8. Descriptive Research
A friend of yours is working part-time in a local pet
store. Your friend has become increasingly interested in
how dogs normally communicate and interact with
each other, and is thinking of visiting a local veterinary
clinic to see how dogs interact in the waiting room.
Is this a good research design? Why or why not?Can
you suggest a better method to study this research
question?
10. CORRELATIONAL RESEARCH
1. DESCRIPTIVE:
a. No hypothesis
b. Describe behaviors that are observed or measured
2. CORRELATIONAL:
a. Hypothesizes & tests for specific relationships
b. Two or more measured variables
11. CORRELATIONAL RESEARCH
CORRELATION: a measure of the degree of association
between two variables
CORRELATION COEFFICIENT (r): a number that expresses the
relationship between two variables
– Strength of the relationship
– Type of relationship: positive or negative
-1 +1
0
Strong
correlation
Strong
correlation
Zero
correlation
17. CORRELATIONAL RESEARCH
Correlation DOES NOT imply causation
• We can say that a relationship exists.
• We can’t say that one variable causes changes in the
other.
THIRD VARIABLE PROBLEM
18. Positive or Negative Correlation?
1. Amount of clothing worn & temperature outside
2. Cigarettes smoked & risk of lung cancer
3. MPH & amount of time it takes to arrive at your
destination
4. Income & education
5. Hours spent studying & grade in course
6. Amount of time spent playing World of Warcraft
& number of dates
7. Number of books in the home & child IQ
19. Consider the following:
Scott has a theory that the amount of caffeine he drinks
makes him a more productive writer. For one month,
he keeps track of how much caffeine he drinks and
how many words he writes each day. At the end of the
month, he calculates the correlation coefficient and
the result is +.89. What pieces of information can he
learn from this number?
Pleased with himself, Scott goes around and brags to all
of his friends that drinking caffeine causes him to write
more. Is his conclusion correct? Why or why not?
22. EXPERIMENTAL RESEARCH
1. DESCRIPTIVE:
a. No hypothesis
b. Describe behaviors that are observed or measured
2. CORRELATIONAL:
a. Hypothesizes & tests for specific relationships
b. Two or more measured variables
3. EXPERIMENTAL:
a. Hypothesizes & tests a causal hypothesis
b. Manipulated & measured variables
c. Random assignment to groups
23. EXPERIMENTAL RESEARCH
EXPERIMENTAL RESEARCH: manipulate one
variable to look for an effect on another
• INDEPENDENT VARIABLE (IV): variable that is
manipulated→ THE CAUSE
– EXPERIMENTAL GROUP: subjects who
receive some special treatment
– CONTROL GROUP: subjects who do not
receive the special treatment
24. EXPERIMENTAL RESEARCH
• DEPENDENT VARIABLE (DV): what is
measured → THE EFFECT
• OPERATIONAL DEFINITION: defining a variable
by how it will be manipulated or measured
• CONFOUNDING VARIABLE: anything that
could affect the DV besides the IV
29. EXPERIMENTAL RESEARCH
An experimenter wanted to know whether exposure to bright
light during the winter months would lead to improved
mood among people with Seasonal Affective Disorder. Half
of the participants were assigned to a condition in which
they were asked to sit in front of a special lamp 1 hour a
day. The other half of the participants were exposed to
normal levels of light. The experimenter measured mood
by administering a questionnaire.
Identify the following: IV, DV, Experimental Group, Control
Group
32. BIASES
• SOCIAL DESIRABILITY: subjects present
themselves in a way that is more positive than
reality
• Attitudes may not match behaviors
33. BIASES
• PLACEBO EFFECT: behavior changes in the
absence of any manipulation
• EXPECTANCY EFFECT: a researcher’s
expectation causes them to unconsciously
influence the results of a study
34.
35. BIASES
• SINGLE BLIND STUDIES: participant does not know
hypotheses
• DOUBLE BLIND STUDIES: neither participant nor
experimenter know hypotheses
• RANDOM ASSIGNMENT
• REPLICATION
There is no substitute for
good experimental design.
36. SUMMARY: GOALS OF RESEARCH
• DESCRIPTIVE: Describe the quality or quantity
of a behavior.
• CORRELATIONAL: Predict behavior. Determine
the relationship between two variables.
• EXPERIMENTAL: Explain behavior. Establish
cause and effect between two variables.
37. SUMMARY: IDENTIFYING
METHODS
Is the study looking at the relationship between 2 or more
variables?
NO
Descriptive
YES
Have people been randomly assigned?
NO YES
Correlational Experimental