Final evaluation and research design

Quantitative Design

Agiolela’i Penitusi

Walden University

MMPP 6281-1

Quantitative Design

Quantitative design when conducting a research is a standard method of experiment most used in scientific disciplines. In some cases, these experiments get referred as true science; they use statistical and traditional mathematical means in order to measure results in a conclusive manner.

Field experiments are types of quantitative design. Field experiment is experiment done in the real world or outside a laboratory. In other words, field experiments take place in real-life setting, for example work place, classroom or a street. In these types of experiments, subjects get randomized into treatment and control groups where the results of the two groups are compared.

When selecting treatment and control groups one should select two groups of people. People in these groups should be representative of a similar population. For example, select two groups of people presenting similar symptoms of a particular illness. One group should be the control group and the other group should be the group in which the treatment is applied in full. Members in the control group receive standard treatment, while those in the complementary group receive full treatment. By doing this one will be able to tell the effectiveness of the treatment by studying the changes in both the control group and its complimentary group (Hinkelmann, & Kempthorne, 2008).

A real-life example of this experiment is conducting treatment for fever amongst children. Children of the same age presenting symptoms are grouped into two; one group, the control gets standard treatment (such as, home remedy), while the other group receives hospital treatment. The result produced after a given period of time will conclude on the most effective type of treatment.

Quasi-experiment design is an experimental design that involves choosing groups, where a specific variable gets tested, in the absence of random processes of pre-selection. For example, testing the impact of alcohol on pregnant women; even though other factors such as socio class and income levels may play role in such a situation; this experiment design fails to consider limitations of the outcome that might be caused by such factors.

In order address selection bias in quasi experiment, the researcher must be ethical. For example, when conducting the impact of alcohol on pregnant women, it would be unethical to group pregnant women in order to conduct the research, as the research should be generalized. Therefore, the biasness in the selection of this particular group can only be addressed by being ethical (Center for Innovation n Research and Teaching, 2016).

Nonexperimental designs is part of research designs where the experimenter either describes or explains about a group or studies or examines the correlation existing between preexisting groups the group members are not assigned randomly and an independent or self-determining variable is not influenced by the person conducting the experiment; therefore, no conclusion regarding informal correlation between variables in the experiment can get drawn. In general, limited attempt is carried out in order to control for possible threats when it comes to internal validity during nonexperimental designs. These types of designs are used to answer questions regarding groups or whether there exist differences among groups. The conclusion made from nonexperimental experiments is mainly expressive in nature. Any efforts to make conclusions regarding casual relationships on the basis of nonexperimental research get done thereafter. In this regard, in order to address internal validity in nonexperimental design all subjects should be selected by random assignment and random sampling in order to make all the groups equal. This helps in making the groups comparable or in treatment (Michael, n.d).

References

Center for Innovation n Research and Teaching. (2016). Benefits & Limitations of Quasi-Experimental Research. Retrieved from https://cirt.gcu.edu/research/developmentresources/research_ready/quasiexperimental/benefits_limits

Hinkelmann, K. & Kempthorne, O. (2008). Design and Analysis of Experiments. Hoboken: John Wiley and Sons, Inc.

Michael, S.R. (n.d.). Threats to Internal & External Validity [PDF Document]. Retrieved from http://www.indiana.edu/~educy520/sec5982/week_9/520in_ex_validity.pdf