Lessons from the 2016 and 2020 polling failures

• Nonresponsive people who refuse to participate in research can affect the accuracy of the findings.

• The recruitment method affects the accuracy of research findings, even for probability samples.

Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 66 Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

Introduction

• How we collect representative data is fundamental to criminal justice research.

• Sampling is the process of selecting observations.

• It is not typically possible to collect information from all persons or other units you wish to study, nor is it necessary to collect data from everyone out there.

• Although probability sampling is central to criminal justice research, it cannot be used in many situations of interest.

• In these cases we use nonprobablity sampling.

Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 88 Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

The Logic of Probability Sampling

• In sampling

• We select samples to represent a larger population of people or other things.

• We may want to generalize from a sample to an unobserved populations the sample it is intended to represent.

• In probability sampling, each member of the population has a known and equal chance of being selected into the sample.

• Since we are not completely homogeneous, our sample must reflect— and be representative of—the variations that exist in the population.

Conscious and Unconscious Sampling Bias

• Bias is likely when sample is selected casually.

• Researchers must be conscious of sampling bias.

• This may occur when sample is not fully representative of the larger population from which it was selected.

• Possibilities for inadvertent sampling bias are endless and not always obvious.

Representativeness and Probability of Selection

• A sample is representative if the aggregate characteristics of the sample closely approximate the same aggregate characteristics in the population. This is limited to characteristics relevant to the study.

• This is called the Equal Probability of Selection Method (EPSEM).

• Advantages of probability sampling

• Probability samples are typically more representative than other types of samples.

• Probability samples permit us to estimate the accuracy or representativeness of the sample.

Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 1212 Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

Probability Theory and Sampling Distribution

Probability Theory and Sampling Distribution (1 of 2)

• Probability theory permits inferences about how sampled data are distributed around the value found in a larger population.

• We must understand four concepts:

• Sample element: who or what we are studying

• Population: theoretically specified grouping of study elements

• Population parameter: the value for a given variable in a population

• Sample statistic: the summary description of a given variable in the sample; we use sample statistics to make estimates or inferences of population parameters

Probability Theory and Sampling Distribution (2 of 2)

• The purpose of sampling is to select a set of elements from a population in such a way that descriptions of those elements (sample statistics) accurately portray the parameters of the total population from which the elements are selected.

• The key to this process is random selection.

• Serves as a check on conscious or unconscious bias

• Can draw on probability theory to estimate population parameters and how accurate our statistics are likely to be

The Sampling Distribution of 10 Cases

• A sampling distribution is the range of sample statistics we will obtain if we select many samples.

• Sampling distribution example: Figures 8.3–8.6

From Sampling Distribution to Parameter Estimate

• Example: studying the population of Placid Coast, California, to assess levels of approval of proposed law

• Target population: all adult residents

• Sampling frame: voter registration list

• A sampling frame is a list of all elements in our population.

• Element: individual voters on Placid Coast

• Variable: attitude toward proposed law; approve and disapprove

• Random sample: 100 people

• Use expanding sampling distribution to find the parameter in the population

Estimating Sampling Error

• Probability theory tells us that if we select many independent random samples from a population, then the sample statistics provided by those samples will be distributed around population parameter in a known way.

• We have a formula for estimating how closely the sample statistics are clustered around the true value: standard error

• This is a measure of sampling error × =S P Q

• Where S is the standard error, P and Q are the population parameters for the binomial and n is the number of cases in each sample.

• The standard error tells us how sample statistics will be dispersed or clustered around a population parameter.

Confidence Levels and Confidence Intervals • Probability theory specifies that 68 percent of large number estimates will fall within

one standard error of the population parameter, and 95 percent will fall within two standard errors.

• This logic permits us to construct a confidence level. • Confidence levels express the accuracy of our sample statistics in terms of a level

of confidence that the statistics fall within a specified interval from the parameter.

• For example, we are 68% confident that our sample estimate falls within one standard error of the parameter.

• The confidence interval is the range of values within which the population parameter is likely to be found.

• The logic of confidence levels and confidence intervals also provides the basis for determining the appropriate sample size for a study.

Discussion Question Activity 1

• How would you respond to someone who told you that they were 100% confident in an interpretation of their survey results?

• What questions would you have for them?

Probability Theory and Sampling Distribution Summed Up

• Random selection permits the researcher to link findings from a sample to the body of probability theory so as to estimate the accuracy of those findings.

• All statements of accuracy in sampling must specify both a confidence level and a confidence interval.

• The researcher must report that they are x percent confident that the population parameter is between two specific values.

Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 2121 Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

Probability Sampling

• Different types of probability sampling designs can be used alone or in combination for different research purposes.

• As consumers of research, we must understand the theoretical foundations of sampling.

• Researchers have a number of options in choosing a sampling method outside of simple random sampling.

Populations and Sampling Frames

• One key feature of all probability sampling designs: the relationship between populations and sampling frames

• Sampling frame: The quasi-list of elements of the target population from which a probability sample is selected.

• Example: We want to study the attitudes of corrections administrators about a new law.

• Sampling frame: membership roster of the American Correctional Association

• Sampling frames serve as a real-world version of an abstract study population.

Simple Random Sampling

• The basis of probability theory and statistical tools we use to estimate population parameters, standard error, and confidence intervals.

• Steps

• Establish a sampling frame.

• Each element in a sampling frame is assigned a number.

• Choices are then made through random number generation (table or computer program) as to which elements will be included in your sample.

Systematic Sampling

• Elements in the total list are chosen (systematically) for inclusion in the sample.

• This is more commonly used than simple random sampling.

• Example

• List of 10,000 elements, we want a sample of 1,000, select every tenth element

• Choose first element randomly.

• Danger: periodicity—A periodic arrangement of elements in the list can make systematic sampling unwise.

Stratified Sampling

• Two methods of selecting a sample from a list: random and systemic

• Stratification represents a modification of their use.

• Stratified sampling: ensures that appropriate numbers are drawn from homogeneous subsets of the population

• Method for obtaining a greater degree of representativeness— decreasing the probable sampling error.

• Choice of stratification variables depends upon what variables are available and important for research questions.

Disproportionate Stratified Sampling

• Disproportionate stratified sampling purposively produces samples that are not representative of a population on some variable.

• It is a way of obtaining a sufficient number of “rare” cases by selecting a number disproportionate to their representation in the sampling frame.

• Example: General Social Survey—Canadians’ Safety (2021)

• Oversamples individuals who identify as First Nation, Inuit, and Métis to provide more accurate victimization estimates for the Indigenous population.

Multistage Cluster Sampling

• May be used when it is either impossible or impractical to compile an exhaustive list of the elements that compose the target population (e.g., all law enforcement officers in the U.S.)

• Involves the repetition of two basic steps

• Listing

• Sampling

• Sampling units are population elements or aggregations of those elements.

Multistage Cluster Sampling with Stratification

• We may use stratification techniques to refine and improve the sample being selected for our multistage cluster.

• Once the primary sampling units have been grouped according to relevant, available stratification variables, either simple random or systematic sampling can be used to select the sample.

• The more homogenous the strata of clusters, the lower the sampling error.

• The primary goal of stratification is homogeneity.

Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 3030 Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

Illustration: Two National Crime Surveys

• Two national crime surveys show different ways of designing samples to achieve desired results.

• The National Crime Victimization Survey (NCVS) conducted by the U.S. Census Bureau

• The Crime Survey for England and Wales (CSEW)

• Both are surveys that use multistage cluster samples, but use different strategies for sampling to produce sufficient numbers of respondents in different categories.

The National Crime Victimization Survey (1 of 2)

• Seeks to represent the nationwide population of persons 12+ living in households

• First stage: define primary sampling units (PSUs)

• Defined as large metropolitan areas, nonmetropolitan counties, or groups of contiguous counties

• All PSUs in large Core-Based Statistical Areas (CBSA) are included as self-representing = 330 self-representing PSUs

• Smaller, non-CBSA PSUs are grouped to provide accurate victimization estimates = 203 non-self-representing PSUs

The National Crime Victimization Survey (2 of 2)

• Second stage: designating two different sampling frames within each of the 542 PSUs

• Housing unit frame

• Group quarters frame

• The 2021 NCVS yielded 238,043 individuals living in 150,138 households.

• The sampling design for the NCVS is scheduled to be updated in 2026.

The Crime Survey for England and Wales

• CSEW aims to produce national estimates of victimization for people 16 years and older

• Simplified by the existence of a national address database

• Sampling process is designed to reflect police force areas:

• The list of addresses within the postcodes are stratified and sampled to conduct about 600 interviews within each police area.

• After household selected, one resident aged 16+ were listed, and one was randomly selected by interviewers.

Discussion Question Activity 2

• If you were administering a national survey to assess crime victimization, which model would you prefer: one like the NCVS or the CSEW?

• Why did you choose that particular model?

• Would you make any changes?

Probability Sampling in Review

• Depending on the field situation, probability sampling can be very simple or extremely complex, time-consuming, and expensive.

• Probability sampling avoids conscious or unconscious bias in element selection.

• It permits estimates of sampling error.

• Despite advantages, it is sometimes impossible to use standard probability sampling methods or may not be appropriate to do so.

Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part. 3737 Michael G. Maxfield, Earl R. Babbie, Amie Schuck , Research Methods for Criminal Justice and Criminology, 9 Edition. © 2025 Cengage Learning, Inc. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.

Nonprobability Sampling

• In many research applications, nonprobability samples are necessary or advantageous.

• Nonprobability sample is sampling in which the probability that any given element will be included in the sample is not known.

• There are four types: • Purposive or judgmental sampling

• Quota sampling

• The reliance on available participants

• Snowball sampling

Purposive or Judgmental Sampling

• Purposive sampling is selecting a sample on the basis of your judgment and the purpose of the study:

• May be used to select study elements that exhibit a particular attribute

• May be used to represent patterns of complex variation

• Can also be used to pretest questionnaires

Quota Sampling • In quota sampling, units are selected so that total sample has the

same distribution of characteristics as are assumed to exist in the population being studied.

• Has two inherent problems: • The quota frame must be accurate, and it is often difficult to get

up-to-date information for this purpose. • Biases may exist in the selection of sample elements within a given

cell, even though its proportion of the population is accurately estimated.

• Quotas and purposive sampling may be combined to produce samples that are intuitively, if not statistically, representative.

Reliance on Available Participants or Other Units

• Misleadingly called “convenience sampling” • Often used due to the ease and low cost of convenience methods in

comparison to others • Best used for polling of sample when a specific time and location is

important to the study • Example: Painter’s (1996) study of street lighting conditions in

London • Convenience sampling may be combined with probability models • Online samples offer a new and creative way to select high-quality

convenience samples.

Snowball Sampling

• In snowball sampling, you interview some individuals, and then ask them to identify others who will participate in the study.

• It closely resembles the available-participants approach.

• Commonly used for field observation studies or qualitative interviewing.

• Helpful in sampling a population that may be difficult to access or distrustful of others

Discussion Question Activity 3

• How would you explain the strengths and weakness of snowball sampling?

• What is a research subject or a participant group that snowball sampling may be appropriate to use?

Nonprobability Sampling in Review

• Snowball samples are variations on purposive samples and samples of available participants.

• May be necessary when the target population is difficult to locate or even identify

• Advantageous when a probability sample would be ineffective, costly, and inefficient

• Sampling plans must be adapted to specific research applications.

Summary

• Click the link to review the objectives for this presentation.

• Link to Objectives

Research Methods for Criminal Justice and Criminology, 9e
Chapter Objectives (1 of 2)
Chapter Objectives (2 of 2)
Sampling and Election Polls (1 of 2)
Sampling and Election Polls (2 of 2)
Introduction
Introduction
The Logic of Probability Sampling
The Logic of Probability Sampling
Conscious and Unconscious Sampling Bias
Representativeness and Probability of Selection
Probability Theory and Sampling Distribution
Probability Theory and Sampling Distribution (1 of 2)
Probability Theory and Sampling Distribution (2 of 2)
The Sampling Distribution of 10 Cases
From Sampling Distribution to Parameter Estimate
Estimating Sampling Error
Confidence Levels and Confidence Intervals
Discussion Question Activity 1
Probability Theory and Sampling Distribution Summed Up
Probability Sampling
Probability Sampling
Populations and Sampling Frames
Simple Random Sampling
Systematic Sampling
Stratified Sampling
Disproportionate Stratified Sampling
Multistage Cluster Sampling
Multistage Cluster Sampling with Stratification
Illustration: Two National Crime Surveys
Illustration: Two National Crime Surveys
The National Crime Victimization Survey (1 of 2)
The National Crime Victimization Survey (2 of 2)
The Crime Survey for England and Wales
Discussion Question Activity 2
Probability Sampling in Review
Nonprobability Sampling
Nonprobability Sampling
Purposive or Judgmental Sampling
Quota Sampling
Reliance on Available Participants or Other Units
Snowball Sampling
Discussion Question Activity 3
Nonprobability Sampling in Review
Summary