Behavioral Research Methods. College Level – Research methods Controlled Experiments to study human behavior Homework: Read 4 pages (click arrows) by

Embed Size (px)

Text of Behavioral Research Methods. College Level – Research methods Controlled Experiments to study...

Behavioral Research Methods

Behavioral Research MethodsCollege Level Research methodsControlled Experiments to study human behaviorHomework: Read 4 pages (click arrows) by Friday: http://allpsych.com/psychology101/intro.htmlQuiz Friday! Do practice quiz on allpsych

Quick class experiment

Copy text pages X35Random AssignmentWas there any in this experiment?Double or single blind?Neither. Both the subjects and the researcher (me) were aware of which treatment group they were in.

Single blind the researcher knows who is getting what treatment, but not the subjects.

Double blind neither researcher nor subject knows who is getting what treatment ex: all subjects are administered a pill, but dosage is not knownResearch methods IICase studySurveyNaturalistic (field) observation

Define eachCreate pro/con chart for each typeComplete worksheetHow could this experiment be imrpovedMore subjectsMultiple trialsRepeat experiment for reliabilityBetter constants

Random assignment ensures subjects are not grouped in a way that will affect outcome of study (no bias introduced)

WARNING: Carrots Can Kill YouNearly all sick people have eaten carrots. An estimated 99.9% of all people who die from cancer and heart disease have eaten carrots.99.9% of people involved in car accidents ate carrots within 60 days of their accidents.93.1% of juvenile delinquents come from homes where carrots are served regularly.Carrot eaters born between 1900 & 1920 have been noted to have wrinkled skin, brittle bones and failing eyesight.Among the people born in 1839 who later dined on carrots, there has been a 100% mortality

Correlation a relationship between two variablesThe happier people are, the more likely they are to do volunteer work As the temperature increases, vodka sales increaseThe more you exercise, the less crabby you feelSweaty people have more friendsYour example?

Correlation can have a value:

1is a perfect positive correlation0is no correlation (the values don't seem linked at all)-1is a perfect negative correlation

The value showshow good the correlation is(not how steep the line is), and if it is positive or negative.Example ice cream salesThe local ice cream shop keeps track of how much ice cream they sell versus the temperature on a given day. http://www.mathsisfun.com/data/correlation.html

Correlation is NOT causationA relationship between 2 variables does not show cause and effect

Ice cream sales and sunglass sales both increased in the same way, but one did not cause the otherBoth were probably caused by the heat waveBUT! You cant prove that the heat wave caused anything unless you perform a controlled experiment (IV, DV etc)Confounding variablesAconfounding variable(alsohidden variable) is anextraneous variablein astatistical modelthat correlates(positively or negatively) with both thedependent variableand theindependent variableTeen pregnancies correlate positively with socioeconomic status and education level but you cant assume any one causes the other

Books out. Reading homeworkChapter 7Total college-level reading assignment for unit (youll read this over next month): DUE MONDAY: Pages 246 254You will be quizzed regularly on reading

Quiz Behavioral Research MethodsGood luck!