Exam 2 Study Guide Exam 2 will cover Chapters 4 7 45 questions Bonus Question Kevin Conroy voices Batman True false multiple choice matching Topics to Focus on Conceptualization the process by which we identify what we mean by a concept the product is a specific agreed upon meaning for a concept o Concept construct derived by mutual agreement from mental images Operationalization the process of developing an operational definition o Operational definition definition in terms of specific operations measurement instruments or procedures Typologies produced by the intersection of two or may variables to create a set of categories or types 3 styles of policing o watchman style emphasized order maintenance and ignore minor violations while focusing on big stuff o Legalistic style focused on law enforcement not just maintaining order They will make an arrest and record that arrest no matter what the crime is o service style focus on both law enforcement and order of maintenance but are less likely to make an arrest Reliability thing is being measured o Reliability requires that the indicator gives the same result each time the same o Inter rater reliability the reliability of two raters looking at the same behavior Example two TAs grading the same paper o Intra rater reliability reliability of one person over time One professor grading papers is it the same over time o Representative reliability reliability across subgroups or subpopulations ask students their GPA then cross reference with school records to see if women or men lie more o Internal consistency reliability multiple questions or indicators to measure some concept to see if they produce the same result For example If you have 10 questions give them 5 at a time then compare the consistency between the two sets of questions o Equivalence reliability same as interrater reliability o Know the different ways you can improve reliability Conceptualize clearly Hard to measure ambiguous topics need to be more specific Increase the level of measurement Try to measure at the highest level of precision possible Ordinal rather than nominal if possible Use multiple indicators More indicators means higher reliability Use pretests and pilot studies Want to do a test run Use established measures So you don t have to demonstrate reliability Training of research workers interviewers Make sure everyone is on the Validity same page o Validity measures the extent to which the measure actually reflects the real meaning of the concept it is supposed to measure o Internal validity vs external validity Internal you re doing a study This looks at if your conclusion actually reflects what happened External how well does your conclusion generalize to other settings o Generalizability how can the results of the study apply to other situations where the factors are the same how applicable are the results o Construct validity based on the logical relationships among variables How well your measurements and your variables generalize to real world concepts o Criterion validity compare the results of the measure to some trustworthy alternative measure Concurrent validity indicator must be associated with a preexisting indicator that is judged to be valid For example you make your own intelligence test and give it to a group of people and the average is 100 To see if it is a good test give them a basic standardized IQ test and if the average is also about 100 your test has concurrent validity o Predictive validity where an indicator predicts future events that are logically related to a construct such as SAT ACT LSAT MCAT o Face validity does the measure make sense Just look at it and see if it logically makes sense o Threats to internal validity Compensation researchers treat the control group differently because they feel bad for them The way you treat them could have different affects than if you would ve treated them the same Compensatory rivalry they know what group they are in so people in the control group try and compensate or work harder to measure up to the experimental group o Nominal indicate only that there is a difference between categories o Ordinal indicate that there is a difference plus the categories can be ordered or ranked If you say the statement research methods is the best class and the answer choices are strongly agree agree neutral disagree strongly disagree o Interval indicate categorical differences that can be rank ordered but they also specify the distance between categories Levels of measurement Temperature in Fahrenheit Monday it was 70 and today it is 50 there is a 20 degree difference that is an interval measure IQ of 80 and wife has an IQ of 120 there is a 40 point difference there interval measure o Ratio include everything in the previous 3 levels plus there is a true zero make proportions ratio possible Income you could make 0 10 000 or 100 000 a year Distance between two points it could theoretically be 0 feet The classic experiment o Gold standard in social science research o Use experimentation in our everyday loves Develop generalized understanding o Independent and dependent variables IV experimental stimulus Treatment Only has two conditions traditionally it is present in the experimental group and absent in the control group Dichotomous variable has two and only two attributes such as gender The independent variable is a dichotomous variable o Pre test and post test Measure DV in pre test expose subjects to IV measure DV is post test Without a pretest we cannot be as confident that the IV directly affected the Experimental group pretest DV IV posttest DV Control group pretest posttest outcome o Experimental and Control groups Experimental group exposed to IV experimental stimulus Control group not exposed to IV experimental stimulus Study in 1970s testing productivity in a factory Wanted to see effect of lighting on productivity turns out it was the effect of the supervisors not the lighting so they needed a control group Putting them in the groups must be RANDOM assures that the two groups are statistically identical o Fourth element that really makes the experiment is the randomization of putting people into those two groups experimental and control o Topics appropriate for experimentation limited and well defined concepts you want to know very specifically what you re talking about Hypothesis testing Explanatory research questions trying to determine causation Examining small group interaction among people
View Full Document
Unlocking...