How do psychologists develop new knowledge

Document Sample
How do psychologists develop new knowledge Powered By Docstoc
					Research Methods-PART TWO
An outline to guide student notes

I. Types of Psychological Research Methods

A. Descriptive studies
To systematically and objectively describe the characteristics of individuals or groups

B. Correlational studies
To determine and describe the strength of relationship (if any) between two or more variables (events or

C. Experiments
The deepest explanation of all methods used to determine if the manipulation of the independent variable
causes the predicted change in the dependent variable

D. Depth of Explanation

II. Descriptive Studies:
           A. Case Studies -Observe one or a very few subjects in great depth, usually over a
           long period of time
                        1. Advantage: The only method appropriate for very unusual cases
                        2. Disadvantage: Problems with generalizing the results
        B. Naturalistic Observation-Observe behavior in its natural setting, attempt to
        avoid influencing or controlling it
                        1. Advantage: Good way to collect normative data
                        2. Disadvantage: Must wait for the behavior to occur naturally
                        3. Note lecture ‘call out’ on defining conceptualization and operationalization of
        C. Surveys -Collect data from groups of people using questionnaires or
        interviews. Data is useless unless sample is representative.
                        1. Advantage: Can collect information such as attitudes and beliefs
                        2. Disadvantage: Subjects may lie or mislead
        D. Standardized Tests
                        Require people to answer a series of written or oral questions
                        Individual’s test score is totaled to yield a single score
                        This single overall score is compared to the scores of many similar people
        E. Archival Records
                        A form of research that relies on existing records of past behavior

III. Correlational Research
        A. Goal of describing the strength of the relationship between two or more          events or
characteristics (variables)
                        1. Variables -Any characteristic on which individuals differ.
                                 i. Examples: Aggressiveness, intelligence, self-esteem,.
                        2. The more highly two variables are related, the more we can
predict one from the other.
        B. Correlation Coefficient
                        1. Statistical measure of the extent to which two variables are
                        2. r varies from -1.0 to +1.0 (-.9, -.64, +.55, +. 76)
                                 a. Strength
                              Higher absolute values, indicate stronger relationships
                              (-.91 > +.66).
                              Higher scores on VAR1 go with higher scores on VAR2
                              Higher scores on VAR1 go with lower scores on VAR2
       C.. Correlation does not imply causation.
                      1.Three possible causal explanations
                        Start with 3 variables, (X, Y, & Z) where X and Y are correlated:
                        X might cause Y
                        Y might cause X
                        X might be correlated with Y, which causes Z
                       Correlations show patterns, not causes

                       Description and prediction possible
                      3. Disadvantages
                        No control over variables
                       Cannot imply causality

IV. Experimental Method
        A carefully regulated procedure in which one or more factors (independent          variables) believed to
influence the behavior being studied (dependent      variable) are manipulated and all other factors are held
       A. Determining cause by measuring effect
               1. Independent Variable (IV)
                      a. the variable being manipulated or controlled by the
                      experimenter (to see if it causes an effect)
                      b. Independent variable (IV) must have at least two levels:
                              1. Experimental or treatment condition (experimental
                              2. Control condition (control group)
               2. Dependent Variable (DV)

               3. Random Assignment - Assignment of subjects from the representative sample to
               experimental and control groups by chance
               4 . Representative Sample - (opposite of sampling bias)
                  The characteristics of the participants in the experimental and control
                  groups corresponds closely to the characteristics of the larger
                  population being studied

       B. Experimental Research Design Examples
               A. Example 1
               B. Example 2
               C. Summary Note
                 1. A hypothesis is made (experimental hypothesis = a prediction)
                 2. The population is identified
                 3. A representative sample is selected
                 4. The representative sample is randomly assigned to the experimental
condition or the control condition.
               5. The investigator varies some factors (independent variable)
              Experimental groups get the different values of the independent variables. Control groups do not
               6. The investigator keeps some factors constant
               7. The investigator measures the effects on the dependent variable.
               8. Dependent variables supposedly affected by independent variables.
               9. Statistical Significance established
              10. Conclusions about causality are possible due to the controlled methods.
              11. Replication/repeating research strategies to confirm original findings are possible.

       C. Cautions : Bias
         Bias can affect the way an experimenter designs a study, collects data, or interprets results
              1. Types of observer bias:
                       a. . Personal bias
                               i. . The influence of the experimenter’s own beliefs, preferences assumptions,
                               prejudices, on the outcome of the research
                               ii. Often not obvious to individual
                               iii.. May notice only evidence confirming their hypothesis
                               and ignore contradictory evidence
                       .b. Expectancy bias
                       The influence of one’s expectations on one’s observations and
                       looking for certain outcomes.
              2.. Participant bias
                       Placebo effect
                       Occurs when participants’ expectations, rather than the
              experimental treatment, produce the desired outcome
                       Placebo bias – When a subject in an experiment receives no
              treatment but still believes that the treatment is working
                       Placebo – Substances or events that appear to be the real thing
              (e.g. drugs) but are not. Placebos cause a placebo effect in some people that experience them.

              3. Controlling Bias
                     a. Double blind study
                       A study in which neither the subjects nor the experimenters know which subjects are in
                       the experimental group and which are in the control group until the results are
                     b. Single blind study
                        Keeping subjects uninformed about the purpose of a study so they are less likely to
                       perform to the researchers expectations.
                     c. Deception and debriefing
                       Provide subjects with a full and honest account of the true purposes and assumptions
                     of the research study as required by American Psychological Association when
                     deception is used

D. Confounding Variables
             1. Researchers must also attempt to control extraneous or confounding
             2. Any variable(s) other than the independent variable that seem likely to
      influence the dependent variable in a specific study.
             3. Use of random assignment helps to avoid extraneous variables
E. What is the Scientific Method?
•      The scientific method is an approach to knowledge that relies on a systematic method of asking
questions, generating hypothesis, collecting data, and interpreting the findings.

        Scientific method –
A five-step process for empirical investigation of a hypothesis under conditions designed to control biases and
subjective judgments
The First Step of the Scientific Method is______________________________________
       Define: Hypothesis
                Operational Definition
The Second Step of the Scientific Method is____________________________________
       Define: Independent Variable
                Experimental Group
               Control Group
The Third Step of the Scientific Method is _____________________________________
       Define; Data
               Dependent Variable
The Fourth Step of the Scientific Method is ____________________________________

The Fifth Step of the Scientific Method is _____________________________________

F. Thinking Critically About Behavior
   •Making sense of behavior claims
   •Find out how important terms are defined and measured
   •Understand the nature of the sample before you generalize
   •Do not predict individual behavior from group results
   •Resist interpreting the results of a single study as definitive
   •Differentiate correlation research from experimental research
   •Evaluate the credibility of the source
   •Be skeptical of simplistic claims
   •A single study usually is not the defining word about an issue or problem

Shared By: