Class Notes (838,049)
Criminology (722)
CRM 204 (16)
Lecture 3

# CRM204: Week 3 (Sept. 19th).docx

4 Pages
102 Views

School
Department
Criminology
Course
CRM 204
Professor
Smita Tyagi
Semester
Fall

Description
CRM 204 – Introduction to Criminal Justice Research and Statistics (September 19, 2012) Research process 1. Identify stakeholders 2. Introduce claims, concerns and issues: Ask for reactions 3. Focus further information on above relating to areas of disagreement 4. Negotiate with stakeholders about information collected 5. Attempt to gain consensus Experimental designs The Language of Experimental Designs - Independent and dependent variables (length of prison sentence -> re-offending rates) - Confounding variables --> (See class hand-out; Variables in hypothesis testing) - Hypothesis testing - Experimental and Control groups - Pre-tests and post tests - Random assignment Hypothesis and Causality - At least 2 variables - Expresses a cause-effect relationship between variables - Can be expressed as a prediction - Is logically linked to a research question and theory - Is falsifiable; capable of being tested against empirical evidence and shown to be true or false Null Hypothesis - A null hypothesis is a hypothesis that states that there is no relationship between the two measured variables - You accept or reject the null hypothesis based on your data. If data show a statistically significant change you reject the null hypothesis. - The alternative hypothesis is one that states that there is a relationship between the two variables - Research question - Definition of variables - Study design - Experimental and comparison group - Pretest and post-test group - Random assignment / Matching - Data analysis Experimental and quasi Non-equivalent control group design - Uses experimental and treatment groups - Groups are not created by random assignment - Comparison group selected to be as comparable as possible to treatment group - Individual matching or aggregate matching - One shot case study design - One group - One treatment - One post-test -No random assignment Easy to carry out but difficult to say for sure if treatment caused the dependent variable to change Example: One group pre and post-test design (Before and after design; Fixed sample panel design) -One group -A pre-test -A post-test -No comparison group -No random assignment Ex post facto control group design -Control and comparison groups are designated after (ex post facto) treatment is administered Equivalent time series -One group -One treatment, pre-test then treatment and post-test, then treatment and post-test (Interrupted time series: One group; multiple measures pre and post treatment) (used for rehabilitation programs, sample is small) *beware of research design; no withholding treatment as it is not ethical. Latin Square design -Research question focuses on how several treatments given in different sequences in time affect the dependent variable A B C B C A C A B This is 3 x 3 Latin Square. This pattern can be extended to any size. Question: “What is the most effective sequence for the maximum effect?” Factorial designs -More than one group -Effects of more than one independent variable -Every combination of categories in variables (factors) is examined -As each variable has more than one factor
More Less

Related notes for CRM 204
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.