Summary mathematics and ecology survey

Total: 937 responses

What are you using mathematics for?
Statistics
878     96%
Theory
361     39%
Decision making
217     24%
People may select more than one checkbox, so percentages add up to more than 100%.
What is your background? As an undergraduate you were studying
Biology
781     83%
Physics
17     2%
Applied or pure mathematics
36     4%
Other
103     11%
Rate your feeling towards using equations
you really dislike it

you really like it
1 -
you really dislike it
67     7%
2
146     16%
3
246     26%
4
247     26%
5 -
you really like it
225     24%
Rate your involvement in the process of ecological modelling in your field
You do not use models

Modelling is your specialty
1 -
You do not use models
102     11%
2
206     22%
3
267     28%
4
207     22%
5 -
Modelling is your specialty
152     16%
Are you satisfied with your understanding of the mathematics behind the models used in your field?
Yes
243     26%
No
690     74%

In the general ecology courses you have followed, how would you describe the level of mathematics (in retrospect)?
Too low
702     75%
Just right
208     22%
Too high
18     2%

Do you think more mathematics classes (statistics not included) during the ecological curriculum would be good?
No
81     9%
Yes, at undergraduate level
129    14%
Yes, at graduate level
149     16%
Yes, at both levels
571     61%

Do you think more classes teaching statistics during the ecological curriculum would be good?
No
48     5%
Yes, at undergraduate level
70     7%
Yes, at graduate level
107     11%
Yes, at both levels
707     75%
Should classes in statistics and mathematics be merged with or separated from classes in programming?
Merged
581     62%
Separated
339     36%

What percentage mathematics, statistics, and programming should approximately cover of the university curriculum of an ecologist, in your opinion?


0
        18             2%
10
        85             9%
20
        306             33%
30
        314             34%
40
        113             12%
50
        66             7%
60
        14             1%
70
        18             2%
80
        2             0%
90
        1             0%
100
        0             0%

If you are lacking knowledge in one or severals areas of mathematics that would be useful to you as an ecologist , please indicate which
Probability
515     58%
Calculus
504     57%
Linear algebra
489     55%
Graph theory
463     52%
Geometry
194     22%
Other
122     14%
People may select more than one checkbox, so percentages add up to more than 100%.
A little more on yourself. You are currently
PhD student
394     42%
Postdoc
187     20%
Lecturer and above
183     20%
Other
173     18%
You are
Male
522     56%
Female
404     43%
People may select more than one checkbox, so percentages may add up to more than 100%.
You completed mainly your studies in
Asia
71     8%
Africa
6     1%
Australia and New Zealand
42     4%
Europe
400     43%
South America
26     3%
North America
387     41%

Any suggestion on mathematical training for ecologists?

[Selection of emblematic comments]

"The mathematical training needs to be specifically tailored for ecologists in such a way that they can see the relevance of quantitative skills from the very beginning. Too many students choose ecology because they think it is a science that doesn't require maths, and too often they are not exposed to the importance of quantitative methods until too late in their training. Instead there are simply general requirements for a maths course that students take reluctantly, and rarely pay too much attention to until it is too late."

 "I double-majored in math and biology, and I'm finding that it put me at a huge advantage compared with the math-abilities of my peers."

"The teacher of statistics should be an ecologist with good knowledge of statistics, not a professional mathematician. Also, tie the theory with real ecological examples. "

"Statistics classes must move away from P-values and arbitrary judgments about "significance."  These approaches have roots going back over 100 years and are now only of historical importance. Current/modern methods should be taught and taught well."

"There is almost 0 training in programming.  Of all things, this needs to change.  Algorithmic thinking is key to solving novel ecological problems with big data sets.  We need to begin teaching this along with a solid foundation in statistics and math that teaches us that both of these are evolving disciplines, not a set of rules set in stone."

 "mathematical training should involve applying theory to ecological problems - this makes it much easier for the students to understand the concepts"

"I really wish I had taken more math and stats courses as an undergrad (and had more opportunities to do so now as a grad student), and I'm really thankful I took a programming course on a whim.  I think a lot of ecological concepts are at least somewhat intuitive, so they are a lot easier to pick up later than the fundamental math/stats that turns data into concept!"

"it should be explained early in the bachelor degree WHY it so important, especially for those who wants to pursue graduate studies"

"As an undergraduate, the only requirements for my ecology degree were one semester of calculus (no integrals) and basic stats. Now I am trying to catch up with the state of the art, especially with multivariate analyses and species distribution modeling. Let's raise the bar in our math education. "

"All ecologists need to be conversant in math/stats.  Not all of them need to be fluent or even very good.  There are still multiple paths that ecologists can follow.  However, students should be aware that more and more of these paths require a solid understanding of math/stats.  The number I suggested above (40%) is misleading.  Some students should have 60-70% of there training in math/stats, whereas others should have 10-20% (many, however, would probably benefit from something in-between these two extremes, hence my suggestion).  It all depends on what path they intend to follow."

"I think practical examples are always important for understanding mathematical concepts. Turn typical ecology-related data analysis problems into exercises in programming / statistics / mathematics. This is how I have had concepts stick through my own learning (i.e. patching up holes in my knowledge that should have been filled during university courses). This is not to say that I hadn't come across some of these mathematical topics at earlier times; just that they were learned through the mathematics dept. and thus did not include a link to biological / ecological problems that I would encounter later in my career."

"It has to be taught in a context relevant to and understood by the ecologist. For example, if a teacher is just using algebra and equations alone to explain a mathematical technique, at least 75% of students (particularly undergrad!) will either switch off, give up or get confused as to why they are having to learn this. Most going into a biological undergrad degree won't even realise how much math and statistics they are going to have to learn. If, however, the technique is demonstrated using an ecological question or theory, I think more people will repond positively to the training and see how it can be useful. "

"More of it! Lots more! There is a huge gap in ecologists' biologists' and other life scientists' quantitative training and subsequent understanding. It is doing all of us and the field of quantitative ecology a huge disservice. Thank you."

"I think it is important that ecologists not only take classes on statistics, but that they take statistics courses taught by ecologists.  Being taught by someone who understands the math but also knows how you will be using it in the field makes a huge difference in learning the methods well, and I think make statistics easier and more fun to learn."

"I have a very strong background in biology and ecology, but most of the positions out there need modelers, which I find frustrating. I want to find out how to supplement my doctoral training with modeling so I can have the tools I need to do the research required today."

"Have courses jointly taught by mathematicians with some real-world experience in ecology and practitioners of ecology who happen to be quantitative.  The courses I took were all theoretical, using hypothetical examples constructed from perfect data. 
I also think it's important to start training ecologists to be "bilingual" in math:  show the mathematical equations, but also the same equation as R code, to get in the habit of going back and forth between the two.  We're usually taught to think about math in ecology in either mathematical notation or as code:  to the instructor, it's equivalent.  To the student, the translation can be overwhelming."

"I am building simulation models. My studies of probability theory are proving most useful at the moment. "

[ Interesting practical suggestions ]

1. Create videos of modelling courses/tricks, so that they can be browsed by anybody who has an internet connection [comes up several times]

"It would be great if somebody (or everybody) set up a youtube channel where we could submit "how to" videos of coding particular statistics, or using particular types of maths.  For example, I recently was trying to learn about determining lambda from a Leslie matrix (i.e. determining the dominant eigenvalue).  It would be wonderful to have an ecology youtube channel where I could go and find information about population ecology, matrix calculations, etc.  There are some videos like this, but it would be great to have a collection of videos that are specifically geared for ecologists (using examples from ecology, etc).  Also, having somebody walk through the steps of building a model in R would be great, or performing path analysis, etc.  The data could be available online, and anyone could follow along with the videos to teach themselves particular materials as needed. "
[In some universities, maybe courses could be recorded? As in e.g. http://oyc.yale.edu/]

2. More workshops organised to keep up with mathematical and statistical techniques needed for ecology
[comes up several times]

3. "One way to reach those who are reluctant regarding statistics, especially at the undergraduate level, may be approach the subject as "data visualization"  This aspect frames the beginning of any set of analyses anyway, and is a great way to introduce the cool capabilities of programming software like R."

[Those who disagree - though in a way it rejoins previous comments, i.e. we should produce/teach more models inspired by concrete ecological problems]

"Leave more room for naturalists/biologists. It is sad to see how ecology and conservation biology are becoming more so the realm of mathematicians and physicists. Without empirical understanding of the natural world (often with MANY hours spent monitoring the study organisms in the field), we often end up with fancy but rather useless models that distract from the tough reality of today: unprecedented loss of biological diversity."

Number of daily responses