Chapter 2: Research Methodology


Scientific Method

            Hypothesis/Formulate a question

            Method/Develop a way to test the question





Quantitative Method (Positivist Social Science)

Qualitative Method  (Interpretive Social Science)



General Principles of Conducting Psychological research

            Operational Definition

            Samples of the Population (4)


            Objective Measurement

            Experimenter Bias/Blind Studies

            Blind Observer


            Multiple Measures (concept)


Observational or Descriptive Research Designs

            Naturalistic Observation

            Survey Method

            Case Histories


            Correlation Coefficient (r)


Rules for Correlation


1.      Positive or Negative = Direction of effect/relationship

2.      Range (one to negative one) = Size of effect/relationship

3.      r = 0, then there is NO relationship or the relationship is random

4.      Correlation DOES NOT EQUAL causation


Correlations:  Positive or Negative?


For each of the correlations described below, determine the direction (positive or negative).


1.  A study of married couples showed that the longer they had been married the more similar their opinions on social and political issues were.


2.  An intelligence test was given to all the children in an orphanage.  The results showed that the longer the children had lived in the orphanage, the lower their IQ scores.


3.  In a study of American cities, it was found that as the number of violent crimes increase so did the number of stores selling violence-depicting pornography.


4.  A college professor found that the more class absences students have, the lower their grade in the course tends to be.


5.  A politician running against a candidate who has been in office for eight years pointed out that violent crime had increased steadily during those eight years while the administration appropriated more and more money to fight crime.


6.  It was found that elementary school children scored highly on a vocabulary test also tended to score highly on a test of physical strength and muscular coordination.


7.  In golf, the better one’s ability, the lower one’s score.




Independent Variables and Dependent Variables

            A condition is a level of the independent variable. 

            Experimental Group, Control Group, and Random Assignment


How experiments can go wrong:  Demand Characteristics


Ethical Considerations in Experimentation



The key question to be answered here is when is the researcher justified in risking harm (physical or otherwise) for the sake of knowledge?



Measuring and Analyzing Results


Descriptive Statistics: Descriptive Statistics are “Mathematical summaries of results.”


1.  Measurements of the Central Score:  Mean, Median, and Mode.

            a.  the mean is the “sum of all the scores divided by the total number of scores.”  It is the most useful measure, especially if the scores approximate a normal distribution.


            b.  A normal distribution is “a symmetrical frequency of scores clustered around the mean."” The mean can be misleading if the distribution is not normal. 


            c.  The median is the middle score of a distribution, and is affected less by extreme scores.


            d.  The mode is “the score that occurs most frequently.”


2.  Measures of variation: To fully capture a distribution, we also need some measurement of spread around the mean. 


            a.  One such measure, the range, is simply  “a statement of the high and low scores.”  It is not very useful because it only reflects the extremes. 


            b.  The standard deviation is “a measurement of the amount of variation among scores in a normal distribution”,  and provide a useful means of comparing scores on two different tests (which tests **).


Evaluating Results:  Inferential Statistics


1.  The 95% confidence interval is “the range within which it is 95% certain that the true population lies. “  It is an example of an inferential statistic, “a statement about a large group based on an inference from a small sample.”  The usual convention is that the probability that a result occurred due to chance is less that 5%, that the result is considered “statistically significant.”