LECTURE 3 Cross-Sectional Design:
- Different age/cohort
groups compared at the same time
- Measures age
differences, not age changes
- Age effects and
cohort effects are completely confounded
Longitudinal Design:
- One cohort is studied
repeatedly over a period of months or years
- Measures age changes
- Age effects and time
of measurement effects are completely confounded
Sequential Design:
Combines cross-sectional and
longitudinal methods; untangles effects of age, cohort and time
Experimental
Method:
Used to determine cause and effect
relationships, i.e., to establish unambiguously that one thing causes
another
Experiment:
Isolate a particular variable
hypothesized to affect behavior or development, and then manipulate its
influence within part of the sample (the experimental group), while
simultaneously controlling for or holding constant all other variables
Independent Variable
The variable that is manipulated by
the experimentor (the causal variable)
Dependent Variable:
The aspect of behavior or
development that is presumed to be affected by the various levels of the
independent variable
Random Assignment
Placing research participants in
experimental conditions in an unbiased way so that the resulting groups are
not systematically different from one another
Limitation of the
Experimental Method
Some questions cannot ethically be
studied experimentally
Correlational Method
- Used to determine
whether two (or more) variables are related
- But can NOT determine
cause and effect
A Correlation Shows:
- The direction
and strength of relationship between two variables (that is, the
degree to which two variables are related)
- Direction:
Positive (+) or Negative (-)
0.00-1.00
Positive correlations:
Can always be expressed as
"the more ... the more," or "the less ... the less"
Examples of Positive Correlations:
- The more TV the
child watches the more aggressive the child
- The more
"reality-based" shows (e.g., America's Most Wanted or Cops)
children watch the more behavioral problems
- The more TV children
watch the more obese they are
- The more years of
schooling the higher the IQ
- The fewer years of
schooling the lower the IQ
Negative Correlations:
Can always be expressed as
"the more...the less," or "the less... the more"
Examples of Negative Correlations:
- The more TV the child
watches the less self-control the child shows
- The more TV the child
watches the lower the grades in school
- The fewer alcoholic
drinks consumed per week the higher the GPA
Limitation of the Correlational
Method
- Although it can show
that two variables are related, it cannot unambiguously establish a causal
relationship between them
- Variable A may cause
B, but it is entirely possible that the cause-effect relationship is
reversed: B may cause A!
- Or, the association
between the two variables may be due to some third variable
Reliability
Repeatability; the extent to which
a research measure yields consistent information from occasion to occasion or
between different observers
Validity
- How accurately a
test measures what it is supposed to measure
- More generally, the
extent to which a research measure actually measures what it is supposed
to measure rather than something else
END
|