Sampling, Group Designs, and Survey Methods in Criminology

Population

All possible units in a set being studied.

Census

Systematic process of collecting data from all members of a given population.

Sampling

Process of selecting a subset of individuals from a larger population to represent the whole group in a study.

Sample

Subset of the population under study through research.

Population parameters

Statistics for the whole population.

External validity

How well the results of a study generalize beyond the specific conditions and participants involved in the study.

Proximal similarity model

Researchers or consumers of research consider generalizability by matching contexts that are more similar to and less similar to the context of the study.

Sampling model

Model of sampling that proposes that utilizing proper sampling techniques or procedures will yield a representative sample of the population.

Theoretical population

Includes everyone to whom the researcher assumes the findings of a study could generalize.

Study population

Contains everyone from the theoretical population whom the researcher can feasibly access.

Sampling frame

List of individuals or units who are part of the study population.

Respondents

Individuals in the sample who respond to a survey or some other form of research that requires a response.

Selection bias

Occurs when the sampling procedure yields a sample that is not representative of the population that it purports to represent.

Response rate

Percentage of those who received a survey who actually completed the survey.

Nonresponse bias

Concern that those in the sample who do not respond to a survey are different than those who do reply to the survey.

Probability Sampling

Techniques that utilize random sampling.

Random sampling

Involves some component of randomness to select individuals from the sampling frame to be part of the sample.

Simple random sampling

A method in which each unit in the sampling frame has an equal chance of being selected as part of the sample.

Probability Sampling as a Gold Standard

Preferable over nonprobability sampling.

Nonprobability Sampling

Sampling procedures that do not involve random selection of elements from a frame.

Snowball sampling

Study participants or respondents are asked by the researcher to help in recruiting others to participate in the same study.

Convenience sampling

Involves accessing research participants who are easy to access with minimal logistical preparation.

Quota sampling

Type of nonprobability sampling that involves limiting the number of people from given groups who are able to participate in the research study.

Unit of analysis

Elements that the researcher is studying.

Wrongful Conviction Cases

Theoretical population includes all cases in which someone was wrongfully convicted on the basis of a mistaken eyewitness.

Study population (wrongful conviction)

All cases of wrongful conviction listed in the National Registry of Exonerations that include a mistaken eyewitness as a cause.

Sampling frame (wrongful conviction)

The specific cases, likely named by the name of the person who was wrongfully convicted.

Sample (wrongful conviction)

Should overlap completely with the sampling frame, but there might be caveats.

Group design

Design that is primarily intended to make comparisons between or within groups

Independent variable in a group

Categorical variable where groups can be formed

Dependent variable in a group

A variable that is measured by the researcher

Treatment or intervention

Changed in a research study or that we think might influence the dependent variable

Pretest

Measure of the dependent variable prior to exposure to the independent variable

Posttest

Measure of the dependent variable after exposure to the independent variable

One-Group Pretest, Posttest Design

A group design with a single group that receives a treatment or intervention; the value of the dependent variable is measured both before and after that treatment and intervention

Design formula

Pretest (measurement) → treatment → posttest (measurement)

Covariability

Can be demonstrated when there are changes from the pretest to posttest; allows researcher to demonstrate temporal precedence

Two-Group Posttest Only Design

Group design that compares the posttest values of the dependent variable across 2 different groups

Comparison group

A group that we compare to the treatment group; usually a control group

Control group

A group that receives no intervention

Random assignment

A process used in experimental research to place participants into different groups by chance

Experimental design

A type of group design where the researcher manipulates the different attributes (levels) of an independent variable and measures the dependent variable

Quasi-experimental design

An experiment that has a comparison group, but it lacks random assignment

Probabilistic equivalence

Assumption that groups are equal at the start of a study

Experimental group

The group in the design that receives the treatment or intervention

Two-group experimental design

An experimental design with a single independent variable that has two conditions

Conditions

Attributes of the independent variable in an experimental design

One-way factorial experimental design

An experimental design in which there are 3 or more conditions of a single independent variable

Two-way factorial experimental design

An experimental design that utilizes two independent variables

Main effects

Isolated effect of a single independent variable on a dependent variable

Interaction

Occurs when the effect of one variable changes with a condition of another variable

Nonequivalent groups design

A design where a treatment group is compared with a comparison group, but there is no random assignment to groups

Laboratory experiment

An experiment that takes place in a sterile, laboratory setting

Field experiment

An experiment that takes place in the field (real world) rather than in a laboratory setting

One-Group Pretest, Posttest

A design where only the treatment group is tested before and after treatment.

Treatment used

The intervention applied in a study.

Pretest allows observation of change

A pretest enables researchers to see how much change occurs after treatment.

No comparison group

Absence of a group that does not receive the treatment for comparison.

Introduces third variable

A factor that may influence the results but is not controlled for.

Pretest itself could influence posttest

The initial test may affect the outcomes of the subsequent test.

Two-Group Posttest Only

A design with both treatment and comparison groups tested only after treatment.

Depends

Indicates variability based on context or conditions.

Two-Group Pretest, Posttest

A design with both treatment and comparison groups tested before and after treatment.

Probabilistic equivalence is an assumption

The assumption that groups are similar at the start of the experiment.

Highest level of internal validity

The strongest assurance that the results are due to the treatment.

One-Way Factorial Design

A design with multiple treatment groups and potentially a control group.

More than 2 groups allows for further comparisons

Having multiple groups provides additional information for analysis.

Two-Way Factorial Design

A design with multiple independent variables allowing for complex interactions.

Interactions can demonstrate complexity in relationships

Interactions between variables can reveal intricate connections.

Non-Equivalent Groups Design (Quasi-Experiment)

A design with treatment and comparison groups that are not randomly assigned.

Survey

A means of collecting data by asking people questions on topics of interest.

Measure

The survey instrument and the design of its usage.

National Crime Victimization Survey

A survey estimating rates of crime victimization in the US.

Questionnaires

A set of written questions used to gather information from individuals.

Items

Structured prompts designed to gather specific information.

Interview

A data collection method where questions are asked directly to a respondent.

Focus groups

Interviews with a small group discussing a particular topic guided by a moderator.

Need for Clarification

The degree to which respondents may need assistance while answering questions.

Speed of Data Collection

The rate at which data can be gathered using different methods.

Cost of Data Collection

The financial implications of using different data collection methods.

Research Approach (Nomothetic or Idiographic)

Nomothetic focuses on general trends, while idiographic emphasizes individual narratives.

Types of Questions

Various formats including yes/no, multiple-choice, and open-ended questions.

Ease of Data Analysis

The simplicity of analyzing data based on question types.

Social Desirability

The tendency for respondents to answer in a socially acceptable manner.

Interviewer Effects

Changes in responses caused by the interviewer's characteristics or behavior.

When to Use a Questionnaire

Ideal for collecting data from a large sample quickly and anonymously.

When to Use an Interview

Best for exploring complex issues or gathering rich descriptions.

Structured Survey Items

Items that offer predefined answer choices for uniform responses.

Open-ended survey items

Questions allowing free responses to gather qualitative data.

Unstructured Survey Items

Open-ended items that provide rich and personalized data.

Data Analysis

The process of interpreting data collected from surveys.

Dichotomous survey item

A question with only two possible choices.

Filter Questions

A question asked to determine whether a respondent is appropriate or eligible for the survey.

False respondents

Individuals who lie on the filter question and get past it when their data should not be included in the study.

Contingency Question

Question intended to determine whether someone should answer certain questions or if those questions do not apply.

Double-barreled question

Survey item that asks about two different topics in a single question without a means of addressing each individual part.

Telescoping

A memory effort in which someone believes something happened more recently or less recently than it happened.

Leading question

Question on a survey that tries to influence the respondent to answer in a particular way by suggesting there is a preferred or expected answer.

Loaded question

A question on a survey that contains an assumption about the respondent that forces the respondent into answering in a way that might not be fair or neutral.

Social desirability bias

Answering based on what people think is the right answer.