CRCJ 3001 Final: CRCJ 3001 ONECLASS study guide

28 Pages
Unlock Document

Carleton University
Criminology and Criminal Justice
CRCJ 3001
Donna Mailloux

CRCJ3001– QuantitativeMethodsin Criminology (Prof.Donna Mailloux) Introduction Is Ottawa safer today than 5 years ago? In 2015… • 32,500 charges laid (excl. traffic) • 40,000 in 2011 • 20% decrease! Ottawa police service employs 1,272 officers – 25 charges/officer/year • No breakdown of what types of charges were laid • Shift in focus of police efforts (i.e. locations) • How many charges actually led to conviction • Multiple officers on one charge • Reliable = every time you run the same statistics, you achieve the same results • Valid = actually measures what it is intended to measure The Science News Cycle John Oliver video – reporting on scientific studies • Context, methodology • Question process & results • Avoiding confirmation bias • Importance of replicating studies • Peer-review process Frances Rauscher (1993) – studied students who listened to Mozart • “Students who had listened to the Mozart sonata scores sig. higher on the spatial temporal task” – i.e. being able to flip things in 3 dimensions • “Important to note that we did not find effects for general intelligence just for this one aspect of intelligence. It’s a small gain and it doesn’t last very long” • Next… o Headlines  “Mozart makes you smart” o Generalizations to children that listening to Mozart from a young age makes you smart o Marketing  “Baby Einstein” How do we know things? • Tenacity: knowing by force of habit o I believe it is true because it has always been true. o Repeating ideas increases belief o Flawed way of knowing, as many beliefs are contradictory o E.g. “birds of a feather flock together” vs. “opposites attract” • Authority/faith: knowledge gained by a respected source o I believe it is true because an expert says it is true. o Truth depends on credibility o Saves time in evaluating evidence ourselves but often flawed source and flawed process o E.g. “my meditation guru says babies should only eat eggs the first year of life” • Rational: relying on logic and rationality o I believe it is true because it is logically derived. o Rules of logic applied so that a reasonable conclusion can be derived o More rigorous, however only as good as the assumptions that underlie it 2 o E.g. “All poodles are dogs; Toast is a poodle; therefore, toast is a dog.” Vs. “All poodles are afraid of hot air balloons; Toast is a poodle; Therefore, Toast is afraid of hot air balloons.” • Intuition: knowing by a gut feeling o I believe it is true because I feel it is true. o No direct involvement of the senses but rather a hunch o Psychics o Lacks supporting evidence • Personal experience: knowledge gained by experience o I believe it is true because I experienced it. o Verydifficulttoargueagainst anddifficulttoconveytoothers;subjective;isittypicalorrare? o E.g. “I know what divorce does to children because I experienced it when I was 10.” • Empiricism: knowledge is acquired via senses and observation o I believe it is true because I measured it. o Can be verified o Biases? o Unreasonable to expect something only exists if we can observe it • Science: process of systematically gathering and evaluating rationally empirical evidence to answer questions and test ideas that are publicly verifiable o Limitedbyempiricismbecauseifwecannotdeviseawaytomeasuresomethingthenwecan’t use science (e.g. is there a soul? Is there an afterlife?) o When science cannot create a way to measure, that leaves only reason and those topics become the domain of philosophy or theology What is science? • Science: process of systematically gathering and evaluating empirical evidence to answer questions and test ideas that are publicly verifiable • Scientific disposition = skepticism, parsimony, connectivity • Scientists seek to describe, explain, predict, and control events • 4 objectives/goals of science 1. To describe a. Most fundamental task b. Young areas of science will first spend time describing the subject c. Takes on many forms 3 d. Most common form of science in the area of criminology 2. To explain a. Why those phenomena occur b. Initial explanation of what causes an outcome = hypothesis c. Usually involves reasoning d. As empirical evidence accumulates, scientists then build theories which explain the how and the why events are related 3. To predict a. Oncetheareahasbeenwelldescribed&explained,sciencemovesontomakepredictions from the explanations b. If the predictions are not confirmed, then the explanation is considered faulty and must be revised c. Strongest means by which scientists determining whether their explanations, or hypotheses, for events are correct d. Important in the application of science 4. To control a. Being able to decide the how, when, what, and where of the study and the factors within the study design b. Try to control the phenomenon c. In criminology, it may mean controlling the peers in an effort to reduce recidivism since science demonstrated that antisocial peers describe, explain, and predict criminal behavior Principles of Science 1. Is empirical and systematic – focuses on testable questions (operational definition) 2. Is objective (unbiased) 3. Is testable (operationism) 4. Is publicly verifiable (peer review) 5. Is tentative 6. Is self-correcting (replication) Steps in the Scientific Research Process 1. Find a research idea 2. Convert idea to hypothesis 4 3. Define and measure variables 7. Conduct the study 4. Identify participants 8. Evaluate the data 5. Select a research strategy 9. Report the results 6. Select a research design 10. Refine your research data Generating Research Ideas RESEARCH QUESTIONS • All good research begins with a question  not always stated explicitly 1. Pick your topic you are interested in 2. Think about all the possible aspects of the topic and pick one 3. Think about possible questions for that topic and pick one 4. Focus the question (who, what, where, when)  should not have an obvious answer 5. Refine the question  should include your foci and not have an obvious answer Research Question to Hypothesis • Hypothesis: a statement, designed to answer your research question, and which predicts what you believe will occur when you randomly assign participants to various conditions; must be… o Must be testable o Must be logical (must have a rationale & justification) o Must be refutable (must be able to achieve results that refute your hypothesis) o Must be a positive statement about the existence of something o Must be stated before you start collecting data • Hypothesis statement must answer… o Who are the participants? o What are the groups? o What are you comparing? o What direction are you predicting the results to be? • E.g. o RQ: “Does treatment work for provincially incarcerated first time violent women offenders?” 5 o Hypothesis: “Provincially incarcerated first time violent women offenders who participate in cog behavior treatment will recidivate less than those who do not participate.” FINDING & USING BACKGROUND LITERATURE Literature search: ultimate goal is to find a set of published research reports that define the current state of knowledge in an area and to identify gaps in that knowledge base that your study will attempt to fill Journal articles can be empirical or review articles… • Empirical: report results for the first time, contains detailed methods, stats, and numerical results • Reviews: summarize a body of literature that has been done in a particular area of study o Some use quantitative techniques (meta-analysis) o Others just tell the story Literature you refer to can be classified as… 1. Primarysources:firsthandreportsofobservationsorresearchresultsthatarewrittenbytheindividual who actually conducted the research and made the observations (e.g. research reports published in scientific journals) 2. Secondary sources: description or summary of another person’s work; written by someone who did not participate in the research or observations being discussed (e.g. books, textbooks, introduction of research reports, newspapers, magazines) Many sources for conducting literature searches… • PsycINFO • ERIC • Social Science Abstracts Defining & Measuring Variables 1. Hypotheses 2. Operational definitions 3. Measurement validity 4. Measurement reliability 6 5. Modes of measuring 6. Scales of measurement 2 groups in hypothesis  comparison Variables • A variable is something that varies so that it must have at least 2 levels or values • Scientists examine the causal relationship between variables in an experiment by manipulation, controlling, and measuring variables • Independent variable: the manipulated variable in an experimental model (or grouping variable in quasi-experimental model) • Dependent variable: the measured variable (quantitative in this course) • Controlvariable:keptconstantacrosstheindependentvariablesothatitdoesnotbecomeaconfound variable (one that adversely affects the dependent variable) Designs Experimental Model Quasi-Experimental Model Mixed Method Model Random selection Random selection Random selection Random assignment to the Pre-existing membership in the More than one IVof which there levels of the IV w/ a control levels of the IV is both random assignment and group pre-existing membership in the IV Pre-post measure Pre-post measure Pre-post measure A descriptive study (or non-experimental research) only measures variables but does not manipulate them Shy people are better at reading facial expressions than non-shy people. • Independent variable = shyness  two levels = shy & not shy • Dependent variable = reading facial expressions • Hypothesis = shy people are better • Quasi-experimental = people are already shy or not shy 7 Males who participate in anger management treatment will demonstrate greater reductions in their hostile attributions than males who do not participate in treatment. • Independent variable = treatment  2 levels = anger management & control • Dependent variable = hostile attributions • Hypothesis = greater reductions • Experimental = two values that are being randomly assigned Cognitive behavioral treatment for violent offenders is more effective at reducing recidivism than psychotherapy. This effect will be stronger for males than for females. • Independent quasi-variable = gender  2 levels = female & male (pre-existing category) • Independent variable = treatment  2 levels = cognitive behavioral & psychotherapy • Dependent variable = recidivism • Hypothesis = stronger for males than for females • Mixed methods = 1 quasi and 1 experimental variable Offenders who participate in anger management treatment will demonstrate greater reductions in their hostile attributions than offenders who do not participate in treatment. • Confound variables… o Other treatmentat the same time o Institution where they’re housed or before o Peer associations o Different definitions of “hostile o Gender attributions” o Age o Outside stressors o Ethnicity o Type of offenders What you can’t hold constant  random assignment Operational definitions • The act of operationalizing is the describing of how a concept (or variable) will be measured • Converting an abstract idea into something measurable • Can be something observable, and if not, then taking something conceptual and making it observable • E.g. smart, recidivism, self-confidence, rehabilitate 8 • All variables must be described precisely so that other researchers can replicate your findings • From past research • Use conventional methods for defining & measuring variables • Critically examine the procedure and see if there are better measurement techniques Criminal justice theories… • Biological o A person’s physique is correlated to the type of crime one commits o Criminality is genetic o A chemical imbalance in one’s brain can lead to criminal behavior • Psychological o Criminal behavior is the result of an inadequately developed ego o Inadequate moral development during childhood leads to criminal behavior o Criminals learn their behavior by modeling it after other criminals • Sociological o Socializing w/ criminals produces criminal behavior o Society’s labeling of an individual as deviant or criminal breeds criminality o Failure to reach societal goals through acceptable means leads to criminality Measurement… • Many different measurement options which can influence the interpretation of the data and the outcome of your study Modalities of measurement 1. Self-report a. Types of questions… i. Open-ended ii. Restricted questions iii. Rating scale questions 2. Physiological a. Using physiological performance/measures to get at some underlying construct b. E.g. measuring heart rate for anxiety or level of physical fitness c. Advantage = extremely objective; equipment is reliable and accurate 9 d. Disadvantage = expensive; unnatural situations; questions about construct 3. Behavioral a. Measure some indicator of the construct (e.g. reaction time as a measure of mental alertness after a drug has been administered) b. Can also measure behaviors directly by observation c. Advantage = can measure many hypothetical constructs this way that are hard to see d. Disadvantage = behaviors you are observing may be temporary or influenced by the context Reliability & validity of measurements… • Validity: does the measurement procedure accurately capture the variable that it is supposed to measure (e.g. brain size, bumps on skull, IQ tests); 6 types of measurement validity… 1. Face validity: idea that a test should appear superficially to test what it is supposed to test a. IQ should include questions appropriate for measuring intelligence b. Researchers often disguise questionnaires in order to reduce face validity (and decrease the adjustment of answers to produce a better self image) 2. Concurrent validity: scores obtained from a new measure are directly related to scores obtained from a more established measure of the same variable a. E.g. new test to measure psychopathy differentiates individuals like the standardized psychopathy test 3. Predictive validity: when a measurement of a construct accurately predicts future behavior 4. Construct validity: the property of a test that actually measures the constructs it is designed to measure, and no others a. Tests the construct (leadership and not extraversion) b. Should not measure theoretically unrelated constructs (tests leadership and not reading ability) c. Should predict results related to the theory (e.g. leadership ability should predict who would be a good leader) 5. Convergent validity: using 2 different methods to measure the same construct and then showing a strong relationship between the measures obtained from the 2 methods 6. Divergentvalidity:demonstratingthatwearemeasuringonespecificconstructandnotcombining 2 different constructs in the same measurement process a. Convergent + divergent validity = construct validity 10 • Reliability: repeated measurements of the same individual under the same conditions produce nearly identical values; stability and consistency of a measurement o Typeofmeasurementerrorthataffectreliabilityofmeasurementofthedependentvariable… 1. Observererror:theindividualwhomakesthemeasurementscanintroducesimplehuman error into the measurement process (e.g. stopwatch) 2. Environmental changes: small changes in the env’t from one measurement to another making it impossible to obtain 2 identical env’t conditions (e.g. time of day, temperature, weather, lighting) 3. Participant changes: participant can change b/w measurements producing what appears to be inconsistent or unreliable measurements (e.g. hunger) o Measures of reliability… 1. Test-retest reliability: the degree to which the same test score would be obtained on another successive occasion  good measure has high test-retest reliability 2. Inter-rater reliability: the degree of agreement b/w two or more separate observers who simultaneously record measurements  good measure has high inter-rater reliability Scales of measurement… • Nominal: labels items (e.g. university major, jersey numbers) • Ordinal: orders people, objects, or events along some sequential continuum (e.g. class rankings) • Interval: legitimate differences b/w scale points (e.g. thermometer) • Ratio: a scale with a true Zero refers to the absence of the thing being measured Sampling – Research Strategies – Validity Modality: the in which variable is measured  e.g. self-report modality for laugh experiment Research strategy: general approach to research determined by the kind of question that the research study hopes to answer 1. The descriptive strategy: purpose is to describe variables as they exist 2. The correlational strategy: measures two variables for each individual in order to describe the relationship b/w the variables (weakness = doesn’t talk about causality; “third-variable problem”) 3. The quasi-experimental strategy: compares groups or conditions but lacks one of the critical components, either manipulation or control, that is necessary for a true experiment 11 4. The experimental strategy: to establish the existence of, and demonstrate a, cause-and-effect relationship between 2 variables; attempts to show that changes in one variable are directly responsible for changes in a second variable POPULATIONS & SAMPLES Population: defined by the interest to the researcher Sample: a set of individuals selected from the population and intended to represent the population • Sample is selected from population • Results from sample are generalized to the population Target population: the entire set of individuals of interest to the researcher Accessible population: an accessible group that represents the population and are used for drawing the sample Representativeness: how accurately we can generalize the results from a given sample to the population Bias is a major threat  characteristics that are noticeable different from those of the population Due to selection bias (or sampling bias): sampling procedures favored the selection of some individuals over others Size of the sample is also a major threat  as the sample size becomes larger, the sample results tend to be more representative of the population (30 subjects minimum for each level) • 1 IV, 1 level  n=30 • 1 IV, 2 levels  n=60 • 1 IV, 3 levels  n=90 Sampling Methodology… 1. Probability sampling: odds of selecting a particular individual are known and can be calculated • 3 conditions… a) Exact population size must be known and all individuals listed 12 b) Each individual must have a specifiable probability of selection c) Random selection process • Methods… a) Simple random sampling: a group chosen from an entire population such that every member of the population has an equal and independent chance of being selected in a single sample th b) Systematic sampling: a probability sample that is obtained by selectingevery n person from a list containing the total population, after a random start c) Stratified random sampling: a random sample in which 2 or more sub samples are represented and the sample is to contain equal proportions of each sub sample d) Proportionate stratified random sampling: structuringthe sample sothatitscomposition matches the composition of the population e) Cluster sampling: a group selected by using clusters or groupings from a larger population a. Advantages = quick and easy; group measurements b. Disadvantages = independence of scores • Combined-strategy sampling: researchers often combine 2 or more strategies to optimize the chances that a sample is representative of a widely dispersed population 2. Nonprobability sampling: odds of selecting a particular individual are not known because the researcher does not know the population size or the members of the population; greater risk of producing biased samples a. Convenience sampling: a non-random sample that is chosen for practical reasons i. Most commonly used ii. Compensate for weaknesses by combining w/ other sampling techniques and being cautious about interpretations iii. Also referred to as haphazard sampling b. Snowball sampling: we are led to other participants from previous participants 13 c. Quota sampling: researcher establishes a quota for individuals to be selected from each subgroup (non-randomly) VALIDITY Validity: determined by the q
More Less

Related notes for CRCJ 3001

Log In


Don't have an account?

Join OneClass

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

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.