Chapter 2 Research Methods Announcements PQ 2 Due Sat Contrib 2 Due Sat Chapter 3 Wed Tech problems or concerns Enrollment Deadlines Drop Deadline Sept 16th Anything from the chapter is fair game on the exam Overview Empirical Method Correlational Evidence Causal Evidence Ethics Hypothesis testing and theory development Measurement Empirical Method Empirical Method Empiricism The belief that the best way to learn is through direct observation As opposed to dogmatism Theory An explanation of physical natural phenomena Hypothesis A falsifiable prediction Testable falsifiable and specific Empirical Method Population The broader group that the researchers want to learn about Sample The group of people from the target population that participate in the study Random All in population have equal chance of being in study Convenience Sample is recruited via non random means email poster PSYC 104 NOTE Do not confuse random sampling with random assignment to groups or conditions Measurement Measurement Operationalization or an Operational Definition The specific way in which a phenomenon is defined and measured for the purposes of the study For example depression symptom severity may be operationalized as total score on the HDRS a depression questionnaire Construct Validity Definition should accurately describe the target phenomenon Contrib 2 Pt 1 1 Prof B is running a study in which he hopes to learn if people who are introverted are at higher risk of insomnia sleep difficulty How might Prof B operationalize each of his variables of interest 2 Prof B has also been hired to run a behavioral intervention with a young man Bubba in which he ll have independent observers count Bubba s behaviors Prof B interviews Bubba s teachers and they tell him that Bubba has been hitting other students Operationally define hitting such that the independent observers will agree about when the target behavior occurs Measurement Takeaways There is often more than one way to measure a variable Be wary interpreting questionnaires or indirect measures Flawed measures can bias results and lead to unfair consequences Be clear and specific Study Designs Correlational Evidence When at least two measures are given to participants and data points are compared Types of studies that yield correlational evidence surveys longitudinal studies quasi experimental designs 1 Did the researcher change a variable or subject participants to a condition 2 Were the participants randomly assigned to conditions for between group designs If no to either of the above then it s likely correlational evidence Correlational Evidence When interpreting correlation coefficients The or symbol indicates ONLY direction not magnitude Closer to 1 or 1 means larger magnitude closer to zero is smaller magnitude 0 15 0 29 is weak 0 3 0 59 is medium 0 6 or greater is strong Setup NOT A PART OF CONTRIBUTION Consider your opinions regarding each of these issues DO NOT WRITE WHAT YOUR STANCES ARE IN YOUR ASSIGNMENT Gun Control vs Gun Rights Juvenile court mandated Measles vaccines vs Parental right to choose Privatized Health Insurance vs Public Health Insurance Interventionist Foreign Policy vs Isolationist Foreign Policy Preventative Climate Pro Green Energy vs Reactive Climate Anti Green Energy Policy Activity NOT DUE You are introduced to an average undergraduate student Joe Shmo with similar public speaking abilities and opinions as you You are both informed that you are going to debate in front of an audience of people mostly undecided on each of these issues Your goal is persuade the audience towards your side except your going to be defending the OPPOSITE of all of your opinions and Joe Shmo is going to be defending your preferred positions You have as much time as you like to prepare your arguments but you cannot access the internet or otherwise obtain new information All swans are white Popper Limitations of Correlational Designs Cannot falsify alternative causal explanations for the relation observed Factor A may cause change factor B if temporal precedence isn t established Factor B may cause change in factor A if temporal precedence isn t established The third variable problem Factor C may be causing the changes observed in either A or B or both Strengths of Correlational Evidence Convenient and low cost Explore relationships and avoid ethical concerns with random sampling E g study of the relationship between smoking and birth complications Types of Experimental Causal Designs Group Experimental Designs Between groups One condition of the IV experienced by each group Within Groups Each participant undergoes each condition of the IV Random Assignment to condition participants have equal chance of being assigned to any condition Not to be confused with random sampling And also single subject case experimental designs but they get ignored Causal Evidence Elements of an Experiment Independent Variable IV variable that s manipulated at least two conditions Dependent Variable DV outcome value depends on the condition or level of the IV in theory Experimental and Control Conditions Groups Random Assignment to conditions Measure the Dependent Variable DV Statistically compare the data from the two or more conditions Causal Evidence Qualities of Experimental Evidence Internal Validity How well experiment rules out alternative explanations for the results Strength of experiments over other types of studies Double blind When neither the researcher nor the participant knows what condition the participant is in External Validity How well results apply to real world situations Often times it s difficult to recreate real world scenarios and hide the manipulation Contrib 2 Pt 2 Prof B wants to know if tanning beds that use UV light are more positively reinforcing than tanning beds without He recruits participants and on Day 1 has the participants use a tanning bed with or without UV light On Day 2 he has participants use the opposite tanning bed Neither the participant nor the observing researcher knows which tanning bed they re using on either day On Day 3 the participant is admitted into the room with both tanning beds and they are allowed to select which one they use not knowing if it has UV or not only that the beds are in the same locations as they were before Is this a within groups or between groups design 5 6 What is the Independent Variable IV 7 What is the Dependent Variable DV 8 What strategy did Prof B use to limit the effects of bias or demand
View Full Document
Unlocking...