Transcript April 21

BHS 204-01
Methods in Behavioral
Sciences I
April 21, 2003
Chapter 4 & 5 (Stanovich)
Demonstrating Causation
Figure 4.5. (p. 93)
Two different distributions with the same range and mean but different
dispersions of scores.
Standard Deviation


Average distance of scores from the mean.
Calculated by taking the square root of the
variance.


The variance scores were squared so that the
average of positive and negative distances from
the mean could be combined.
Taking the square root reverses this squaring and
gives us a number expressed in our original units
of measurement (instead of squared units).
Graphing Data

Line graph – used for ordinal, interval, ratio
data.



Independent variable on the x-axis
Dependent variable on the y-axis
Bar graph – used for categorical data.
Figure 4.6. (p. 97)
Effects of room temperature on response rates in rats.
Figure 4.7. (p. 97)
Effects of different forms of therapy.
Transforming Data

Sometimes it is useful to change the form of
the data in some way:




Converting F to C temperatures.
Converting inches to centimeters.
Transformation let you compare results across
studies.
Transformation must preserve the meaning of
the data set and the relationships within it.
Standard Scores

One way to transform data in order to
compare two data sets is to express all scores
in terms of the distance from the mean.



This is called a z-score.
z = (score – mean) / standard deviation
z-scores can be transformed so that all scores
are positive:


This is called a T-score
T = 10 x z + 50
Measures of Association

Scatter plot – used to show how two
dependent variables vary in relation to each
other.


One variable on x-axis, the other on y-axis.
Correlation – a statistics that describes the
relationship between two variables – how they
vary together.

Correlations range from -1 to 1.
Figure 4.9. (p. 102)
Scatter diagram showing
negative relationship
between two measures.
Figure 4.10. (p. 103)
Scatter diagrams showing various relationships that differ in degree and
direction.
The Problem with Testimonials



The Placebo effect
The “vividness” problem.
The P.T. Barnum effect.
Correlation and Causation



The “third variable” problem.
The directionality problem.
Selection bias.