CCJS 300 09 04 2014 Approaches to Research 1 Positivist approach of the natural science from Auguste Comte 2 Interpretive Understand everyday lived experiences from Max Weber 3 Critical Generating liberating knowledge for a better world form Karl Marx Sigmund Freud etc Scientific Research and Time 1 Cross sectional Research Snapshot of a collection of people at one time 2 Longitudinal Research Time as a variable a Time series Observe different people at multiple times b Panel observe the exact same people at 2 or more times c Cohort observe people who shared an experience at 2 or more times d Case study observe a small set of people intensely across time Purposes of Research 1 exploration 2 description 3 explanation 4 evaluation and application The Research Process Deduction Hypothesis operationalization Research design measurement Data Gathering analysis Findings induction Theory General steps in Empirical Research 1 Problem formulation 2 Research design 3 Data collection methods 4 Analysis and presentation of findings 5 Conclusions interpretations limitations of the study Ethics in Research 09 04 2014 Definition Doing what is morally and legally right in the conducting of research The most serious question in research Biomedical examples Social science examples Stanley Milgram obedience to authority People learn things correctly when you make a mistake how much punishment is necessary for learning IRBs 40 psychiatrists how many people would go all the way 1 1000 of people in the condition predicted it was actually 50 The Experimental Model 1 The experimental model as a research design The Gold Standard 2 Developed in the Natural Sciences experimentation in the social sciences began in Psychology perceptual psychology behaviorism mental measurement human factors 3 Experimentation is best suited for 2 of the 4 purposes of Research explanation and Evaluation 4 Topics in experimental CCJS research effectiveness of a arrests for domestic violence b Drug education programs like DARE c Correctional Boot camps etc 5 The 3 steps to establishing causality linking independent and dependent variables a Demonstrate that a relationship exists between key variables b Specify the time order of the relationship c Eliminate rival causal factors expose spurious relationships 6 Rival causal factors two general types A internal validity did variable X the IV produce a change in variable Y the DV 7 Internal factors variables related to internal validity a History what else is going on out in the world other than our experiment that could effect it b Maturation c Testing d Instrumentation e Statistical regression extreme tends to moderate over time f Selection bias randomization g Experimental Mortality people dropping out h Selection maturation interaction 9 18 2014 8 External factors variables related to external validity A interactive testing effect B interaction of selection bias and experimental variable C reactive effects of experimental arrangements Hawthorne effect D multiple treatment interferences repeated measure design 3 Two other factors A Post Hoc Error B Placebo Effect 10 Experimental Designs A True Experimental Designs use random assignment to assure group equivalence B Quasi Experimental Designs No random assignment maybe use matching for grouping equivalence C Pre experimental designs No group equivalence 11 The classic experimental designs include A equivalence of groups start randomization B Pretests and protests C experimental and control groups 12 Examples of classic experimental designs see Hagen A The Provo and silver lake experiments B The Kansas city preventive patrol experiment C The Minneapolis Domestic Violence experiment How do the design of these studies fit with the experiment examples control experimental 13 Advantages and disadvantages of experiments A advantages 1 The big One Control 2 Relatively quick and inexpensive 3 Manageable can be replicated by other researchers B Disadvantages 1 The Big One Artificiality 2 May be hard to recruit respondents 3 Experimenter effects biases Data analysis and statistics 09 04 2014 CCJS is a science Stats is a tool we use to answer our questions different types or branches of statistics Descriptive describe a variable or variables Inferential statistics infer something relationship or association Explanatory research vocab language Variable changes Theories hypothesis Validity causality IV DV Levels of measurement NOIR nominal ordinal interval ratio Measures of central tendency What is going on in the middle of a data distribution Mode most frequently seen actual number or value Median middle number Mean X bar is for samples Mu is for population Advantages takes all values of the variable into account An efficient estimator of central tendency Used in subsequent calculations Disadvantages may be changed radically by extreme scored or outliers Measure of dispersion speadoutness classic distribution VR variation ratio Range highest value lowest value IQR interquartile range Variance Statistics depends on the field of probability as its infrastructure and origin Rules for probability bounding rule Central Limit Theorem Don t worry about the underlying shape of the popular distribution the sampling distribution provides a link between samples and populations Data Analysis and Statistics Point estimation confidence interval Hypothesis testing Step 1 assumptions null Status quoi nothing going on and alternative hypothesis something happening perhaps a statistically significant difference Step 2 Appropriate statistic and that statistics distribution Step 3 Alpha level critical value directionality rejection region Step 4 Calculation Step 5 Decision reject or fail to reject the null Interpret your results Using theory Different Hypothesis Tests Z test we have some parameter information T test direct comparison of means between groups independent and dependent samples 3 or more groups need different statistics test compare means simultaneously AOVA F statistic F test T tests and ANOVA IV is groups categorical DV is interval or ratio level variable Advances Statistics Regression multiple regression Build in a sense of causality Exam 09 04 2014 All notes discussions handouts from class includes elegant universe drug journal 2 from Campbell and Stanley book table internal external validity and Milgram film What does the R mean in the table Random randomization characteristic of experiments NOT ON EXAM doping for gold SPSS programming
View Full Document