Strengths and Weaknesses of Quasi-Experimental Designs - Assignment Example

The wide applicability of quasi-experimental design in real-life situations makes it possible to study a phenomenon that cannot be possibly studied under experimental design due to the inability to exercise pure control on research variables. The design is also able to apply selective control on the investigated variables in circumstances where experimental design cannot. The design also has strength in its ability to counter the emerging trend of resistance to randomization. This is particularly applicable when people are involved as participants with consideration of the principle of free consent.

While participants would resist randomization into a control group, the quasi-experimental design does not involve randomization and is therefore easily acceptable to people. Consequently, sample generation and sampling are easier in a quasi-experimental design (Polit and Beck, 2007). Further, the design allows some level of control in the research process that may be used to control expected threats. A researcher can for example control when to collect data to allow room for sufficient treatment effect (Evans and Keenan, 2009).

The design, however, has a number of weaknesses. It for example considers many sets of hypotheses that must be investigated by the researcher. This, therefore, makes it more hectic and probably time-consuming than other designs. This is because other possibilities must be explored to consider the effects of a lack of control. Another significant weakness of the quasi-experimental design, which originates from its lack of a control set up, is its susceptibility to confounds that a researcher cannot explain or account for.

In the presence of a control set up, a researcher can identify variations in the control group to make appropriate adjustments to the treatment group and therefore account for the possible confounds. This, therefore, means that the observed results in the quasi-experimental designs are possibly influenced by confounds. As a result, there is a loss of confidence over the validity and reliability of results from quasi-experimental designs (Polit and Beck, 2007).