Title: Research Strategies and Validity
1Chapter 6
- Research Strategies and Validity
2Research Strategies
- Descriptive Strategies
- Correlational Strategies
- Experimental Strategies
- Quasi-Experimental Strategies
3Descriptive Strategy
Descriptive strategy simply describes variables
as they exist.
4Correlational Strategy
The correlational strategy looks at the nature of
the relationship between two or more variables.
It does not attempt to determine the cause of
that relationship.
5Experimental Strategy
The experimental strategy attempts to discover
the cause of a relationship between two (or more)
variables by manipulating independent variables
and observing the consequence on the dependent
variables.
6A true experiment will have the following
components
- Manipulation
- Measurement
- Comparison
- Control
7Quasi-Experimental Strategy
The quasi-experimental strategy compares
pre-existing groups on one or more dependent
variables. Although it takes the form of an
experiment, it actually is a type of
correlational research and suffers from the same
limitations.
8Research Validity
Valid research is free from methodological flaws
that would limit the researchers ability to draw
clear conclusions and that would limit the
researchers ability to generalize the results to
other groups or situations.
9Two types of validity
- Internal validity Whether there are
methodological flaws that would limit the
researchers ability to draw a clear conclusion. - External validity Whether there are
methodological flaws that would limit the ability
of the researchers ability to generalize the
results to the population of interest.
10Threats to internal validity
- Extraneous variables variables in a research
study other than the independent and dependent
variables. - Confounding variables An extraneous variable
that changes systematically with the independent
and dependent variables.
11Classes of extraneous variables
- Participant variables Participants differ with
regard to many characteristics. - Environmental variables The environmental
conditions under which a participant is tested
may vary. - Measurement variables The process of taking
repeated measurements from a participant can
introduce change.
12Threats to external validity
- Non-representative samples
- Selection bias
- Volunteer bias
- Species studied
- Unique features of the study
- Laboratory environments
- Participant reactivity
- Demand characteristics
13Threats to external validity contd.
- Researcher characteristics
- Measurements obtained
- Different measurements
- Measurements taken at different times
14The relationship between internal and external
validity
There is a tradeoff between internal and external
validity. Often times the effort to increase
internal validity results in a loss of external
validity and visa versa.
15Internal and external validity of the different
research strategies
- Descriptive research External validity tends to
be high, but internal validity tends to be low - Correlational research External validity tends
to be high, but internal validity tends to be low - Experimental research External validity tends
to be low, but internal validity tends to be high - Quasi-experimental research External validity
tends to be high, but internal validity tends to
be low
16Distinguishing among research strategies,
research designs, and research procedures
- Research strategies General approach determined
by purpose (e.g., correlational, experimental) - Research designs A general plan for
implementing the research strategy. - Single subject or group subject designs
- Within subject or between subject designs
- The number of variables to be manipulated
- Research procedures Specific details of how the
study will be conducted