CJ 280-002 Exam 3 Study Guide
CJ 280-002 Exam 3 Study Guide CJ 280
Popular in Research Methods
Popular in Criminal Justice
This 11 page Study Guide was uploaded by Jennifer Gintovt on Monday April 18, 2016. The Study Guide belongs to CJ 280 at University of Alabama - Tuscaloosa taught by Matthew Dolliver in Spring 2016. Since its upload, it has received 71 views. For similar materials see Research Methods in Criminal Justice at University of Alabama - Tuscaloosa.
Reviews for CJ 280-002 Exam 3 Study Guide
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 04/18/16
CJ 280002 Exam 3 Study Guide Quantitative research Nomothetic o Looking at generalizability and link cases Role of numerical modeling o The use of numbers o Type of measurement is key Wanting to generalize and link things together o Using numbers up front 2 major types Observational Experimental Best for standardization o Why does it matter? You don’t want somebody to be different “all things being equal” courts, opinions, sensitivity, characteristics what numbers are good for ___% is white ___ amount of population is against death penalty Survey methodology: an interview with closed ends o closeended questions & highly structured thought about questions and standardized closed ends so answers are standardized binary scale o usually only have 2 options o male/female; agree/disagree o good because we can code easily & then do stats easily or 1 o false dichotomy have to be careful about if it is really as straight forward as yes or no Likert scale o “feeling thermometer” numerical value of your feeling agree 0510 disagree in 2 styles let somebody be neutral o 110 make somebody answer o 19 Developing good questions: exhaustive V. (mutually) exclusive o want to be both o exhaustive – avoided false dichotomy by giving people all of their options EX: income A.) 30k60k B.) 60k90k C.) 90k + Problems: o You still might not be included which means you failed to be exhaustive o If you make exactly 90k you don’t know how to answer so you make sure categories don’t overlap Exclusive o Making sure you only offer 1 category for people to fall into Doublebarreled problem o Asking 2 questions but only giving one response for one of the questions EX: should gov. abolish death penalty and release all prisoners on death row? th Short, quick, and clear (7 grader rule) o Avoiding attrition dropping out o Want questions to be on a 7 grade reading level o Do it partly so people can go through it pretty quickly Bias o Wording and halo effect o Halo effect is a fallacy of authority EX: many researchers think…what do you think? o Can be subtle or profound Administering surveys 3 styles o selfadministered; mail, email, in person o inperson – interview o telephone rapid and high response rate Observational Designs II Secondary analysis What is secondary data analysis? Primary vs. secondary o Primary: data that you collect directly o Secondary: data collected by someone else VALIDITY All about asking right questions Number of sources o Official gov’t stats EX: UCR o Nonpublic agency records EX: Court case load stats o Primary collection data Researchers reaching out to other researchers EX: crime and victimization Right time, right place Evaluation and the dark figure of crime o Figures that don’t get reported o Matching questions + measures Asking right question o Definitions, procedures, and responses EX: homicide in early 1900s o Keep track of unites EX: forming groups o Errors increase with volume The more data you collect, the more likely you are to encounter errors When do we use secondary data? Money, time, energy, information o Secondary data allows you to span all these things without ever having to take part in them Gives access to data you may not have had access to any other way o EX: UCR – SHR, court records Able to fill in the gaps by combining information. o EX: dates, locations Probability 1 Probability is important! Counterpart to randomness o Randomness is a subtle but incredibly powerful concept Any one particular even is unknown (the outcome is unknown), and yet the outcome of aggregates is very well known How do you know something is random? Because you don’t know what is going to happen ahead of time Humans have free will Our own measures contain errors Generally we conduct analysis on a sample Probability Independent random events grouped together, always produce order/patterns A measure of how likely an event is to occur o Ranges from 01 The higher the probability, the more likely a particular outcome is Definitions… Experiment: o Repeatable procedure for making an observation or measurement Outcome: o A possible result of an experiment Sample space: o The set of all possible outcomes of an experiment Event o Any subset of the sample space of an experiment Mutually exclusive events: Events = mutually exclusive if there is no overlap between the two events o Ex: in Venn Diagram, they do not share a common region Which means events can occur simultaneously EX: getting both a 1 and a 6 on a roll of a die Probability 2 Randomness: Probability helps describe randomness o Individual results v. aggregate calculations Law of Large numbers o Tells us that random events don’t care o Gamblers fallacy The coin doesn’t care about the outcome, but we do I feel like I’m due for a win, because I know I have a 50/50 shot, so I continue to bet Viewing some events that are actually equally likely as being more rare EX: rolling a die and getting a 2 Order arises from groups of random events! o Think about coin flipping example Implications for equal selection We converge on probabilities o I know I will get a little bit of everything if my sample is big enough Small events matter Rare events Must happen because of chance o Rare events happen through chance alone Winning the lottery even though you literally have no chance of winning So why does it matter? Lurking variables Controls for bias The Experimental Design, Part I Treatment: If you have a treatment you are running an experiment The thing in our control as a researcher that we get to introduce or take away o Looking for cause and effect relationship (True/Classical) Experimental Designs o Treatment A new way of doing research o The role of cause and effect Causation’s 3 parts and 2 conditions Hume 3 parts o Correlated o Time order o No outside factors 2 Conditions: o Necessary must be present o Sufficient – numerous ways to do something The goal is to get everything The use of control v. experimental groups o Comparison The only difference between control v. experimental group is the treatment If we see that the group that got the treatment experienced a different result than the group that did not receive the treatment, then we can conclude that the different result was due to the treatment The Savana Effect o Randomness – selection or assignment? – its about both! When we take a sufficiently large random sample we don’t have to worry about getting absolutely every different characteristic Group Pretest x Posttest 1 X X X 2 x x 3 x x 4 x x The large random sample ensures that we get the differences we need 4 step process: Assignment Pretest o Gives you a baseline Treatment Posttest Several common variations o The Solomon four–group design This gives you every possible combination You really know if what you think is happening is really happening Double v. Single Blind designs Blindness refers to what you understand about the study o Singleblind = participants don’t know what group they’re in o Doubleblind = participants AND researcher don’t know the groups The Classical Experiment, Part II Science in the “real” world “Ceteris paribus” – all things being equal – the model experiment o Controlling outside factors in the lab History Certain events cause people to act differently, such as being on high alert Maturation Peoples views, preferences, etc. begin to change Attrition, and others Process of gradually reducing the strength or effectiveness of someone or something through sustained attack or pressure o Controlling inside factors Pretest The Hawthorn effect Are we getting a socially desirable response Finding the perfect match: QuasiExperimental Designs o Experiment with a key difference Randomization o What about threats to internal validity 2 equal groups Someone can always find something that doesn’t quite match between two groups/subjects o Why use it? Ethical considerations EX: abused children turn into criminal adults? Or police officers coming in and out of the academy Nonequivalentgroups design o Matching on key factors o Matching on “types”/unites EX: classes at the academy Natural Experiments Experiment with a key difference o Treatment is “natural” (or unplanned) o Randomness is still the key but with a twist Threats to internal validity o Time or groups Why use it? o Unethical or impractical Think about scale EX: smoking ban and death penalty IRB’s, Ethics, Grants, and Funding Ethics: Professional code of conduct o Human (and animal) subjects Harm, risk, and rewards balancing act o Different from morality because morality is a person’s belief IRB: Internal Review Board o Maintaining and adhering to ethical standards o Gov’t and nongov’t organizations o Judge risk/reward AND safeguards to be used o Focus on Harm, risk, rewards EX: medical study o Internal to structure of institution o Review all studies that are coming through o Weigh out: Potential harms and benefits o Composed differently based on institution Research at one institution may get past an IRB while it wouldn’t at another institution Stanford Prison Experiment Remuneration: Incentivizing people (ex. Giving gift cards for participation) Paid for your work: (Research) Grants: o Monetary award Competitive; you make your proposal o Government, Nonprofit organizations Contributed to intellectual and social development o Major sources include NSF (National Science Foundation), NIJ (National Institute of Justice), NIH (National Institute of Health; Drug research) NIJ COPS office o Focused on departments nationwide that wanted to institute Community Oriented approach UA internal grant funding (seed funding; they give you a little money now with the hope that it will turn into a lot of money later) OSP and Pivot Giving a Presentation/Writing a paper The Hour Glass Paper formatting o Starting with large overarching idea o Bringing it in to more specific details o Coming back out to answer “so what?” or “why should I care?” Creating a conversation: You start with a question Use of “elaborative encoding” to bring them in Helpful tips: Rewards for progress Someone to check work Stop in the middle of line or section Skip around, let work come to you Pocket paper Although, nevertheless, because, thus o Many argue, and I agree…however o New evidence, old way o Old evidence, new way o Old evidence and old approach, new way Quiz: Key Terms 1._____ Ceteris Paribus 10._____ Treatment 2._____ Likert Scale 3._____ exhaustive 4._____ Exclusive 5._____ Doublebarreled problem A. making sure you only offer one category for people to fall into B. data collected by someone else C. “feeling thermometer”/ numerical value placed on feelings D. data collected directly E. using authority to direct a response 6._____ Halo effect 7._____ Primary data 8._____ Secondary data 9._____ Mutually exclusive F. the thing in our control as a researcher I. avoid false dichotomy by giving all that we can induce or take away possible options G. asking two questions but only giving a J. there is no overlap response to one question H. all things being equal Quiz: Short answer/ Multiple Choice 1. Quantitative research focuses on _____________. A. Statistics B. Interviews C. Numerical Values D. Meaning individuals attach to things 2. What are the two styles of the Likert Scale? 3. Which of the following is not one of the benefits of using secondary data? A. Money, time, energy, information B. Access to data you may not have had access to otherwise C. Secondary data is not collected directly and therefore may not be correct D. Able to fill in gaps by combining information 4. What are Hume’s 3 parts of causation? 5. Research grants… A. Are a major source of funding for researchers B. Typically come from governmental or nonprofit organizations C. Are competitive D. All of the above
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'