In: Statistics and Probability
give an Explanation of the following comparison through some examples:
Quasi-experimental and true experimental research designs both attempt to determine causal relationships by applying a treatment or condition to one group and comparing the outcome with a control group. A true experimental design is the gold standard in assessing causal relationship because it requires that subjects be randomly assigned to the groups to avoid bias and it controls all extraneous variables
In order to understand the difference , We need to understand the basic definition of the two types of experiments firstly.
A true experimental design is a design that includes the manipulation/change of the various independent variables and then comparison of groups in randomized fashion.All the key factors are controlled in a true experimental design. The subjects are randomly assigned to control group and the treatment group.
whereas
A Quasi-experimental design is a design that involve manipulation/change of the various independent variables and then comparison of groups ( control and treatment ) but which does not have the element of randomization assigned to the subject/samples.
Hence we can easily observe from the definition of the two that both a true experiment and a quasi-experiment carry comparison of experimental group and a control group.
but
a true experiment has probability samples(randomized) and a quasi-experiment involves a non-probability sample( pre assigned).
Let us try to understand the difference between the two with the help of an example.
Suppose we visit a society and observe the athletic ability of students . While observing so we also note if the parents of the students provide them with meat or not ? And in order to study the effect we choose a random sample of 30 families from the society given. Now This can appear to be a true experiment wherein we are trying to find the athletic performance of a student based on the meat intake taken by him/her.
But now think !
There can be many families who dont take meat due to their rleiious beleives.
Now Think what will happen when we identify these families before the start of experiment itself and we choose the families for our experiment form the set which do ont include these families.
Clearly all families even if when selected randomly from the new set ,will fail to capture the impact of meat eating on athletic performance since all families in this set provide meat to their kids.
This will be a quasi experimental design.