Sampling Methods and Research Techniques in Social Studies

Population

Entire group that someone wants to draw conclusion about

Sample

Process of selecting some units of larger population in a study

Reasons to Sample

Not possible to gather data from everyone & Not necessary to collect data from everyone

Unit of Analysis

Individuals & Groups & Organizations & Social artifacts

Unit of Observation

Exactly who is providing information about unit of analysis

Problems with Sampling

Ecological Fallacy & Exception Fallacy

Ecological Fallacy

Draw conclusion about individuals based on research conducted on larger unit of analysis

Exception Fallacy

Draw conclusion about large groups after learning about exception circumstances with few individuals

Goal of Sampling

Select sample which represents the population as we generalize findings based on sample

Sampling Bias

Sample is not representative of larger population (This can be intentional or unintentional)

Creaming

Intentionally selecting cream of the crop (being bias)

Sampling Frame

Comprehensive list of everyone in population from which we will select sample

Types of Probability Sampling

Simple random sampling & Systematic random sampling & Stratified random sampling & Multistage cluster sampling

Simple Random Sampling

Randomly selecting participants → names out of a hat

Systematic Sampling

Elements in total list are chosen for inclusion in sample

Stratified Sampling

Ensures that appropriate numbers are drawn from homogenous subset of the population

Multistage Cluster

Divide sample into cluster and take random sample from each

Nonprobability Sampling

Sampling plans that violate Equal probability Selection Method

Types of Nonprobability Sampling

Convenience/Accidental Sampling & Purposive Sampling & Snowball sampling

Convenience/Accidental Sampling

Study characteristics of people who are passing the sampling point at a specified time

Purposive Sampling

Selecting sample on basis of your judgment and purpose of the study

Quota Sampling

Selecting proportions of individuals based on characteristics of interest

Snowball Sampling

Researcher increases sample size by getting referrals from research participants

Sampling Error

Difference between sample and population - Want to minimize the error

Survey

Standard instrument in systematic way to take measures from large number of units

Cross-sectional Design

Study with just one wave of data collection

Counting Crime

Asking people about victimization & counters problems of data collected by police

Self Reports

Dominant method for studying the causes of crime

Perceptions & Attitudes

Learning how people feel about crime and CJ policy

Targeted Victim Survey

Used to evaluate policy innovations and program success

Types of Surveys

Structured, Unstructured & Semi-structured

Structured

Close ended questions → Pre-defined choices

Unstructured

Open-ended question → tell, explain, describe

Semi-Structured

Mixture of close ended and open ended

Avoid Issues in Survey Construction

Avoid leading questions, Clear and concise and avoid vague wording

Survey Formatting

Careful of question orders, demographic questions for the end, start interesting but not controversial

Written Survey

Make questions visible, leave room for written responses, organize question with matrix

Interview Schedule

Structure interview may have predetermined questions or topical areas to be discussed

Focus Group Interviews

Guided discussion led by facilitator, useful to learn group dynamics, used to generate hypotheses

Question Rules

Create outline of categories, nested set of topical areas and questions worded can affect response

Reflexivity

Subjectivity and meaning you give to information

Recording Data

Audio, video recording, note taking, transcribe dialogue verbatim, write field notes no later than the morning after

Field Research Two Methods

Direct Observation & Asking Questions

Field Research

Allows researchers to engage in natural settings and processes

Inductive Reasoning

Collect data and develop theories and hypotheses testing another time

Grounded Theory

Develop theory based on gathered data

Ethnography

Immersing yourself in particular community or organization to observe their behaviour and interactions

Reactivity

Research subjects change their behaviour as reaction to being studied

Role of Observers

Observe or Participate

Complete Participant

Participates fully, true identity/purpose is not known to the subjects

Participant as Observer

Participates fully but makes it clear they're doing research, mostly in ethnography

Observers As Participant

Identifies themselves as researchers, interacts with participants in social process

Complete Observer

Observe without participating

Sampling Field Research

Impossible to observe everything, controlled probability sampling used rarely

Recording Observations

Note taking or recording, usually for before and after changes

Structured Observation

Observers mark close ended forms producing numeric measures

Environmental Surveys

Applying observational instruments in systematic ways, recording behaviour

Organization Gaining Access

Get a person known and respected by the agency as people may be hesitant to trust

Subcultures Gaining Access

Known, respected member of the subculture

Impact of Ethnographic on Researchers

Time/Money, Personal Cost, Danger

Low Reliability of Field Research

Often very personal

Validity of Field Research

Intended meaning of things observed captured accurately

Less Obtrusive Methods

Content analysis, Secondary data analysis, Official records analysis, Simulation, Crime mapping

Obtrusive

Subjects are aware of data collection

Unobtrusive

Subjects are not aware of data collection

Modes of Obtrusive

Experiment, Surveys, Interview, Focus Groups, Field observation

Modes of Unobtrusive

Content analysis, Agency records, Secondary data Analysis, Systematic reviews

Content Analysis

Study of messages and meanings of those messages

Stages of Coding

Establish universe of data, Identify units of analysis, Sample the data, Start coding

Manifest Content

Visible, surface content similar to using closed ended survey questions

Latent Content

Underlying meaning

Advantages of Content Analysis

Directly examines communication using text, Allowing both qualitative and quantitative analysis

Secondary Data Analysis

Re-analysis of data gathered for other reasons

Sources of Secondary Data

Databases on internet, library, websites

Primary Data

Raw Data

Secondary Data

Data already analyzed

Simulation

Attempted replication of real life scenarios in laboratory setting

Crime Mapping

Map locations of crimes or other items of interest with intent of finding patterns

Systematic Reviews

Collection and summarization of empirical evidence on specific topic

Steps for Systematic Reviews

Identify research questions, Decide on inclusion criteria for selection of studies, Conduct exhaustive literature, Preliminary Review.

Meta Analysis

Systematic review with statistical analysis of the results of included studies

Advantage of Systematic Review

Account for strength of statistical relationships found in each individual study

Publication Bias

Published studies to have positive findings than unpublished work

Applied Research

Systematic attempts contributing to solutions of practical social problems through research

Program Evaluation

Evaluating progress and impact of interventions for decisions making

Problem Analysis

Designed to help public officials choose alternative future actions

Evidence Based Policy

Actions of justice agencies linked to evidence used for planning and evaluations

Policy Process

Demands, Goals, Outputs & Impacts

Process Evaluation

Determine whether a program was implemented as intended

Impact Evaluation

Determine whether program achieved its intended

Theorys of Change

Explains how activities undertaken by interventions contribute to chain of results, leading to intended or observed impacts

Topics to Evaluate in CJ

Policing, Corrections, Criminal courts, Juvenile Justice

Evaluability Assessment

Pre-evaluation, researches determines whether requisites conditions are present

Logic Model

Establishing variables to be measured and expected relationships between them

Methods for Program Evaluation

Randomized Evaluation, Quasi Experimental, Interviews, Focus groups, Field research, Content analysis & secondary data

Problem Solving

Fundamental tool in problem oriented policing applying problem analysis to policing

Career in Program Evaluation

Canadian evaluation society, Non-profit organizations, special topics course

Resources to Use in Problem

How to do it guides, problem and response, case studies, evidence based policing societies