CH.8 QUASIEXPERIMENTAL AND NONEXPERIMENTAL DESIGNS
- Randomization allows us to label single and multifactorial experiments as true
- Independent variables like gender, race, age, ethnicity are some examples that can’t be
directly manipulated so the experimenter selects participants with these characteristics or
participants who have been exposed to specified events or living in certain geographic
- When this occurs the experimenter is interested in studying the effects of these subject
variables on a dependent measure
- When exposures to events, situations or settings that emanate from the real world defined
hw participants are selected, it is called natural treatment
- The subject variables and natural treatments belong to a distinct clss of independent
variables termed quasi-independent.
- Quasi-experiment is one that investigates the effects of quasi-independent variable
on a dependent variable. Quasi meaning as if or to a degree. It resembles a true
experiment except for the degree to which an experimenter can directly control and
manipulate one or more of the independent variables.
- A true experiment would be the first option if its ethical and practical for problem. Quasi-
experiments offer a fertile design for investigating some of the most important and creative
questions in psychology.
2 types of quasi-experiment
1. Non-equivalent control group designs: it has experimental and comparison groups that
are designated before the treatment occurs and are not created by random assignment
2. Before and after designs: before and after designs have a pretest and post-test but no
comparison group. In other words, the participants exposed to the treatment serve, at
an earlier time, as their own controls.
- It might be helpful to think of non-equivalent designs and before-and-after designs as
belonging to a large and extended family of quasi-experimental research approaches. There
are 3 types of before and after designs: interrupted-time series design, multiple-group
before and after design, and repeated-measures-panel design.
Natural Treatments as Quasi-Independent Variables
- A natural treatment is created by events in the real world over which an experimenter has
no control. Ex: 911 attack; we can’t directly control or manipulate via random assignment
exposure to 9/11.
- The researcher can provide only an after-the-fact or ex post facto analysis (done
afterward) of its effect on particular dependent variable such as memory, suicide
rates, or grades.
Subject Variables as Quasi-Independent Variables - If you want to study 9/11 again but with gender you have 2 quasi-independent variables
such as a natural treatment and the other a subject variable which is gender
- Manipulated independent variable is incorporated with a non-manipulated quasi-
independent variable in some rsrch designs
- Mixed factorial design combined both between and within subject factors. One
independent variable is varied between subjects and the other independent variable
is varied within subjects
Non- Equivalent Control Group Designs
- Random assignment provides us with the best chance of ensuring that the groups would be
equivalent at the beginning of the experiment before the treatment or the manipulation of
the independent variable.
- With quasi-experiments, control and experimental groups aren’t formed by random
assignment. The groups are selected on the basis of pre-existing, immutable subject
characteristic or exposure to some kind of natural treatment
- Control group can never be considered to be equivalent to the experimental group.
- The non-equivalent control group designs have experimental and control groups that have
been predetermined or pre-designated by either an existing subject characteristic or an
already occurred natural treatment and aren’t created by random assignment
- 2 methods of selection of a control group can be used: individual matching and aggregate
- Individual matching: individual cases in the treatment group are matched with
similar individuals in the control group. In many studies it may bot be possible to match
on the most important variable
- When random assignment is not possible, the second method of matching makes more
sense: identifying a comparison group that matches the treatment group in the aggregate
rather than trying to match individual cases. This is known as aggregate matching by
finding a comparison group that has similar distributions on key variables like the
same average age, percentage female, and so on. For this to be a quasi-experimental,
individuals must not have been able to choose whether to be in treatment or control group
- Researcher matches on a variable that is highly related to the dependent variable.
Oftentimes, a rsrch can collect pretest measures and then match participants either
individually or in the aggregate on pretest scores
- Regression artifact is a phenomenon that can occur anytime a pretest measure is
used for matching. It is a threat to internal validity that occurs when subjects who are
chosen for a study because of their extreme scores on the dependent variable become
less extreme on the posttest due to natural cyclical or episodic change in the variable
- Regression to the mean: trend for extreme scores on a measure to move closer to the
group average when retested due to inherent unreliability of measurement
- Regression to the mean may be a confounding factor for interpreting results of studies that
match groups on pretest measures. Before and After Designs
- The common feature of before and after designs are the absence of a comparison group.
Because all cases are exposed to the experimental treatment, the basis for comparison is
provided by comparing the pre-treatment to the post-test measures.
- They’re useful for studies of interventions that are experienced by virtually every case in
some population such as total coverage programs like social security or studies of the effect
of a new management strategy in a single organization.
- Interrupted time series design: it is often used in quasi-experimental research to
examine observations before and after a naturally occurring treatment. In the
simplest time series design, there is a single experimental group for which we have
obtained multiple observations before and after a naturally occurring treatment.
- Such a design might be used to examine