PSYCH 100 1st Edition Lecture 4Archival Research- Use other peoples data instead of your owno Hospital records, police records, autobiographies Research Design: Descriptive (Studies that are non-experimental)- Correlational approacho Researcher does not control variables or random assignment- Correlationo Relationship between two naturally occurring variableso Positive, negative, or no correlation, computed by correlation coefficiento Variable: Any entity that can take on different values Does not have to be quantitative Ex: Gender, age, country- Weakness- does not prove causation- Correlation Coefficient (What makes them descriptive)o Helps to determine how much the variables vary togethero Pearson’s “r” o -1 to +1o The sign (+ and -) tells directiono The # tells the strength of the relationshipo There are different values of correlation coefficients- Correlation does Not = Causation- Directionalityo Does A cause B?o Does B cause A?- There is also a third variable to considerResearch Design: Experimental Research- Experimento Researcher controls procedureso Participants are randomly assignedo Allows for statements of cause and effecto Manipulate the variable of interesto Hold other factors constant- Quasi-experimento No random assignments- Variableso Independent- what is manipulatedo Dependent- what is measured o Confounds- what can go wrong in experiment - What’s involved in experimentso Control group o Experimental group- Typeso “Blind”o Double-blind Participant doesn’t know and researcher doesn’t knowo Placebo effect Using a real variable and a fake variable - Random Assignmento “The great equalizer”- Randomly assigning groups to get rid of confounding variables o Define characteristic of an experiment Its what gives you controlo Establishes equivalent groups BEFORE manipulation- Random assignment vs. Random samplingo Random sampling gives the researcher the sample of populationo Random assignment determines what we do with the sample Who goes where Meta-Analysis:- Define: Statistical analysis of many previous experiments on a topic- Studies that have already been done- They are done to show patternso Psych info.Studying Across Time ( Observations )- Cross-Sectional (At one point in time)o Normally have something in common Ex: Facebook datao Looking for age, gender, etc. and explore differenceso Something specific- Longitudinalo Studying the same group of people at a point in timeo Really high drop out/cost rates- Mixed Longitudinal o Shorter, about 5 yearso Mixture of cross-sectional and
View Full Document