Descriptive statistics
Descriptive statistics allow researchers to make precise statements about the data
o Two statistics are needed to describe the data
o One number can be used to describe the central tendency or how participants
scored overall while
In chapter 8 we looked at classic experimental designed in which participants are
assigned to experimental controls and a variable is manipulated
This process has high internal validity
This chapter will look at 3 types of special research circumstances
Regression Toward the Mean
Aka statistical regression, regression toward the mean is likely to occur when
participants are selected because they score extremely high or extremely low on a
particular variable. When they are tested a 2nd time, their result
Quasi-Experimental Design
6 quasi-experimental designs will be explored
While reading, compare the design features and problems with the randomized true
experimental designs
The first three described are sometimes called pre-experimental designs to dis
Multiple Baseline Designs
the reversal of some behaviours is unethical or impossible (eg is it unethical to reverse
treatments that reduce dangerous behaviour or a treatment may provide a long-lasting
effect that is not reversible)
in such cases multiple
Control Series Design
one way to improve the interrupted time series design is to find a control a control
series design.
in the case of passing the driving law, a control was possible because other states had
not implemented this law, therefore could b
Sequential Method
a compromise between cross-sectional and longitudinal study
this method, along with cross-sectional and longitudinal, are illustrated in figure
11.6 on page 219*
in the figure, the goal is to minimally compare 55 and 65 year olds.
Th
Sampling distributions
You can infer using intuition that getting 7/10 answers vs. 2/10 answers correct on the
ESP experiment is unlikely
Look at table 13.1 on page 251
o The probabilities shown were derived from a probability distribution called the
bi
Structural models
Structural model: an expected pattern of relationships among a set of variables
o The pattern is based on a theory of how the variables are causally related to one
another
This research approach is called structural modeling or structu
Effect size
Effect size: a general term that refers to the strength of association between variables
o The pearson r correlation coefficient is one indicator of effect size
An effect size correlation coefficient can be calculated for experiments like th