1. Scientific Methodology in Psychology

2. The Scientific Method
rules for conducting research
prescribed methodology for the collection, analysis and interpretation of data

3. Why use the scientific method?
reduces chances for error
provides procedures for repeatability and subsequently, reliability
self correcting

5. Basic Steps for Conducting a Scientific Experiment
Observe and Identify the Problem
(ask a question)
Formulate an Hypothesis
this is a statement of your question and a guess about the answer
Design the Experiment
operationally define variables, select and assign subjects,

6. Conduct the Experiment (collect data)
Analyze and Interpret Data
utilize statistical analysis to determine significance
*Report Findings (publish data)
*Why is this last step so important?

7. Consider the following example:

8. Observations
observations of school children and surveys concerning preferred breakfasts indicate that children eating sugar sweetened cereals performed better on standardized tests throughout the day
observations of animals in learning tasks show that those eating Froot Loop cereal learned to solve the maze in less time than those receiving non-sweetened food reinforcers

9. Hypothesis Consuming sugar enhances memory
Where did this hypothesis come from?

10. Observation in animal experiments and humans in natural environment
Logical progression of information from other fields or other areas of the same field

11. Design and Conduct Experiment
Dependent Variable (DV) = that which the experimenter measures
Independent Variable (IV) = that which the experimenter manipulates
Operational Definition (OD)= the exact procedure used to measure an intangible variable

12. Must be randomly selected and randomly assigned to either:
Subjects:
Must be randomly selected and randomly assigned to either:
Experimental Group (E)
those subjects who receive the treatment
Control Group (C)
those subjects who receive a placebo
What does "random" mean as used here?

13. Which group receives sugar? Which group receives saccharine?
We want to give some students sugar and others a placebo (saccharine) and test their ability to recall a list of nonsense syllables.
What is our DV?
What is the OD of our DV?
What is our IV?
How many levels of the IV are there?
Which group receives sugar?
Which group receives saccharine?

14. Controlling Error Extraneous Variables
conditions which might affect the outcome but can be controlled by the experimenter.
What are some possible extraneous variables in our experiment?
Why did we have the control group consume saccharine? Why not have them consume nothing at all?

15. confounding variable
uncontrolled extraneous variables which render results invalid
what are some possible confounding variables in each of our experiments?

16. Experimental Design Single Blind Experiment
The subjects are unaware of the group to which they belong
Can control
subject bias
self-fulfilling prophecy

17. Double Blind Experiment
Neither the subjects nor the experimenter know to which group the subjects belong
Can control
experimenter effects

18. Results:
Fifty students were randomly assigned to the either the experimental (sugar sweetened water) or control (saccharine sweetened water)group. After consuming 8 oz. of the drink, the students were given a list of 50 nonsense syllables to memorize for 1 hour.
They were then given a math task to control for rehearsal effects. The number of nonsense syllables each recalled after 1 hour was measured. Results were as follows:
Experimental Mean # recalled = 33
Control Mean # recalled = 24

20. What can we conclude from our data?
Note: although it appears to support our hypothesis, we must conduct statistical analyses to determine if effect is true or occurred just by chance alone

22. True Experiment
Example represents true experiment because IV is controlled and subjects are randomly assigned
Allows us to draw causal conclusions.

23. Inferential vs Descriptive Statistics
inferential statistics
use sample data to draw conclusions about the population
experimental designs allow us to draw causal conclusions about the population
descriptive statistics
numerical values that summarize sample data

24. Descriptive Statistics
Measures of Central Tendency
Mean (the average score)
Median (the middle score)
Mode (the most frequently occurring score)
Skewed distributions
Negative/Positive
Measuring Variability
Standard Deviation

25. An Observational Study
How long do people wash their hands after using the bathroom?
You observe approx. 50 subjects over the course of 3 days
You time each subject’s hand washing
You report your findings

27. So would you say that knowing the average gives you sufficient information on which to make a decision?
e.g. The average number of hours of study necessary to get an A is 3 per week
e.g. On average, you have to “crash” the toll booth 5 times before you’ll get a ticket

28. Positive Skew Mean is higher than the median

29. Negative Skew
Mean is lower than the median

30. Correlational Studies

31. Correlational Studies
provide information about the relationship between two variables
can be used to make predictions
cannot be used to draw causal conclusions

32. Why conduct correlational studies? Why not conduct experiments?

33. Consider the following examples:
relationship between
IQ and income
race and traffic tickets
weight and incidence of diabetes
blood cholesterol levels heart attacks
cigarettes smoked and incidence of lung cancer
note that these are variables which cannot be randomly assigned for practical or ethical purposes

34. Correlations may be: Direct or Positive Correlation
As A increases, B increases
Indirect or Negative Correlation
As A increases, B decreases

35. negative
positive B B A A
no correlation B A

36. Negative Correlation
Positive Correlation
No Correlation

37. Correlation Coefficient
indicates degree and direction of relationship
most common is the Pearson r
values range from
cannot be < -1
cannot be > +1
0 = no correlation

39. Which provides more information, a negative correlation or a positive correlation?
What does a correlation coefficient of indicate?
Why can’t we draw causal conclusions from such data?

40. Great care must be taken when considering the applicability of correlational studies to the general population.
errors in sampling can lead to incorrect conclusions:
Sigmund Freud
1948 Presidential Election
XYY males and violent behavior

41. "Other Types of Research"

42. Single Subject Design can be correlational or experimental
case studies
most common type
observation and testing of an individual with unusual characteristics such as disease or injury
eg. Phineas Gage
eg. Patient HM

43. Phineas Gage

44. Pt. H.M.

45. Naturalistic Observation
observations of humans or other animals in their natural environment
observer effects problem
people change their behavior when they know they're being watched
Dr. Jane Goodall