What impact would online
discussion have on student engagement?
Jennie Williams
Stevenson University
Introduction
In my field experience, I am
observing 2 AP Biology classes filled with mostly Seniors and a few Juniors.
The students frequently work cooperatively with one another and interact with
other students when given time to work freely. However, during a lesson or when
discussing a specific topic, students do not offer up comments or questions
unless prompted extensively; in short, they are not engaged in class. When I
talked with the teacher of this class, he indicated that several other teachers
in the school have encountered a similar issue with their Senior classes; more
so than past years. This teacher uses timed essays frequently in class to
prepare students for the AP exam, and he has seen a lower rate of improvement
on the essays this year compared to previous years. In addition, in cases where
their scores dip, the teacher has seen a much larger drop affecting more of the
class than in previous years. He indicated that this trend might be related to
the students’ unwillingness or inability to engage in class discussion. Since
the essays require extensive critical thinking and problem-solving, discussing
topics with classmates would give students a more complete view of concepts and
may help students think more deeply on essays.
I believe that in this case,
lack of student engagement may be a result of the students spending much of their
time online. Since these students are more comfortable interacting and sharing
their ideas online, they may not feel that it is necessary or comfortable to
engage in class discussion. I think that some potential solutions could be to
have the students engage in online discussion or share their thoughts through
other means. The purpose would be to find a method of discussion, idea-sharing,
and critical thinking that allows them to think outside of the box that the
students would be more willing to engage in.
I
believe that an online discussion would be a good way to engage this particular
class because of their affinity for online communication. In past studies,
researchers have found that online discussion, while not very “conversational,”
can lead to higher cognitive engagement and critical thinking (Thomas, 2002).
In addition, online discussion can improve student grades when students
consistently contribute new posts (Palmer, Holt, & Bray, 2008). Since it is
easier to monitor student participation and new contributions in an online
format where there is a record of each student’s involvement, online discussion
may be more beneficial in this aspect than class discussion. A teacher can
require a certain amount of new posts or use bonus points to encourage a higher
level of participation.
I chose this problem for
improvement because engaging students is something that has a huge affect on
learning. I would like to be able to effectively engage my students in class
discussions and class topics, so focusing on it now will provide me with ideas
and methods to use in the future. Through this research project, I hope to find
that using alternate methods of class discussion, specifically online
discussion, will increase student engagement in the class.
Methods
For my study, since there are two
classes of AP Biology students taught by the same teacher, I would use one
class as a control (no change) and the second as a test group (online
discussion). Each class has approximately 20 students, and the class averages
indicate that the students in each class are comparable. Both classes would
receive the same discussion prompts for each lesson, but the control group
would use in-class discussion and the test group would use online discussion.
Based on results from previous studies, I would require students to create
several posts or comment several times in class (Palmer, Holt, & Bray,
2008), make the prompts personally relevant to the students (Skinner, 2009),
and make the discussion boards structured (Salter & Conneely, 2015). Since
several of these studies indicate that using online discussion can potentially
increase student grades, motivation, and engagement, I will analyze my results
in several different ways (Thomas, 2002; Palmer, Holt & Bray, 2008; Salter
& Conneely, 2015).
I
will monitor student grades, participation, and self-reported engagement. Since
student grades in both classes are comparable now, any significant differences
during the study are relevant. I will monitor participation by the amount of
times each student contributes a new post to the online discussion or a new
comment to in-class discussion. I will also use a self-report to determine how
engaged students felt in each class. This will be in the form of a survey that
will allow students to indicate how engaged the students felt, how well they
understood the material in each unit, and how much they felt the discussion
helped them learn. Past studies indicate that extraneous variables such as how
the discussion is structured (Salter & Conneely, 2015), differences between
faculty members (Junco, et al., 2010), or students’ connection to discussion
prompts (Skinner, 2009) could skew results. Using three measures of student
engagement, as well as my study’s design, will reduce these sources of error.
Use of Results
For this study, I will need to determine if online
discussion will improve student engagement. This can be judged by student
grades, participation in discussion, and self-reports from the students. I will
collect student grades on individual assessments as well as overall class
grades and compare my control group with my test group. I will monitor student
participation in discussions by the number of posts a student makes on an
online discussion board or the number of comments a student makes in an
in-class discussion. A self-report assessment will include a direct question
about how engaged the student felt in class as well as questions about how each
student felt about the discussion topics and mode of discussion. This
self-report will give student the opportunity to rate their feelings about the
questions through a Likert scale. When reviewing the self-report data, I will
look at the number of positive responses as well as how positive each response
was.
All
of my data will be displayed comparing the control group with the test group. I
will also include a graph for student grades and participation over time to see
if the test group improved more than the control group over time or if there
was no significant difference. My data will be best displayed in graphs to
highlight any possible differences between the control and test groups.
Example Prompts
1. Describe a
situation where you have seen humans subjected to natural or sexual selection.
Respond to 3 other students’ situations with discussions of evolutionary
relevance and how this selection will change the gene pool.
2. What species
concept do you feel is most relevant and why? Respond to 3 other students’
posts constructively with evidence to support their species concept or to
(respectfully) disagree with their species concept.
Self-Report Questions
Students will rate
how much they feel these questions apply to them from using the following
scale:
1. Completely
Disagree
2. Somewhat
Disagree
3. Somewhat
Agree
4. Completely
Agree
Directions: Put a
number 1-4 beside each question to indicate how much the question applies to
you. Be honest, there are no right answers. Do not put your name on the paper;
answers will be kept anonymous.
________ I feel that my
AP Biology class was interesting and easy to pay attention to.
________ I feel that
I was not engaged in class.
________ I was
interested in the discussion topics given for the unit discussions.
________ I thought
the discussion topics were boring.
________ I enjoyed
discussing topics through in-class/online discussion.
________ I would have
preferred using in-class/online discussion.
Results
The
raw data that I simulated for this study can be found here.
The data was simulated based on what I expect to see as a result of this study.
Discussion
Based on the simulated data, the
test group has higher grades, more participation, and a higher reported
engagement and satisfaction with online discussion than the control group has
with in-class discussion. Looking at the class average grades, the simulated
data shows that both class averages on the tests are similar at first, but the
test group average is higher than the control group average over time. I expect
this to happen if the online discussions are successful, because it will take
some time for students to get used to online discussion if they have not done
it before.
Looking at the participation data,
the students in the test group consistently contribute new ideas than the
students in the control group. For this simulated data, I added several
students in each group that participated very little, if at all. However, based
on what I have seen of in-class discussions with these classes, there were more
students in the control group that did not participate than in the test group.
In addition, since contributions in a face-to-face discussion are less
concrete, I predicted that there would be more students in the control group
that did not contribute the required 3 times and included that in the simulated
data. I also included several students in both sets of data that contributed
more than the required amount. Based on this simulated data, the test group had
a higher average amount of participation than the control group.
From the simulated data, the test
group scored higher overall on the self-report than the control group. However,
there were a few questions that both groups answered similarly on. Since, based
on my predictions, the students should feel more engaged in the test group, I
included that in my simulated data. However, since the students are using the
same discussion prompts, I anticipated that they should score similarly on the
two questions asking about the prompts. Finally, the last two questions ask how
they feel about the mode of discussion used in their class. For the simulated
data, these questions were the widest gap between the classes. I expect the
students to feel more favorably toward online discussion than in-class
discussion based on their previous reactions.
In conclusion, this is what I would
expect the data to show if online discussion is, in fact, a better alternative
to in-class discussion for these students. In a true study of this nature, I
would also look at possible sources of bias for my results. In this case,
better results in the test group may be a result of these students having more
time to think about and reply to their discussions. If online discussion does
not engage these students more than in-class discussion, then all of the data
should be very similar between the two classes or show the opposite trend of
this data.
Junco,
R., Heiberger, G., & Locken, E. (2010). The effect of Twitter on college
student engagement and grades [Electronic version]. Journal of Computer
Assisted Learning, 1-14.
Palmer,
S., Holt, D., & Bray, S. (2008). Does the discussion help? The impact of a
formally assessed online discussion on final student results [Electronic
version]. British Journal of Educational Technology, 39(5),
847-858.
Salter,
N. P., & Conneely, M. R. (2015). Structured and unstructured discussion
forums as tools for student engagement [Electronic version]. Computers in
Human Behavior, 46, 18-25.
Skinner,
E. (2009, July). Using community development theory to improve student
engagement in online discussion: a case study [Electronic version]. ALT-J.
Research in Learning Technology, 17(2), 89-100.
Thomas,
M. J. (2002). Learning within incoherent structures: the space of online
discussion forums [Electronic version]. Journal of Computer Assisted Learning, 18,
351-366.
