200-202
Introduction to Sampling in Criminal Justice Research
Importance of Data Collection
Fundamental to criminal justice research
Quality of research is reliant on how data is collected
Critical Decision-Making
Determine which observations will be made
Example: Studying drug users requires decisions on which users to observe
Overview of Sampling
Sampling: process of selecting observations for study
Two primary reasons for sampling:
Practical limitations of data collection (impossible to observe all)
Ability to generalize findings from a smaller sample to a larger population
Purpose of Sampling
Essential for generalizing conclusions beyond the sample:
Example: Studying proportions of drug users within arrests
Probability Sampling
Allows generalization from a sample to a wider population
Example: Surveying a number of high school students about marijuana use to reflect the whole population
Non-Probability Sampling
Alternative methods when probability sampling is unfeasible
Specific advantages and disadvantages in criminal justice contexts
Goal of Sampling
Reduce or understand potential biases in selection
The Logic of Probability Sampling
General Concept of Sampling
Defined as selecting part of a population
Purpose:
To represent a larger group
To generalize findings to an unobserved population
Importance of Statistical Generalizations
Probability sampling gives each member of a population a known chance of selection
Enables researchers to predict how well the sample reflects the larger population
Characteristics of Sampling
Importance of Variations in Sample
Samples must reflect differences in the wider population to provide useful insights
Biases during selection can jeopardize the representativeness of a sample
Example of Selection Bias
An untrained researcher interviewing convenient subjects results in skewed samples
Potential misrepresentation of population demographics
Bias in Sampling
Conscious and Unconscious Biases
Casual selection of participants leads to non-representative samples
Risks in systematic sampling (e.g., every tenth lawyer)
Consciously seeking a balanced sample does not guarantee representativeness
Online Polls and Self-Selection Bias
Certain methods (blogs, email polls) lead to biased samples due to selective participation
Understanding Representativeness
Defining Representativeness
A sample is representative if its collective characteristics closely mirror the actual population
Not all characteristics must be represented in equal proportions
Focus on characteristics relevant to the study's substantive interests
Importance of Precise Sampling Techniques
Technologies and methods exist to better ensure representative samples.