I had conducted a research investigating students' soft skills practices through learning process between conventional class (Class A) and blended learning class (Class B). Both Class A and Class B attended for 9 courses as well. In Class A, the nine courses were taught by face to face learning (Conventional), while Class B with Blended learning. Since the research aim to compare students soft skills' practices between two learning strategies, then there is no pretest or pre-survey taken. Survey was taken after or at the end of semester by using the questionnaire with likert scale 1-5, measure the practices of soft skills of both classes. And then the result was compared. Your answers will be very valuable for me. Thanks
@Hadiyanto Hadiyanto
Please be certain that it cannot be Experimental Research since there is no way you can control the many variables that can potentially compound the results. Thus, your data cannot be used to test hypothesis or prediction. Your research fits to the Quasi-Experimental (Causal-Comparative) Research method of the Quantitative Research approach. Your two groups are named as "treatment" and "comparison" groups or classes. Having pre-test data could have made your findings or claims stronger.
In this context,depending on whether you carried out randomization of subjects to treatment conditions or not, your appropriate design is posttest non equivalent groups quasi experimental design assuming randomization was not possible otherwise it will be posttest randomized controlled trial experimental design.
If the students were already in Class A and Class B with these conditions (face to face and blended learning, respectively), the "treatment" and "control" conditions were out of your hands as a researcher, so this is essentially a natural experiment.
The justification for not having a baseline measure is not really sound, and it would have helped to strengthen inferences (or to identify whether there were meaningful differences between the two classes before the 9 sessions, which is important), but nothing can be done about that now.
In this case for me, it is a post-test only experimental design with a control group. In post-test only studies with a control group, the subjects are randomly assigned (R) to both the control and the experimental groups. The groups are not previously
submitted to a pre-test. A group is exposed to the treatment (X) or a series of treatments (X1, X2), with the subsequent post-test being carried out to the two groups.
I think, I can get conclusion from your argumentation. It seems my research fit to non randomized sampling of Quasi experiment without pre-test or with post test only. Pre-test is not relevance be taken due to this research investigate students natural practices of Soft skills during learning process between blended learning class and conventional learning. Thank you very much for your all contribution. Very helpful for me. I appreciate very much. Regard.
As research is suggesting some results, it's better to have experiment research always for good results.
Although, the research focus is not on the impact of the strategies on the acquisition of soft skills practices, I suggest you still adopt en experimental design without pretest.
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning