Topics to Focus on Exam 2 Study Guide Chapters 4 7 Concept concept derived by a mutual agreement from mental images conceptions Conceptualization process by which we identify what we mean by a concept product is a specific agreed upon meaning for a concept Operationalization Process of developing an operational definition o How are you measuring variables o Operational definition definition in terms of specific operations measurement instruments or procedures Typologies produced by the intersection of two or more variables to create a set of categories or types o Continuous variables have an infinite number of values that flow through a continuum Can be divided in to many smaller increments o Discrete variables have a relatively fixed set of separate values contain distinct categories Indicators a sign of the presence or absence of a concept the variables that represent a concept Verbal abuse loud insults curse o Nommal definition attributes must be exhaustive and mutually exclusive Exhaustive able to classify every observation in terms of 1 of the attributes Marital status single divorced complicated widowed married Mutually exclusive able to classify every observation in terms of one and only one attribute Age 0 3 3 7 7 11 11 15 16 19 3 styles of policing o Watchman emphasis on maintaining order but not through informal methods Focused on resolving disputes and keeping the peace Rural small town blue collar towns and those with a diverse population Wide discretion but is sometimes misinterpreted as discrimination Crime occurred and police responded o Legalistic emphasis on violations of law and relies on threats of arrest or actual arrests Strict enforcement Resolves disputes using formal methods Common in larger cities with diverse populations and well organized governments State police and state highway patrol agencies Little discretion offenses rarely overlooked o Service emphasis on serving the public and helping the community Police agencies work with agencies outside the police such as social services children s aid homeless shelters etc Reliability requires the indicator gives the result each time the same thing is measured Ex Hitting the snooze button every morning o Stability reliability across time Test retest method parallel forms method o Intra rater reliability the degree of agreement among repeated administrations of a diagnostic test performed by a single rater o Representative reliability across subgroups and how they differ Age sex class gpa o Internal consistency reliability multiple indicators for same concept Split half method 5 questions on 1 survey 5 on another o Equivalence reliability among 2 or more people to cover same thing Inter rater reliability agreement among raters Ex 1 professor and 2 TA s giving student the same grade o How to improve reliability Increase level of measurement Conceptualize clearly Use multiple indicators Pretest and pilot studies Use established measures Training of research workers interviewers Validity Measures the extent to which the measure actually reflects the real meaning of the concept it is supposed to measure others Can never be sure about validity but CAN identify measures more valid than o Internal Validity Results accurately reflect what happened in the study Threats to internal validity History major event that can mess up results o Ex 9 11 affecting security checks in airports Maturation change of individuals over study o Become hungry tired Testing effects pretest affects outcome of posttests Instrumentation changes in measurements Statistically regression usually towards average o Measuring extremes over time whether high or low approaches the average Selection bias two groups not the same Usually when cannot randomize them Experimental mortality people drop out of study over time dropping average Compensation researchers treat control group differently because they know numbers called attrition Causal time order clarity in timing o Threats to internal validity they don t receive treatment Cancer pill vs sugar pill group so they compensate compete Compensatory Rivalry subjects in the control group know they are in the control Both lead to a double blind experiment in which one special member of the staff knows which group is which but no one else does o External Validity degree of generalizability of results Generalizability generalizing experimental findings to the real world Construct validity adequate dosage treatment Statistical conclusion are the statistics correct Sample size particularly in small samples Strength of relationship between variables o Construct validity based on the logical relationships among variables o Criterion related validity compare the results of the measure to some trustworthy Concurrent validity indicator must be associated with a preexisting indicator that o Predictive validity where an indicator predicts future events that are logically related to a alternative measure is judged to be valid construct o Face validity does the measure make sense o Reliability is necessary for validity though does not guarantee is easier to achieve Levels of measurement o Nominal indicate only there is a difference among categories Race gender o Ordinal indicate everything in the previous level there is a difference plus the categories can be ordered or ranked Place in a race first second third o Interval include everything in the previous 2 levels indicate categorical differences that can be rank ordered but also specify the distance between categories Temperature IQ o Ratio include everything in the previous 3 levels as well as a true zero makes proportions and ratios possible Cannot use Fahrenheit temp can use Kelvin Income distance between 2 points The classic experiment o Randomization Makes groups statistically identical Important for when sampling an entire population o Different variations of the classic experiment Double blind experiment Post test only design Factorial design multiple experimental groups Solomon four group design does the best job at controlling external validity Classical true experiment o Exp 1 IV post o Control1 post o Post test exp 2 IV post o Control 2 post o The classic experiment generally consists of Independent and dependent variables Pretesting and post testing Experimental and control groups Quasi experimental design only difference is this one does NOT have randomization o No randomization o Time series designs why would you use one over another Measurements
View Full Document
Unlocking...