28 From plain English to a matrix model

28.1 Introduction

This is a short, quick-fire in-class activity for practising the translation of a verbal life-history description into a matrix population model. It corresponds to the appendix chapter From plain English to a matrix model in the main book and reinforces the workflow from the Matrix population modelling exercise.

The whole point is the mapping rule: fecundity terms go in the top row; survival and stage-transition terms go on the sub-diagonal / main diagonal, with the matrix written in “row = stage at \(t+1\), column = stage at \(t\)” form.

28.2 Learning Outcomes

By the end of this exercise, students will be able to: - Convert a verbal life-history description into a projection matrix. - Build and interpret a simple projection from initial stage abundances. - State the assumptions needed to fill in ambiguous transition rates.

28.3 Activity Overview

Suggested Timings: - 5 minutes: Recap the mapping rule (fecundity row vs survival/transition). - 10 minutes: Warm-up 2-stage worked example on the board. - 15 minutes: Students translate the red-tailed hawk description into a 3×3 matrix and project it in R. - 5 minutes: Compare matrices across the class and discuss assumptions.

28.4 Instructions for Facilitating

  1. Do the 2-stage worked example first: “juveniles do not reproduce, adults produce 2 offspring/year, juvenile survival 0.4, adult survival 0.7” gives \[A = \begin{bmatrix} 0 & 2 \\ 0.4 & 0.7 \end{bmatrix}.\] Narrate where each number lands and why.
  2. Hand out the red-tailed hawk life-history description (supplemental/Exercise_Plain_English_To_Matrix_Model_2018.txt in the main-book materials).
  3. Ask students to draw a 3-stage life-cycle diagram (hatchling → juvenile → adult) before writing any numbers.
  4. Have them state assumptions explicitly (e.g. is there juvenile stasis? what is the transition vs stay probability?).
  5. Project with the given initial vector (1000 hatchlings, 150 juveniles, 5 adults) for 20 steps and compute the dominant eigenvalue.

28.5 Questions & Model Answers

  1. Where does each biological statement go in the matrix?
    • “Adults produce X offspring” → top row (fecundity). “Stage A individuals survive and move to stage B with probability p” → sub-diagonal entry (row B, column A). “Individuals survive and remain in the same stage” → diagonal entry.
  2. What assumptions are needed for the hawk model?
    • Whether hatchlings can remain hatchlings (stasis), whether all survivors advance a stage, and how the verbal survival figures split between “survive and stay” vs “survive and grow”. Any reasonable, clearly stated choice is acceptable — the point is to make the assumption explicit.
  3. Does the population grow or decline?
    • Compute the dominant eigenvalue of the assembled matrix; \(\lambda > 1\) means growth, \(\lambda < 1\) decline. Students should read this from the eigenvalue and confirm it against the 20-step projection.

28.6 Teaching Tips

  • Diagram before numbers: insisting on the life-cycle diagram first prevents most mis-placed entries.
  • Expose disagreement: different reasonable assumptions give different matrices — surface this as a feature, not a bug, and tie it back to how real demographers make modelling choices.
  • Reuse the earlier workflow: the projection and eigenvalue code is identical to the Matrix population modelling exercise; point students back to it.

28.7 Common Pitfalls

  • Fecundity in the wrong place: reproduction belongs in the top row, not on the diagonal.
  • Transition probabilities summing to more than 1 out of a stage.
  • Forgetting to state assumptions, which makes the resulting matrix impossible to interpret or compare.