TITLE: Edinburgh Teaching Award
DATE: 2019-05-10
AUTHOR: John L. Godlee
====================================================================


I've just finished a personal development training programme at the
University of Edinburgh called the Edinburgh Teaching Award, which
aims to encourage teachers to develop their skills in teaching and
generally pay attention to how they teach, rather than seeing it as
a chore that they HAVE to do. I received my certificate from the
Higher Education Academy today which says that I'm now an Associate
Fellow. Following that, I figured I would post here the reflective
short articles that formed my submissions for the Teaching Award.

Discussing study design

When facilitating discussion groups for the purpose of designing
student group projects, I find that initially they begin very
quietly, regardless of the students' level of experience. I
considered that this grew from students not knowing what to expect
from this particular discussion, they didn't know what to talk
about as this was their first time meeting together. Gibbs (2014)
suggests that a teacher should aim to set student expectations
early and make students aware of their place in the classroom
hierarchy (A5, K2). To remedy this fear of the unknown and to help
set students' expectations for the tone of the discussion, I
avoided overly open-ended questions like "What hypothesis do you
want to test?", early on in the discussion. From earlier teaching
experience I have found that these kinds of questions favour a
particular kind of student who is quick and confident in their
response, and can lead to the rest of the discussion being
dominated in this way; most students want time to contemplate their
choices before vocalising. Instead I asked more tangible, directed
questions on a narrow topic, though related to the larger subject.
Questions such as "Who can tell me one environmental factor they
think affects forest canopy cover". This helped to build confidence
and get the students to think about the topic in approachable,
testable terms, rather than being overwhelmed by the seemingly
infinite possibilities of designing a study. During a 4th year
discussion I asked one of the students to take note of the answers
to these 'brainstorming' questions, which we then used as a
jumping-off point for part of the discussion, which turned into a
lively debate. In the 4th year discussions, I felt confident
relinquishing my role as leader at this point, as students already
knew the different steps of developing a research question and were
more confident in their approach to group discussion. In these
cases I remained present in the discussion group, acting as a peer
within the group rather than a leader.

In my 2nd year group I found that there was a large disparity in
how interested individual students were in the discussion process,
and also greater uncertainty in debating certain aspects of the
experimental design, such as statistical analysis. I adopted the
advice of Rogers (1969) and attempted to provide a "facilitative
degree of structure" to the study design process. During these
discussions I had to remain more present to encourage deep thinking
and make students aware of the possible lines of enquiry the study
could follow. However, I made sure to allow the students to come to
their own conclusion of what method to follow. In order to ensure
that all students took an active role in the discussion group, even
if they didn't say much during the discussion itself, I asked each
student to produce a concise set of minutes on what was discussed
in the first part of the meeting, and to record anything they
"didn't think of until after the session". These points were then
discussed during the second, shorter part of the meeting. This
post-discussion evaluation helped quieter students to contribute to
the discussion without needing to keep up with the louder students
during the main discussion (V1, V2).

References:

Gibbs, G. (2014). 53 powerful ideas all teachers should know about.
Staff and Educational Development (SEDA) Online blog. Available at:
https://www.seda.ac.uk/53-powerful-ideas/. Last accessed: 19th Sep
2018

Rogers, C. R., & View, L. A. (1969). Freedom to Learn: A View of
What Education Might Become. Columbus, OH: Merrill Pub. Co.

Remaining visible throughout data collection

Harland (2012) suggests that "The chance corridor conversation can
be as critical to a student's learning as the structured tutorial"
(V3). I considered this during my teaching on a field course.
Although the 4th year students were largely left to their own
devices deliberately, and demonstrators were told they didn't have
to follow their groups around, I made sure that I was available and
visible throughout the day, so my students could easily come and
ask me questions when they needed to. I positioned myself in public
areas and aimed to check in with my groups at least once every 2.5
hours during the working day. I also met with my students in the
evening to discuss how the day had gone in an informal manner.
While these periods of "checking-in" appeared to the students to be
just that, I attempted to use the time to stimulate discussion and
prompt self-evaluation of the work so far, with some formative
feedback from me when appropriate (A3). I also tried to prompt
students to think about the details of the methods they had been
using and taught them finer details (K1). One such example being
the use of hemispherical photography of forest canopies, a complex
technique which requires some knowledge of optics to truly
understand. I found that during these post hoc discussions the
students were much more engaged than when I attempted to teach them
before they had used the photography equipment, presumably because
it was easier to link this abstract knowledge to something more
tangible (Boud et al. 1985).

There is often a temptation to get things done as quickly as
possible during fieldwork, with students preferring to stick with a
bad methodology rather than change to a new one due to a perceived
waste of time if they switch. In contrast, critically evaluating
methodology and potentially switching to a new or adapted method is
a valuable learning experience which mirrors real research, and
should therefore be encouraged (V4).

References

Boud, D., Keogh, R., & Walker, D. (1985). Reflection: turning
experience into learning. Learning.

Harland, T. (2012). University Teaching. London: Routledge.

Quantitative Skills in Ecology

The difficulties of effectively teaching quantitative skills in my
field (ecology) are well known (Barraquand et al. 2014). For many
undergraduate ecologists, learning about mathematics and statistics
is not what they signed up for (Duffy 2017). Knowledge of how and
when to apply different statistical methods however, is vital for
graduating ecologists looking to continue in research. It is also a
desirable skill in industry outside of the life sciences, with many
graduate positions preferring students with a background in R, the
current industry standard for statistical computing (TIOBE 2018)
(V4). The question then, is how to support learning of quantitative
skills (A2) and develop effective learning environments for
teaching statistics (A4).

In conversations I've had with undergraduate ecologists, they often
speak of having a 'fear' of statistics. Probing deeper, this fear
is actually a composite of a fear of mathematics and of computer
programming (V2). For many, this fear has developed through
repeated exposure to traditional statistics classes which merely
reinforce that this is an arcane, impenetrable, and fundamentally
boring field.

In 2016, along with a fellow undergraduate, we set up Coding Club
(https://ourcodingclub.github.io), a learning network with an
online presence and in-person workshops which focusses on
collectively teaching each other about statistics, in a
peer-to-peer fashion. We hold weekly workshops based on changing
themes such as linear models or spatial data visualisation.

We found that leaving most of the learning in the hands of the
students is a much more effective teaching model for quantitative
skills, which require practice and critical evaluation at the
student's own pace, which will of course be different to that of
others in the class. Students often said they felt like they had
failed when the group moved on before they had fully understood
what they had coded (V2). It's not that quantitative skills are
necessarily hard or easy by their nature, it's rather that they
take time, and that this time varies between students. This is why
a non-conventional learning environment like Coding Club works so
well. We decided to avoid the model of another popular quantitative
skills program called Data Carpentry (https://datacarpentry.org),
which focusses a lot on "live coding" by a demonstrator. In
practice, while this method can convey a lot of information in a
short period of time, it is debatable whether that learning is
'deep', does the student know why they typed the code (V3)? When a
student has a problem during a Coding Club workshop, whether that
is to do with the prescribed theme or something that they have
brought in from outside, we encourage small groups to form in an ad
hoc manner so multiple people can talk about the issue, which is
often resolved much quicker than a single student staring at code
on a screen and with benefits for both the impromptu teacher and
the student. Conversation helps to rationalise issues which are
initially abstract and difficult to solve (V1). Unfortunately,
current provision of teaching hours and the teacher:student ratio
in many undergraduate quantitative skills classes precludes
teaching in this manner, hence why we felt it was necessary to act
outside of the prescribed timetable.

References

Barraquand, F., Ezard, T. H., Jørgensen, P. S., Zimmerman, N.,
Chamberlain, S., Salguero-Gómez, R., Curran, T. J., Poisot, T.
(2014). Lack of quantitative training among early-career
ecologists: a survey of the problem and potential solutions. PeerJ,
2, e285.

Duffy, M. 2017. Poll results: How mathy are ecology, evolution, and
genetics? [Online] [Accessed 21st Nov 2018]. Available at:
https://dynamicecology.wordpress.com/2017/09/25/poll-results-how-mat
hy-are-ecology-evolution-and-genetics/

TIOBE 2018. TIOBE Index for November 2018. [Online] [Accessed 21st
Nov 2018]. \ Available at: https://www.tiobe.com/tiobe-index/