Monday September 13 Making Observations example An operational definition specifies the operations used to measure something For o Stress hear and respiration rates OR self ratings of tension and anxiety o Hunger number of hours since eating Sampling Selective Observation Usually research studies take samples from a limited portion of the entire population o The sample must be representative of the population in order for the results to be generalizable to the entire population of interest Random selection from the population of interest must be done to ensure that the sample is representative of the population Drawing Conclusions Consider other interpretations o If the observations are inconsistent with the hypothesis Then the hypothesis can be rejected Descriptive Research Descriptive Research involves carefully observing and describing phenomena o Natural Observation refers to the in depth study of a phenomenon in its natural setting For example Studies of primate behavior in the world Observation of children s play o Case Histories are in depth observations of one person or a small group Useful when large numbers of subjects are not available Often employed in clinical research Drawbacks of the case study method include Small sample size hence limited generalizability Susceptibility to researcher bias Survey techniques ask questions of large numbers of persons to gain information on Survey Research attitudes and behavior o 2 main approaches Questionnaire on paper or computerized Interviews in person or by telephone o Disadvantages of survey approach Sampling issues People may not report their thoughts or feelings accurately Intentionally Correlational Analysis or unintentionally Studies where the relationships between 2 variables are measured but not manipulated o The variables upon which correlational analysis is done may come from case studies naturalistic observation or surveys o A correlational coefficient r is calculated Values from 1 to 1 Positive correlation variable High values of one variable go with high values of the other Monday September 13 o In psychology a large correlation is considered to be 50 a moderate correlation to be about 30 and a small correlation to be about 10 o Correlations between variables do NOT establish causality Only experimentation can determine which variable is the cause and Experimental Research which the effect Experiments ask whether systematic variation in one variable produces variation in another variable In other worlds does one thing cause another o Independent Variable IV Manipulated by experimenter o Dependent Variable DV Participant s response Conditions of an Experiment Control Condition o A condition in an experiment is one of the variations of the independent variable o A condition that is as similar as possible to the experimental condition s except that it does not involve any level of the treatment 1 no alcohol consumed Avoiding Bias o Bias in experimentation refers to the systematic variations in behavior i e the DV caused not by the experimental o The Placebo Effect The subject of an experiment may have beliefs about the experimental treatment that alone can result in psychological or physical changes The control condition of the experiment should be designed to account for the effect of subjects expectancies In the previous alcohol example the subjects should not be able to detect the amount of alcohol in the drinks by taste or smell o For example use vodka no smell for alcohol and flavor the drinks with strong syrups In studies of medication give control subjects treatments pills injections etc that seem the same as the real medication but have no active ingredients If the researcher is not careful the subject may get an idea of how the researcher wants the subject to perform The demand characteristics of an experiment are aspects of an experiment that may lead the subjects to get such an idea Demand characteristics may affect the results o Double blind studies control both subject and experimenter expectancies neither the subjects nor the experiment know the condition o Demand Characteristics Quasi Experiments Like experiments but without random assignment o Examples Males vs Females Depressed vs Non Depressed People Limits of Experimental Research Monday September 13 Complex real world issues can not be easily studied in the laboratory o Lab studies may have problems in external validity That is the lab situation may not sufficiently reflect the real world For example prejudice memory recall
View Full Document
Unlocking...