Midterm Review
Quantitative Research
1. Research designs
 Topic, literature review, theory, hypothesis, conceptualization, and operationalization.
 Wallaces wheel: theories hypothesis observations empirical generalizations theories
 Hypothesis: an idea
Criminology 320: Research Proposal
Drug and Alcohol Use in Colleges in the United States
Preventative Measures and Policies to Target Campus Violence
Author:
Rafael Martins
Date:
February 08, 2015
Student Number:
301265592
Organization Affiliation:
Simon
Definition
Measurement the process of assigning a unit of analysis to an attribute or a variable
Unit of analysis the person or thing from which data are collected
Variable a set of logical attributes that are of interest to the researcher
Concrete the
DescriptionStudy No. 2859
TITLE: Clients of Street Prostitutes in Portland, Oregon, San Francisco and Santa Clara,
California, and Las Vegas, Nevada, 19961999
PRINCIPAL INVESTIGATOR(S): Monto, Martin A.
AFFILIATION: University of Portland
SUMMARY
STUDY
from Fox book focus on:
conceptual info
rules for stat tests
steps in analysis
components in the equations
sample size, mean, variance
Exams:
definitions
applied concepts
interpreting and applying SPSS output
Scenario interpretation and statistic
T Tests
Independent Samples t tests
two diff groups (m/f)
Paired samples t tests
measuring same ppl twice
onesample t tests
IQ tests
police chief wants to know actual response time
Testing differences of Means
CJS research wants to analyse diffs
Distributions and Statistical Significance
Objectives
how probability, normal curve rlate to random sample + stats
what can we say about an unobserved pop based on results
observed in a random sample
dfn confidence interval + statistical significance
I
Exam focus
all info from lectures
equations that are examinable
all diagrams
SPSS outputs
Fox et al. text
all definitions and concepts
all conceptual information including steps required in tests
Yockey text/Tutorial:
aids w/ interpretation of SP
Multiple Regression Continued
Use when we have 2/ more independent variables
KEY CONCEPT: ALL ELSE HELD CONSTANT
multiple regression calculates the effect of each independent
variables, while holding all of the other IVs constant at their
means
Functio
Assumptions of Correlation
linear relationship between x and y
cant have a curvilinear relationship
Interval level data for x and y
random sampling
normal distribution of x and y in the population
Correlation
measure degree of association between tw
Regression
Regression as a causal model
explicitly specify IV and DV
Assumptions
straightline relationship
Normal Distribution for IV and DVs or a large sample
Random sample to test significance
errors are independent and normal distribution of your
Measures of Association for Nominal (or Nominal/Ordinal) Data
Themes
what happens when normality cant be assumed
how to measure association for nominal variables
Reminders
variables measured @ higher levels have ALL props of lower
level Variables
can
Organizing and Summarizing Sample Data
SAMPLE distributions=/= SAMPLING distributions
Sample Stats
imprtnce of levls of measurement
affect how data is described
these stats based on sample + describe attributes of the
sample
not interpret beyond samp
sCriminology 320 Lecture 6 T Tests
Todays Themes

Independent samples t tests
Paired samples t tests or dependent samples t tests
Onesample t tests
Testing Differences of Means


Often in CJS research we want to analyse differences on a continuous var
Lecture 10 Crim 320 Correlation
Online readings: read the methods part and stuff with correlations, dont have to read all of it.
Assumptions of Correlation

A linear relationship between X and Y.
: cannot have a curvilinear relationship
Interval level da
Lecture 12 Crim 320 Multiple Regression Continued
Null & Research Hypotheses

There is 1 null for the overall model and one null for the IV
: similar to ANOVA
Regression Output Model Summary
Regression Output Intercept Coefficient

Use the standardized
Student Name:
Student Number:
CRIM 3203
Summer 2013
Online Assignment 5 Clients of Prostitution
Select any reasonable dependent variable and use ANOVA to answer the
following questions:
Exercise A
Is there a relationship between race and the dependent va
Final study guild
t= (observed mean hypothesized mean)/SE of observed mean
Onesample t test is used to compare a sample mean to a hypothesized population mean.
Three assumptions:
1) The test variable must be normally distributed in the population
2) Case
Student Name:
Student Number:
CRIM 3203
Online Assignment 1 Drugs, alcohol and student crime
Exercise A
1.
Run a frequency distribution for the total_use variable, and print
the output.
Frequencies
Statistics
Total Alcohol and Drug Use in Past
Month
N
Va
Student Name:
Student Number:
CRIM 3203
Online Assignment 5 Drugs, alcohol and student crime
Exercise A
Is there a relationship between ethnicity and the dependent variable?
Assumption number 1 The dependent variable (how many times have you used alcohol
Student Name:
Student Number:
CRIM 3203
Online Assignment 3 Drugs, alcohol and student crime
Exercise A
In this exercise, you are going to test determine whether there is a
relationship between a respondents ethnicity and his or her assessment of
whether
Student Name:
Student Number:
CRIM 3203
Online Assignment 2 Drugs, alcohol and student crime
Exercise A
1.
Run frequencies for the three new variables, and print the output for
each.
Frequencies
Statistics
What percentage of students claim
that neither,
Student Name:
Student Number:
CRIM 3203
Online Assignment 4 Drugs, alcohol and student crime
Exercise A
Answer this question: Can we conclude from this sample that students are
more happy than they are unhappy?
Print your output and answer any and all qu
Lecture 4 Crim 320
Parametric tests

have 2 foundational assumptions
: Interval LOM
: Normal distribution in the population

Violation of these assumptions requires the use of alternative tests
NonParametric Tests

Alternatives to parametric tests.
Re
Lecture 3 crim 320 Jan 21
Objectives of Today

Using sample statistics to organize, present & summarize sample distributions
Sample Statistics

Remember importance of levels of measurement.
: They affect how data is describe
These statistics are based p
Start final on lecture 4 Distribution to the end. It would be similar to the midterm but with more applied
and interpretative questions.
The exam is short and long answer.
The majority of the exam is on postmidterm information.
Exam Structure:
1.
2.
3.
4
Review of Methods
know diffs betwn quant and qual (download on canvas)
know assumptions
lom, distributions
stats software does calcs
know how stats tests are rltd/ how theyve evolved based on
limits of other procedures
quant rsrch designs
experimen