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

Behavioral Research Methods

College Level – Research methods

• Controlled Experiments to study human behavior

• Homework: Read 4 pages (click arrows) by Friday: http://allpsych.com/psychology101/intro.html

• Quiz Friday! Do practice quiz on allpsych

Quick class “experiment”

6 volunteers

• The more a person is poked while reading a passage, the less they will remember from the passage

Test hypothesis: Identify IV, Constants, DV

Problems with design

Ethical considerations

What would make this a “theory”?

Random Assignment

• Was 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 known

Research methods II

• Case study• Survey• Naturalistic (field) observation

Define eachCreate pro/con chart for each typeComplete worksheet

How could this experiment be imrpoved

• More subjects• Multiple trials• Repeat experiment for reliability• Better constants

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

WARNING: Carrots Can Kill You

• Nearly 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 variables

• The happier people are, the more likely they are to do volunteer work

• As the temperature increases, vodka sales increase• The more you exercise, the less crabby you feel• Sweaty people have more friends• Your example…?

Correlation can have a value:

• 1 is a perfect positive correlation• 0 is no correlation (the values don't seem linked

at all)• -1 is a perfect negative correlation

The value shows how good the correlation is (not how steep the line is), and if it is positive or negative.

Example – ice cream sales

• The 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 causation

• A 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 other– Both were probably caused by the heat wave– BUT! You can’t prove that the heat wave caused

anything unless you perform a controlled experiment (IV, DV etc)

Confounding variables

• A confounding variable (also ”hidden” variable) is an extraneous variable in a statistical model that correlates (positively or negatively) with both the dependent variable and the independent variable

• Teen pregnancies correlate positively with socioeconomic status and education level but you can’t assume any one causes the other

Books out. Reading homework

Chapter 7

Total college-level reading assignment for unit (you’ll read this over next month):

DUE MONDAY: Pages 246 – 254

You will be quizzed regularly on reading

Quiz Behavioral Research Methods

• Good luck!