Update
I like concept maps, I think they're a valuable tool, but students do need training and I found that they also need motivation. Over the years, I used some of the ideas I obtained from concept mapping to develop a much more flexible and easy to use tool for both learning and assessing. To find out what it is, you can read about it in my book.
I like cmaps because:
They force students to reflect on and generate knowledge (they really have to think to do it and cannot copy from conventional sources)
They are excellent for assessing student learning & comprehension (you can both see student learning easily and quickly)
Predicting success on exams (in conjunction with other measures).
When I first read about the power of concept mapping in Kinchin’s Visualising Powerful Knowledge (2016), I was blown away. But it would take me some time to really get my head around it, understand it, and utilise it well. Concept mapping is not a tool you can read on Twitter today and start using well tomorrow. However, in this blog I will share my tips with you, to make the transition easier and shorter.
At this point in my journey I have used concept maps as a teaching tool (see this blog), a planning tool, and I have used concept mapping as a homework activity with my IB students (KS5). The latter is the focus of this blog, and the reason is because I find them as excellent tools for:
Predicting success
This year I have used two homework activities with my IB students. Firstly, weekly quizzes of five randomly chosen core questions. As students have the questions and answers available to them, this represents a measure of content retention (important in biology) but cannot be used to assess anything beyond rote learning. Secondly, I have set a weekly concept map, which students complete on cmapcloud and then email to me. These are more a measure of understanding, and reflection.
After my end of first-year mock exam (two papers), I correlated the data of the exam scores with my other data, this is what I found (class of 15 students):
R2= 0.5 with quiz data
R2= 0.5 with cmap data
R2= 0.7 with average of quiz and cmap data (very high correlation for social science)
(R2= 0 means there is no correlation, R2= 1 means there is an exact relationship, which is generally impossible in educational data)
In this post I will describe what I have learnt from doing this, and also my tips to teachers, based on those I give to my students.
Building a Novakian concept map and overcoming student problems
There are a few terms that are confused, and most students also have experience with the variety of maps. Here I am referring only to Novakian (invented by Joseph Novak) concept maps. These are generally both hierarchical and top-down in orientation. At first, I questioned the necessity of this, but I came to realise its importance.
The top-down orientation makes maps easier to read, see their structure and compare with other maps.
The hierarchical aspect is also important, as it forces students to reflect more on the organisation of concepts. More inclusive, more encompassing, more important concepts should come higher in the hierarchy, whereas more detailed concepts should appear lower down.
I have learnt that a small degree of flexibility is required with this. For example, some maps will contain concepts with cyclical relationships (Safayeni & Derbentseva, 2005), and so it can’t be totally top-down, and some maps are better structured when an important and encompassing concept is positioned further down a map.
Here is an example of two maps I have made which shows these two caveats. One explains better than the other, but isn't strictly hierarchical. The other, which is strictly hierarchical doesn't explain as well (it's harder to read and follow) despite having similar content. In summary, hierarchy is a good starting point, but flexibility allows for better maps in some cases.
Not purely hierarchical and more explanatory
More hierarchical structure but less explanatory
At first students will mainly struggle with the reflective nature of the map. They will attempt to organise the map in sentences, using grammar as the organising principle, and placing phrases in concept boxes or in the links. This will also produce lots of 'chains' rather than 'networks'. This is probably because that is how they have learnt the knowledge (in prose), and it makes we wonder if students are really discerning between concepts and their relationships.
In a Novakian cmap, only single concepts should be in a box, and the organisation of concepts is the priority. Thus, students need to convert their learnt sentence structures into a map of organisation. My students took a few weeks to understand this. Here’s something I have used with students to help them:
A quick guide for novice students
Initially students will tend to link concepts with lines but without a linking phrase, or the phrase is a single word. They may claim that the links are obvious, maybe for them, but a cmap is constructed to help me understand their thinking, so you have to insist on proper linking. Look at this map from the second week (still a novice mapper):
In fact, the linking phrases are one of the most important, and revealing, components of a cmap. A lot of students will begin making cmaps based on composition, such as this biological component has this, has this, and has that, rather than causal or correlational relationships. The best maps contain more of the latter. Compare this two maps made by two different students on the same topic:
More composition-based map
More explanatory map
Miller (2008) categorised linking phrases into static (often seen as ‘is’, ‘can be’, or ‘has’), and dynamic propositions. Basically, for a good cmap, the more dynamic propositions there are the better.
Examples of static propositions: The sun is a star, Means of transportation include land transport, Panama is located in Central America, Animals may be vertebrates.
Examples of non-causative dynamic propositions: Roots absorb water, Herbivores eat plants, Living beings need oxygen.
Examples of causative dynamic propositions: Cigarettes produce cancer, Rule of law attracts foreign investment, Heat melts ice, Paper comes from trees.
Examples of quantified causative dynamic propositions: Increased transparency in public affairs discourages corruption, Under-activity of the thyroid gland decreases body metabolism, Increased quality of education contributes to greater national development.
When I first read Kinchin’s work (2016) it suggested that one of the most important things about a good concept map is to ask a good question in the first place. This is another point I overlooked for some time. A good concept map is explanatory, but if you ask students to make a map on ‘the circulatory system’ I have found you will likely get more static propositions, and few cross-links, as students focus mainly on composition and function.
You are much more likely to get an explanatory map, and more dynamic links by asking a good question, such as: Why do some multicellular organisms have circulatory system?
Visualising student thinking
Aside from getting students to reflect more on their knowledge, cmaps are an incredibly useful tool for visualising student thinking because they are so quick to read. A written response is linear, just sentence after sentence, it takes longer to read and is harder for the teacher to discern the links the student is making. A cmap is two dimensional, its more concise, and lays out the student-made connections explicitly.
It typically takes me 10-15 minutes to look through them, score them, and chose two or three examples. I then use these examples to give anonymous, whole-class feedback by recording a screencast of the maps and myself talking over them. This is normally around a five-minute video which I then upload and share with students. From here I may give a new cmap question, but sometimes, I ask the students to improve them for the following week.
Assessing the maps
When students hand in their work I read through them and score them. In contrast to the weekly quizzes this is a more qualitative measure and I score them with either 1, 2, or 3. This is a holistic judgement based on two things: Structure and content. Cañas, Novak, and Reiska (2015) suggest that a 2 is given to a good map, which requires both good structure (readable, good cross links), and good content (good concepts, and good links, which are not just static). A 3 is given to an excellent map, which is more of a holistic judgement, a feeling that his map is excellent as it explains well, it’s content and structure and just right in terms of detail and brevity. A 1 is given to students who either don’t have good structure (messy, few links), good content (appropriate level for the course), or neither of the two.
Also, if you're interested, it is worth reading Kinchin's (2018) short paper on 'A ‘species identification’ approach to concept mapping in the classroom' in which he argues that the overall structure of a map is a big indicator of the type of thinking.
I keep the scores on a spread sheet and generate a rolling average as I do with the weekly quizzes.
Final advice
Novakian concept mapping is a fantastic tool, but you have to play with it a bit at first to fully understand its capabilities. It's not a fad, it's not a pick up and go tool, but after a few weeks of working with your students you are not just gaining and good pedagogical tool, you are also providing students with a skill that can help them study in other subjects, and into their future.
I found it invaluable to spend some time making my own concept maps on biological phenomena. The effort I made here allowed me to understand how to get the most out of the tool, what works, and how to explain it to students. I actually quite enjoyed making them, thinking it through.
To get your novice mappers going, give them a good question, give them a guide of how many concepts to include (maybe 12-20), and encourage them to make a list of the concepts they want to include. Part of the skill is discerning the limits of the map and not including too much. From there, see what happens, give feedback, and see how your students develop. Just be ready for them to struggle at first. They will get better and like it more as they improve.
Update
I like concept maps, I think they're a valuable tool, but students do need training and I found that they also need motivation. Over the years, I used some of the ideas I obtained from concept mapping to develop a much more flexible and easy to use tool for both learning and assessing. To find out what it is, you can read about it in my book.
Christian Moore-Anderson
@CMooreAnderson (twitter)
References
Cañas, A. J., Novak, J. D., & Reiska, P. (2015). How good is my concept map? Am I a good Cmapper? Knowledge Management & E-Learning, 7(1), 6–19
Kinchin, I., 2016. Visualising powerful knowledge to develop the expert student. Rotterdam: Sense Publishers.
Kinchin, I., 2018. A ‘species identification’ approach to concept mapping in the classroom. Journal of Biological Education, pp. 1-7.
Miller, N. L., & Cañas, A. J. (2008). Effect of the nature of the focus question on presence of dynamic propositions in a concept map. In A. J. Cañas, P. Reiska, M. Åhlberg, & J. D. Novak (Eds.), Proceedings of the 3rd international conference on concept mapping. Tallinn, Estonia & Helsinki, Finland.