Quantitative designs include experimental, non-experimental and quasi-experimental.
This examines the cause and outcome relationship between the independent and dependent variables in conditions under high control. According to Polit and Beck (2009), an outcome has to be preceded by a cause without any influencing variable so as to confidently conclude the existence of a cause-outcome relationship. With this design, manipulation could not be possible on all variables. The most classic of these designs according to Cannon (2011) is the pre-test/post-test design where subjects would be assigned to a control or experimental group which does not receive and receives treatment respectively. Other designs include posttest only and factorial designs.
This design has been highly regarded due to its strength in demonstrating cause outcome relationship between independent and dependent variables. It gives highest quality evidence concerning intervention effects. Randomization and applications of control group conditions yield almost ideal counterfactual. It offers the greatest corroboration. Even so, it has limitations including the inability of manipulating some variables. The design suffers from ethical constraints when experimenting on humans. It could fail to be feasible as it would be impractical in many healthcare settings. It suffers Hawthorne effect, described by Polit and Beck (2009) as the tendency to change behavior by the people having the knowledge that they are being studied.
The non-experimental design would be applied in situations where manipulation of independent variables could be difficult and randomization uncontrolled but the outcome of some conditions occurring naturally could be studied. Studies adopting this design would be more observational than international.