Conclusion - Limitations
While action research sets out to isolate interested variables in real time and in its natural environment, it isn’t always perfect. There are some underlying factors that could help explain certain findings, that if aren’t considered, could be misleading.
Achievement for project-based learning and traditional lecture style cannot be pinpointed solely on the pedagogy itself in this study.
In looking at the assessment scores of both groups, there are two stories to tell. In phase one, the problem and lecture group prevailed with higher scores. There are possible reasons for this. First, the assessments for the problem and lecture group were differentiated into three categories: Just getting it, Getting it and Challenge me. Students self-selected the assessment they felt was appropriate for their level of understanding. Assessing students at their level would most likely yield higher results than assessing them at a universal level. A second reason for the possibly higher scores for the problem and lecture group could be the project-based group were assessed incorrectly. In phase one a lot of the project-based students chose to do the problem and lecture assessment, when given the choice, not the presentation of learning option that was designed for them. This could have brought down their scores.
A second story that could be told could be that the project-based group heralded higher assessment scores in phase two. One reason is that all the project-based students were assessed through presentations of learning. This assessment is designed for students to explain their learning and how they incorporated it into their design of a flying object. Another reason was the problem and lecture group was challenged more with problems that encouraged them to think more critically and think at a higher level. This could be said to have lowered their average assessment scores. Also, the assessment for the problem and lecture group was not differentiated. Every student was responsible for the same topics to learn, despite their level of understanding of it.
Sample size is small for the project-based group.
In phase one, only seven students or 8% of my students, were in the project-based group. Any assessment data would be vulnerable to being skewed in either direction because of the small sample size. One student in the project-based group could have a greater effect on the data than a student in the problem and lecture group. In phase two, the sample size increased to 17, or 22% of my students. While this is a healthier increase in sample size compared to phase one, it’s not the ideal proportion. Therefore, the influence of one student is distorted and unequal for both groups.
Choice did not provide a random sampling.
In both phases students were given a choice between the project-based and the problem and lecture group. There were a handful of students who chose the project-based group twice, as well as other students who chose the problem and lecture group twice. The limitation is that students aren’t randomly put into two equal groups through choice. A specific type of student would be drawn towards one learning style and repelled from another. While this is perfectly normal, accepted and encouraged in my class, it doesn’t make for a complete random sample. Therefore, any conclusions that are made need to consider that the sampling of the groups are not random.
Achievement for project-based learning and traditional lecture style cannot be pinpointed solely on the pedagogy itself in this study.
In looking at the assessment scores of both groups, there are two stories to tell. In phase one, the problem and lecture group prevailed with higher scores. There are possible reasons for this. First, the assessments for the problem and lecture group were differentiated into three categories: Just getting it, Getting it and Challenge me. Students self-selected the assessment they felt was appropriate for their level of understanding. Assessing students at their level would most likely yield higher results than assessing them at a universal level. A second reason for the possibly higher scores for the problem and lecture group could be the project-based group were assessed incorrectly. In phase one a lot of the project-based students chose to do the problem and lecture assessment, when given the choice, not the presentation of learning option that was designed for them. This could have brought down their scores.
A second story that could be told could be that the project-based group heralded higher assessment scores in phase two. One reason is that all the project-based students were assessed through presentations of learning. This assessment is designed for students to explain their learning and how they incorporated it into their design of a flying object. Another reason was the problem and lecture group was challenged more with problems that encouraged them to think more critically and think at a higher level. This could be said to have lowered their average assessment scores. Also, the assessment for the problem and lecture group was not differentiated. Every student was responsible for the same topics to learn, despite their level of understanding of it.
Sample size is small for the project-based group.
In phase one, only seven students or 8% of my students, were in the project-based group. Any assessment data would be vulnerable to being skewed in either direction because of the small sample size. One student in the project-based group could have a greater effect on the data than a student in the problem and lecture group. In phase two, the sample size increased to 17, or 22% of my students. While this is a healthier increase in sample size compared to phase one, it’s not the ideal proportion. Therefore, the influence of one student is distorted and unequal for both groups.
Choice did not provide a random sampling.
In both phases students were given a choice between the project-based and the problem and lecture group. There were a handful of students who chose the project-based group twice, as well as other students who chose the problem and lecture group twice. The limitation is that students aren’t randomly put into two equal groups through choice. A specific type of student would be drawn towards one learning style and repelled from another. While this is perfectly normal, accepted and encouraged in my class, it doesn’t make for a complete random sample. Therefore, any conclusions that are made need to consider that the sampling of the groups are not random.