© 2013 WedePsych 100 – Lecture 3 Fall ‘13 - WedeLecture 3 – Methodology*Why do the number of assaults increase when ice cream sales increase?The Scientific Method • Steps (details can vary)1. Perceive-noticing something in the environment-Notice when kids watch violent TV they may act out violently2. Hypothesize-possible explanation for why things happen-Find the best theories out there explaining the possibilities of why3. Test-experiments, case studies…4. Draw conclusions-based on test results you decide if things support or reject your hypothesis -Can never prove something is true…everything can be redefined. 5. Report, Revise, Replicate-critical part of sci.6. Repeat –sci is a never-ending process • Research Design:– Specific method a researcher uses to collect, analyze, and interpret data– Three types:• Descriptive research• Correlational research• Experimental researchDescriptive Research-Naturalistic Observation-observe people or animals in their natural environments-Advantages-realistic behaviors (rather than in a lab, they may change behavior)-Disadvantages-when people are being watched they behave differently, people watching may be biased come to rash conclusionsLaboratory Observation-bring people into lab (ex. Classroom) watch them under more controlled conditions-Advantages-specialized equipment is available in labs -Disadvantages-may not get the same behaviors in the natural environment, in the lab people are aware they are being watched and that may give less accurate resultsCase Studies – study one individual or a couple in great details-Advantages-extensive studies on a personal characteristic, a lot of info 1© 2013 WedePsych 100 – Lecture 3 Fall ‘13 - Wede-Disadvantages-we doesn’t know if what we find with those people will apply to anyone else…don’t know if those results will generalize Survey –technique for ascertaining the self-reported attitudes, opinions or behaviors of peopleo Representative sample is key-need this to represent the whole populationo Random sampling – ea. member of a population has an = chance of being selectedo Problems with surveys Wording effect -the way you word statements/questions can affect the outcome- Stay away from double negatives. Ex. Of the holocaust questions Knowledge – people you are surveying need to understand the words used in the question…what you’re asking Summary Case studies, surveys, and naturalistic observation all describe behaviors Generally easy and inexpensive to get data Problem is that they may not be representative of the population you’re interested in Inexpensive, a lot of data a small amt of timeCorrelational researchCorrelation – a measure of the relationship between two variables, ex. How tall you are and the size of your shoes…taller people tend to have larger shoesRange: +1.0 to -1.0Weakest possible correlation is 1.0Variable – anything that changes or variesPositive vs. negative relationships-- + (income and edu), -(smoking and health)A. Scatterplot - a graph comprised of points generated by values of two variables. The slope of pointsdepicts the direction, and the amount of scatter indicates thestrength of relationship. - Important to visually look at your data and scatterplotsallow you to do just that Correlation and Causation- CORRELATION DOES NOT EQUAL CAUSATION- Ice cream sales go up so does assault both are relatedto the outdoor temp. When it’s hotter out people craveice cream but are also more aggressiveIllusory Correlation – *perception of a relationship where none exists(humans try to find the relationships b/t things, always looking for answers)2© 2013 WedePsych 100 – Lecture 3 Fall ‘13 - Wede Superstitions-no relationship has ever been found Stereotypes and prejudiceOrder in random events – People are always looking for patterns in our environmentGiven large number of random outcomes, a few are likely to express order.Experimentation-*backbones of psychology Like other sciences, experimentation forms the backbone of research in psychology. Allow us to isolate causes and their effects*manipulate something of interestExploring Cause and Effect Many factors influence our behavior The goal of an experiment is to manipulate variable of interest, while controlling everything elseVariablesIndependent variable (IV) –variable that’s manipulated by the experimenter Has an effect on the dependent variable The variable that is manipulated!Dependent variables –proposed to change in response to independent variable - Measured by the experimenterRandom assignment – key to experimentation Minimizes individual differences Ea. Person hasequal chance to bein any condition ofthe experimentRandom assignment vs. random Ritalin ex. Independent variable: Ritalin, placebo(control), dependent variable: positiveclassroom behaviors …compare thesetwo groups. There will be some goodand bad in each group. Mass studyingIV: type of studying distributed vs. massedDV: score of exam (memory test)…Any diff view in the score is due to the random assignment **Measuring people’s GPA clicker question: correlation does not equal causation—“studying more causes your GPA to go up” WRONG. CORRELATION DOES NOT = CAUSATION 3© 2013 WedePsych 100 – Lecture 3 Fall ‘13 - WedeDescribing Data Meaningful description of data is important in research Truck graphs…misleading information is represented with the y-axis. You can magnify the diff of data by changing the number range on a graph. Misrepresentation can lead to incorrect conclusions. 1. Measures of central tendency-best number to summarize your data with, some are better than others—depends on the shape of the distribution a. Mode-most frequently occurring score in a distribution (list of all scores) b. Mean-The arithmetic average of scores in a distribution obtained by adding the scores and then dividing by their numberc. Median-middle score in a rank –ordered distribution *WHEN THE DATA IS SKEWED THISIS THE BETTER MEASURE OF CENTRAL TENDENCY 2. Measure of variationa. Range-diff b/t highest and lowest scores in distributionb. Standard deviation-computed measure of how much scores vary around the
View Full Document