CCJ4700 Exam 2 Review Measuring crime Uniform Crime Reporting Program purpose provide uniform definitions of crime data collection content of the data source part I offenses index murder rape robbery aggravated assault burglary larceny MVT arson added in 1979 most likely to be reported can determine that a crime occurred occur in all areas occur with frequency serious crimes 2 different indexes 3 forms of data violent crime index homicide rape robbery aggravated assault property crime index burglary larceny MVT crimes known to police Supplementary Homicide Reports SHR crimes cleared by police characteristics of every single count of crime if two different crimes happen at one time only the more serious crime is reported e g homicide is well reported and larceny in underreported excludes offender characteristics victim characteristics individual level data hierarchy rule factors affecting reporting of crime reactive mobilization process seriousness factors affecting recording of crime type of crime non reporting result in underestimates Organizational pressure Crack downs Organizational changes Diff interpretation Professionalism Discretion acceptable and unacceptable uses of UCR data acceptable homicide and auto theft local agency decisions trends within units unacceptable NOT across jurisdictions one city might report crime more than another CCJ4700 Exam 2 Review National Crime Victimization Survey purpose more reliable estimates situational factors where did it occur what time of day demographic data victims offenders characteristics of both design reference period 6 months bounding interview the data does not become part of the tabulations way of making sure that we are not mistaking what happened in the last 6 months screening questions when asking the interviewee about their victimizations you ask what crimes they were the victim of first then move on to the details of each short cues telescoping to deliver as many possible cues in a short period of time by asking them in rapid succession to one another to help prompt memory of incident moving something out or bringing it into the reference period data content Part I UCR offenses same definitions not homicide not arson add simple assault Criminal Victimization in the United States victimization estimates offense estimates national level individual level data demographic variables victim offender situational variables problems with NCVS problems with interview boring learn to under report CCJ4700 Exam 2 Review social desirability recall problems people forget details phone interviews much less expensive lower response rate less valid information Sampling Chapter 5 population the entire set of elements in which we are interested e g individuals cities states countries prisons schools sampling frame list of all possible elements often hypothetical can create systematic error use best approximation recognize limitations sampling element individuals who are selected and interviewed from the sampling frame sampling ratio would like to get the biggest sample possible ration of sample to population 500 from 50 000 SR 1 in 100 1 parameter vs statistic parameter characteristic of population income average median always unknown statistic characteristic of sample estimates the parameter sampling error any difference between between the characteristics of a sample and the characteristics of the population from which it was drawn the larger the sampling error the less representative the sample is of the population non probability sampling available subjects haphazard accidental convenience sampling conveniently accessible people in a mall people on the street students in a class cheap and easy not representative useful for pre testing purposive judgmental sampling specifically selecting certain cases for a reason CCJ4700 Exam 2 Review based on purpose of study 3 cases where appropriate especially informative study of unusual cases study of a wide range of cases snowball sampling uses sample cases to find more cases exploratory quota sampling match characteristics of population difficult to know population info can still be biased cheaper and easier that probability be cautious probability sampling simple random sampling SRS a method of sampling in which every sample element is selected only on the bias of change through a random process compile sampling frame assign number to each case use random numbers to select rarely used in practice systematic sampling a method of sampling in which sample elements are selected from a list or from sequen tial files with every nth element being selected after the first element is selected ran domly within the first interval similar to SRS sampling interval population sample size potential for systematic bias stratified sampling cluster sampling a method of sampling in which sample elements are selected separately from population strata that are identified in advance by the researcher separate population into strata randomly select within strata representativeness across strata oversampling sampling in which elements are selected in two or more stages with the first stage being the random selection of naturally occurring clusters and the last stage being the random selection of elements within clusters cluster unit containing sampling elements clusters are randomly selected smaller clusters randomly selected advantages no sampling frame reduced cost CCJ4700 Exam 2 Review disadvantages increased error at each stage need enough clusters randomness EPSEM Equal chance of being selected probability proportionate to size PPS example 30 out 30 000 colleges 100 students from each college college 1 400 students chance is 100 400 25 college 2 40 000 students chance is 100 40 000 25 choices to make interview more at large college interview fewer at small college make large colleges more likely to be selected Causation and Experimental Design Chapter 6 criteria of causation deterministic vs probabilistic necessary and sufficient necessary Y will not occur without X sufficient X will always lead to Y temporal order association non spuriousness temporal order cause must occur before effect can be difficult to establish reciprocal effects feedback loops association as X varies Y also varies 3 characteristics strength direction linearity non spuriousness effect of X on Y is not caused by 3rd factor education X criminal offending Y parental SES Z could arrow to education or criminal offending cross sectional vs
View Full Document
Unlocking...