EXAM 1 MATERIAL The Cycle of the Scientific Progress Theory Data Cycles o If then reasoning enables us to test and refine theories o Set of statements from an observation that describes general principles about relationships between variables o Proving a theory cid 224 supported by data cid 224 falsifiable Basic Applied Cycle o If then reasoning that enables us to take basic findings and apply them to more realistic scenarios Empirical Evidence o An observation or measurement that contributes to either verifying or falsifying a claim about what s true o Evidence will be the same no matter who observes it Ex ghost are not objective because we have no evidence to stand for or against them while being shot is objective because we have firsthand accounts o Any situation or procedure that creates empirical evidence which allows a Empirical Testing claim on truth to be verified or falsified o MUST BE Verifiable Replicable Golden Assumption o If a subject exists in nature it is knowable o It is possible to fully describe explain even complex things brain relationships etc o The Object of study is lawful if we could perfectly control all the input variables we could perfectly predict control the outcome The importance of Replication o When others can repeat it and get the same results 3 Types of Claims o Association Argues that one level of a variable is likely to be associated with a particular level of another variable Ex Sexual orientation linked to handedness o Causal o Frequency Goes even further arguing that one of these variables is responsible for changing the other Ex Summer sun may trigger suicidal thoughts Describes a particular rate or level of something Ex 8 million people consider suicide Psychological Construct o Any explanatory variable that is not directly observable or tangible o Must be operationalized Ways to Operationalize psychological constructs o Self Report o Observational Behavioral Recording people s answers to verbal questions about themselves ex questionnaire or interview Pro s Con s Easy low cost large anonymous samples may be the most appropriate format Open to fabrication social desirability biases memory distortions laziness may not be useful for non conscious constructs Recording observable behaviors or physical traces of behavior ex counting the number of times a certain event takes place Pro s Con s More shielded from respondent bias can be recorded w less interference sometimes most appropriate operationalization of a construct experimenters seeing what they want to see confirmation bias discrepancies may arise over whether the behavior was actually produced ethics Can be more complicated to collect subject to Recording biological data believed to be associated with a construct ex dopamine epinephrine levels Pro s Con s hard to consciously control or fake can be quite precise perceived as more credible can be expensive time consuming to collect may require technical expertise w machinery procedures and may be most sensitive to uncontrollable sources of error o Physiological Scales of Measurement o Nominal Categorical o Ordinal Eye color religion birth month Variables in which the values indicate some kind of ranking Central tendency time length area volume mass force energy pressure o Interval Zero does not mean zero Temperature o Ratio Zero means zero Time Validity o Face Seems to plausibly and reasonably capture the construct of interest Ex Measuring intelligence as the speed at which someone can recite the alphabet is not very face valid o Content Must capture all parts of a defined construct Ex measuring intelligence simply by scoring 10 math problems does not fully capture the construct the big picture 3 empirical way to assess validity o Predictive Should predict future outcomes Ex entrance exams SAT ACT BAR o Concurrent RELATED to and correlated with concrete outcomes that it theoretically SHOULD be related to Ex predicting happiness should be recorded by the o Divergent aka Discriminant outcomes of smiles Should be UNRELATED to measures of other constructs that are unrelated Ex Changing intelligence tests to discriminate intelligence from culturally biased knowledge Reliability o Internal People are responding similarly to items meant to assess the same construct Ex liking pizza does not correlate to your mood o Chronbach s alpha Alpha 70 and higher is preferable o High Internal Reliability means people are high ish or low ish on all items Subjects should score similarly at time 1 as they do at time 2 o Test Retest Pearson s R o Inter Rater Scoring behavioral or observational measures Ex measuring aggression o Pearson s R o Pearson s R 70 or higher cid 224 strong 30 70 cid 224 moderate 30 or lower cid 224 weak Open ended questions o Respondents generate their own answers o Advantages Responses can be rich in detail Respondents can take their answers in many directions Allows for greater discovery of related constructs Responses can be hard to code and summarize Responses may not be address the researcher s interest o Disadvantages Close ended Questions o Respondents choose from answers the researcher provides o Advantages o Disadvantages Quicker and easier to score and summarize When responding a person will always provide a relevant answer Preferred to open ended questions when possible and when valid Data do not provide rich detail Response options may not capture the true response square peg round hole o Varying the order of item completion AND TRACKING IT in order to Counterbalancing rule out order effects o 3 2 1 cid 224 3 1 2 cid 224 1 2 3 Social Desirability o Giving socially acceptable responses in an actively deceptive way genuine in their overly optimistic self delusions o The entire set of people of interest o A subset of the population of interest Population Census Sample Survey Representative Sample made Probability Sampling Increases generalization of the sample o Sample that is like the population with respect to measurement being o Ideally works to ensures that every individual in the population has an equal chance of being represented in the sample o Allows generalization rules out alternative explanations of sampling biases Probability Sampling 3 Types of o Simple Random Sampling Every person has a specified probability of being included o Systematic Sampling Every nth person is selected from a list o Stratified Random Sampling Subpopulations are identified and proportions are assessed Final sample reflects the proportions
View Full Document
Unlocking...