Quasi experiments are studies that are unable to use randomization to evaluate e
ID: 3273626 • Letter: Q
Question
Quasi experiments are studies that are unable to use randomization to evaluate effectiveness of interventions. This can make it difficult to tease out possible confounders and assess the integrity of any cause and effect claims. Consider this example adapted from a paper in Clinical Infectious Diseases. ^1 A hospital wants to know if providing alcohol-based hand cleaners for staff will reduce the rate at which bacterial infections occur in the patient population. The hospital needs to design a study that will address this question. Suppose a researcher suggests to place these cleaners in randomly chosen patient rooms and not in others. At the end of the study, the infection rates for each group could be compared. Why might this be an impossible design to implement in practice? Give one reason and explain. The hospital is more likely to adopt a quasi-design that collects data for an extended period of time before hand cleaner dispensers are installed, and then compares those findings with data collected after the dispensers are installed. State and explain two possible sources of confounding that might occur with this kind of design. In the absence of random assignment, arguments addressing confounding often have to be made some other way. What are two arguments that one might be able to make to support results obtained from a quasi-design like the one in Question 2?Explanation / Answer
1. Randomly choosing patient room and placing alcohol based handcleaners for staff is not a viable design, as hospital staff are supposed to enter every patient room, and use the handcleaner wherever available. They can even carry their own handcleaner and use it as and when required. Thus, the very essence of the experiment, that is to determine whether alcohol based hand cleaner reduces the rate of bacteril ainfection cannot be determined.
2. Tw possible sources of counfounding might be:
a> There might be many people who maintain strict hygiene, and were regularly using own hand cleaner before the dispenser were launched. This would affect the internal validity of the experiment.
b> People might be conscious enough to not keep with the bacterial infection if detected during the tenure of the experiment, and might get cured. This too would affect the internal validity of the experimet.
In above two case, the sources of confounding are (rather cannot be controlled) by the reasercher.
3. Since, time cannot be reversed, that is no one can go at the begining of the situation where hand dispenser were not installed and start the experiment from there, one has to accept the way the experimenter has designed the quasi experiment. Another important argument is absence of randomization. Since, the patients room cannot be randomly chosen for placing the hand dsipenser, hence it is suitable to make the decision from quasi experiment.