Geometric Morphometrics (GMM) and Archaeological Science

class: center, middle, inverse, title-slide

# Geometric Morphometrics (GMM) and Archaeological Science
### Dr. Christian Hoggard
### July 2020

---

class: center, middle

# Welcome to these workshops!

---
class: left

# Instructor

**Dr. Christian Steven Hoggard** (Visiting Fellow, University of Southampton)

[<svg style="height:0.8em;top:.04em;position:relative;" viewBox="0 0 512 512"><path d="M459.37 151.716c.325 4.548.325 9.097.325 13.645 0 138.72-105.583 298.558-298.558 298.558-59.452 0-114.68-17.219-161.137-47.106 8.447.974 16.568 1.299 25.34 1.299 49.055 0 94.213-16.568 130.274-44.832-46.132-.975-84.792-31.188-98.112-72.772 6.498.974 12.995 1.624 19.818 1.624 9.421 0 18.843-1.3 27.614-3.573-48.081-9.747-84.143-51.98-84.143-102.985v-1.299c13.969 7.797 30.214 12.67 47.431 13.319-28.264-18.843-46.781-51.005-46.781-87.391 0-19.492 5.197-37.36 14.294-52.954 51.655 63.675 129.3 105.258 216.365 109.807-1.624-7.797-2.599-15.918-2.599-24.04 0-57.828 46.782-104.934 104.934-104.934 30.213 0 57.502 12.67 76.67 33.137 23.715-4.548 46.456-13.32 66.599-25.34-7.798 24.366-24.366 44.833-46.132 57.827 21.117-2.273 41.584-8.122 60.426-16.243-14.292 20.791-32.161 39.308-52.628 54.253z"/></svg> @cshoggard](http://twitter.com/cshoggard) / [<svg style="height:0.8em;top:.04em;position:relative;" viewBox="0 0 496 512"><path d="M165.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6zm-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3zm44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9zM244.8 8C106.1 8 0 113.3 0 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C428.2 457.8 496 362.9 496 252 496 113.3 383.5 8 244.8 8zM97.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1zm-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7zm32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1zm-11.4-14.7c-1.6 1-1.6 3.6 0 5.9 1.6 2.3 4.3 3.3 5.6 2.3 1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2z"/></svg> @cshoggard](http://github.com/cshoggard) / [<svg style="height:0.8em;top:.04em;position:relative;" viewBox="0 0 640 512"><path d="M592 0H48C21.5 0 0 21.5 0 48v320c0 26.5 21.5 48 48 48h245.1v32h-160c-17.7 0-32 14.3-32 32s14.3 32 32 32h384c17.7 0 32-14.3 32-32s-14.3-32-32-32h-160v-32H592c26.5 0 48-21.5 48-48V48c0-26.5-21.5-48-48-48zm-16 352H64V64h512v288z"/></svg> www.sites.google.com/site/christianhoggard](https://www.sites.google.com/site/christianhoggard/home) 
 
[<svg style="height:0.8em;top:.04em;position:relative;" viewBox="0 0 512 512"><path d="M476 3.2L12.5 270.6c-18.1 10.4-15.8 35.6 2.2 43.2L121 358.4l287.3-253.2c5.5-4.9 13.3 2.6 8.6 8.3L176 407v80.5c0 23.6 28.5 32.9 42.5 15.8L282 426l124.6 52.2c14.2 6 30.4-2.9 33-18.2l72-432C515 7.8 493.3-6.8 476 3.2z"/></svg> C.S.Hoggard@soton.ac.uk](mailto:C.S.Hoggard@soton.ac.uk) 
 
Research Interests:
1. Quantitative Archaeology in the R Environment
2. The European Lower and Middle Palaeolithic
3. Cultural Evolution and Taxonomies
4. Open Science Approaches to Archaeology
 
Previous Workshops:
* September 2019: [Sendai (Japan)](https://github.com/CSHoggard/-Morph2019)
* December 2019: [Cologne (Germany)](https://github.com/CSHoggard/-cologne_workshop)
* May 2020: [#StayHomeButStudy (Online)](https://github.com/CSHoggard/-workshopjapan2020)

Publications:
[Click here](https://www.sites.google.com/site/christianhoggard/home/publications)

---
# Workshop Structure

There are **three workshops** in total...

- **Workshop One** (now): Introduction to GMM and Archaeology

- **Workshop Two** (20th July): Landmark-based approaches to GMM (R practical: Geomorph/Tidyverse)

- **Workshop Three** (27th July): Outline-based approaches to GMM (R practical: Momocs/Tidyverse)

About these workshops...

- No prior experience in R is necessary 😄

- All data and R Markdowns (documents and presentations) are on [GitHub](https://www.github.com/CSHoggard/-gmm_liverpool_2020)

- Additional software demonstrations: IDAV Landmark Editor (Checkpoint) and TPSdig2/TpsUtil

- Breaks throughout the workshops will facilitate additional questions and guidance! ✔️

---
class: inverse, left

# What are we going to do today?

I will introduce the subject of geometric morphometrics (GMM), specifically:

1. Shape and shape theory

2. Advantages (and disadvantages) of using GMM

3. History of GMM

4. How we 'do' GMM (data creation, transformation, analysis and visualisation)

---
background-image: url(figures/math.jpg)
background-size: contain

---
class: inverse, middle, center

# GMM & Archaeological Science ✨

---

background-image: url("figures/selden_et_al_2018.PNG")
background-size: 75%
background-position: 50% 30%

.footnote[Selden, R.Z. et al. (2018). Lithic morphological organisation: Gahagan bifaces from the Southern Caddo Area. *Digitial Applications in Archaeology and Cultural Heritage* 10: e00080.]

---
background-image: url("figures/mounier_et_al_2018.jpg")
background-position: center
background-size: 35%
background-position: 50% 30%

.footnote[Mounier, A. et al. (2018).  Who were the Nataruk people? Mandibular morphology among late Pleistocene and early Holocene fisher-forager populations of West Turkana (Kenya). *Journal of Human Evolution* 121: 235-253.]

---
class: left, bottom, inverse

# But what is shape?

---
class: left

background-image: url("figures/circles.jpg")
background-position: center
background-size: 45%
background-position: 50% 60%

# Shape as...

“In general terms, the shape of an object, dataset or image that can be defined as the total of all information that is invariant under **translation**, **rotation**, and **isotropic rescalings**” (Small, 1996: 6)

.footnote[Small, C. (1996). *The statistical theory of shape*. New York: Springer.]

---
class: center

background-image: url("figures/hippo.jpg")
background-position: center
background-size: 35%
background-position: 50% 70%

# But what about size?

Do we use the length? the volume of the object? the mass of an object?

---
class: center, bottom

background-image: url("figures/mitteroecker_et_al_2013.jpg")
background-position: center
background-size: 35%
background-position: 50% 30%

**Centroid size:** square root of the summed squared lengths of the dashed lines 
.footnote[Mitteroecker et al. (2013). A brief review of shape, form, and allometry in geometric morphometrics, with applications to human facial morphology. *Hysterix, the Italian Journal of Mammalogy*: 59-66.
]

---
class: center, middle

background-image: url("figures/ssf.jpg")
background-position: center
background-size: 60%
background-position: 50% 50%

---
class: inverse, left
# Morphometrics 101

* Term: Professor of Zoology Robert Blackith (1957)

* Quantitative study of shape, shape variation and shape covariation

* Two types of morphometric studies:

- Traditional morphometrics (length measurements, angles, ratios…)
  
  - Geometric morphometrics or GMM  (landmarks, outlines, curves, surfaces…)

---
class: left

# Benefits and Drawbacks

.pull-left[###👍 Benefits 👍
- Degree of **shape resolution**

- **Information loss**  (vs. linear measurements)

- **Data can easily be collected** from methods including photographs, drawings and scans

- Abstraction and registration procedure allow an analysis of **shape *sensu stricto***

- Useful **visualisation** tool]

.pull-right[### 👎 Drawbacks 👎
- Required **skill level** (software, statistics...)

- **Proceduralisation** (R is helping!)

- **Landmark subjectivity**]

---
class: inverse, center

# We can use GMM to determine whether…

#### two or more assemblages are different in shape?

#### shape is related to: size? sex? raw material? hominin?

#### shape differences correspond to a hypothesis or a model?

#### mean or median shapes are representative of a site?

#### there are biologically inherent relationships?

#### networks of specific shapes can be identified? (over several sites)

---
class: bottom

background-image: url("figures/time.jpg")
background-size: contain
background-size: 40%
background-position: 95% 20%

#A short history of Geometric Morphometrics...

---
class: left

background-image: url("figures/metmuseum.jpg")
background-size: 50%
background-position: 50% 75%

# Albrecht Dürer (1471-1528)

* Renaissance Paint, printer, theorist and **founder of descriptive geometry**

* Worked on helices, conchoids, epicycloids and the Delian Problem (doubling the cube)

* **Grid transformations** to catalogue and investigate morphological variation

.footnote[*Draughtsman Making a Perspective Drawing of a Reclining Woman* c. 1525 Albrecht Dürer (The Met Museum: Public Domain).]

---
class: left

background-image: url("figures/dwt.jpg")
background-size: 30%
background-position: 95% 50%

### Sir D’Arcy Wentworth Thompson (1860-1948)

* Biologist, mathematician and classics scholar

* Famous for his quotes on the mathematical beauty of nature (inspiring Huxley, Turing, Lévi Strauss and van der Rohe)

### On Growth and Form (1916)

* **Fundamental book on morphological variation in nature**

* Emphasis on **mathemtical structures** accounting for diversity

* Counter-argument to **vitalism** (while  challenging **natural selection**)

.footnote[Professor Sir D'Arcy Wentworth Thompson (after a 1938 original). *David Shanks Ewart (1901-1965). University of Dundee Fine Art Collections.* CC BY-NC-ND. ]

---
class: bottom, center

background-image: url("figures/thompson.jpg")
background-size: 30%
background-position: 50% 25%

Thompson, D.W. (1917). *On Growth and Form*. Cambridge University.

---
class: left

# Transition to Multivariate Morphometrics

* **Fred Bookstein**: Cartesian transformations, Bookstein transformations, etc.

* **David George Kendall (1918-2007)**: Objects of the same shape as seperate points in a geometric space

* **Miriam Zelditch**: Complex shapes, morphological evolution and ecology

* **Ian Dryden and Kanti Mardia**: Development of Fourier-based outline analysis

* **Norman MacLeod**: Eigenshape analysis and palaeontological applications

* **James Rohlf**: Statistical developments, biological applications and software
---
class: left

# Another Ten Minute Break! ⏰
 
## Any Questions?

### Slack Workspace ("workshop_1") and Zoom!

---
class: center, middle
background-image: url("figures/coffee.jpg")
background-size: contain

---
class: left

background-image: url("figures/computer.jpg")
background-size: 25%
background-position: 90% 50%

# Stage 1: Collecting Input Data

### Methods include:

- **CT Scanning**
  
  - **Photogrammetry** and **Structure-from-Motion**
  
  - **Structured-light** Scanning
  
  - **Mono IR** (e.g. Kinect) and **Stereo IR** (Intel RealSense) Scanning
  
  - **MicroScribe** digitisers
  
  - **3D scanners** e.g. NextEngine Desktop
  
  - **Photographs** 
  
  - **Professional Illustrations**

---
class: left, inverse

background-image: url("figures/claude_2008_1.jpg")
background-size: 35%
background-position: 100% 50%

# Stage 2: Landmark Choice

* Landmarks are central to GMM

* Definition: a coordinate point used to represent a shape and/or a homologous point on a structure

* Quantifiable as Cartesian coordinates (x, y / z coordinates)

* Flexibility in what types of landmark are required

* Can be treated as individual points [workshop two] or converted into curves and outlines [workshop three]

.footnote[Claude, J. (2008). *Morphometrics with R*. Springer Publishing.]

---
class: left, inverse

background-image: url("figures/combo.jpg")
background-size: 35%
background-position: 100% 50%

# Stage 2: Landmark Choice

#### Various types of landmarks
* Type I: Homologous biological structures
* Type II: Geometric definition e.g. greatest curvature
* Type III: Point with reference to another point

#### A special case: semi-landmarks 
* Placed using an algorithm
* Can cover the whole or part of a shape
* Equidistant and placed between one or two end points
* A special Type III landmark (*sensu* Bookstein)

.footnote[Ros et al. (2013). Geometric morphometric analysis of grain shape and the identification of two-rowed barley (Hordeum vulgare subsp. distichum L.) in southern France. *Journal of Archaeological Science*. 41.]

---
class: left, inverse

background-image: url("figures/combo.jpg")
background-size: 35%
background-position: 100% 50%

# Stage 2: Landmark Choice

#### Landmarks should...

* sample aspects which are of archaeological interest

* be repeatable and identifiable on all examples (if possible)

* cover as much of the shape as possible

* be sufficient as to not increase the ‘weighting of areas’

* always be plotted in the same order!

---
class: left

# Some observations...

**For archaeological material...** | **For bioarchaeological material...**
--------------------------------------- | ------------------------
Poor morphological correspondence   | Greater morphological correspondence
Less intuitive to examine in 3D     | More intuitive to examine in 3D...
Fewer number of examples to compare with | Greater number of examples to compare with
Fragmentation: major issue  | Fragmentation: major issue
Higher sample size  | Lower sample size is necessary (esp. prehistory) 
Often outline-centric | Often landmark-centric

---
class: left

background-image: url("figures/vestergaard_and_hoggard_2019.jpg")
background-size: 90%

---
class: left

# Stage 3: Landmark digitisation

### A variety of software is available, including:

* TPS Suite (**TpsUtil** and **TpsDig2**)
* CRAN-certified R Packages e.g. **geomorph**, **Momocs**, **StereoMorph** and **shape**
* GitHub-based R Packages e.g. **GUIMorph**
* **SlicerMorph** and **3DSlicer**
* **PhyloNimbus**
* **Stratovan Checkpoint** (£)
* **EVAN Toolkit** (£)
* and many, many more...

Note: IDAV Landmark Editor (no longer available)

Output: various file types including **.tps**, **.nts**, **.csv** and **.txt** formats

---
class: left

background-image: url("https://upload.wikimedia.org/wikipedia/commons/d/d0/RStudio_logo_flat.svg")
background-size: 35%
background-position: 50% 85%

# GMM (and Archaeology) in the R Environment

* **Increasing number of packages** for shape (and archaeological) data

* **Powerful** and **fairly straight-forward** to use

* Last decade: **Archaeology as a more code-literate discipline**

* **Advantages in data visualisation**

* Product: **reproducible** and **replicable** **Open Science** approach

---

background-image: url("https://media.giphy.com/media/l0K4hqqqwgFijgVLa/giphy.gif")
background-size: 60%
background-position: 50% 40%

.footnote[https://media.giphy.com/media/l0K4hqqqwgFijgVLa/giphy.gif]

---

background-image: url("figures/xkcd.png")
background-size: 55%
background-position: 50% 40%

.footnote[https://xkcd.com/1945/]

---
class: left

# Stage 4: Data screening

* Do all my specimens have the **correct number of points**?

* Are all my landmarks in the **correct order**?

* Are the **ID labels correct**?

* Are they to **scale**? (for size-integrated analyses)

* Are **sliders** defined? (if using sliding semi-landmarks)

### Importance of a code-book

* Necessary to **'know'** your meta-data (e.g. #ID, technology, sex or raw material)

* Ensure all your meta-data is **clear**, **easy to understand** and **formatted appropriately**

---
class: inverse, center, middle

# Know your research question!

---

# Ten Minute Break! ⏰
 
## Any Questions?

### Slack Workspace ("workshop_1") and Zoom!

---
background-image: url("figures/coffee.jpg")
background-size: contain

---
class: left

#Stage 5: Data Registration (isolating shape)

A variety of analyses are available in GMM:

1. **Landmark Analysis** (using landmarks)

2. **Outline Analysis** (using semi-landmarks)

* **Elliptic Fourier Analysis** (semi-landmarks of a closed outline)
  
  * **Radii Fourier Analysis** (semi-landmarks of a closed outline)
  
  * **Discrete Cosine Transform** (semi-landmarks of an open outline)
  
3. **Miscellaneous**

* **Eigenshape Analysis** (semi-landmarks of a closed outline)

<center> The registration method changes depending on if you're analysing landmarks or outlines! <center/>

Landmarks: **Generalised Procrustes Analysis** | Outlines: **Fourier-based transformation**

---
class: center

#Generalised Procrustes Analysis (GPA)

#### (Procrustes Superimposition / Procrustes Fitting / Generalised Least Squares)

### **A procedure to isolate shape from a number of variables: rotation, size and translation.**

---

background-image: url("https://media.giphy.com/media/11DFuwckOK9mdG/giphy.gif")
background-size: 40%
background-position: 50% 40%

.footnote[https://media.giphy.com/media/11DFuwckOK9mdG/giphy.gif]

---
class: left

background-image: url("figures/mitteroecker_et_al_2013_2.jpg")
background-size: 50%
background-position: 100% 30%

### Generalised Procrustes Analysis
 
The raw coordinates (for all shapes) are…

* Translated to a **common centroid** (A-B)

* Scaled to a **common centroid size** (B-C)

* Rotated to **minimise the sum of squared-distances between landmarks** (C-D)

* Outcome: **Procrustes coordinates** (shape)

.footnote[Mitteroecker et al. (2013). A brief review of shape, form, and allometry in geometric morphometrics, with applications to human facial morphology. *Hysterix, the Italian Journal of Mammalogy*. pp. 59-66.]

---
class: left

background-image: url("figures/procrustes.png")
background-size: 40%
background-position: 95% 50%

.pull-left[## Why Procrustes?

* Damastes was a son of Poseidon who lived on a sacred way (Attica)

* There he had a bed, in which he invited every passer-by to spend the night. He would set to work on them with his blacksmith's hammer, to stretch them to "fit" the bed

* Nickname: Procrustes ("The Stretcher")

* If the guest proved too tall, Procrustes would amputate the excess length - nobody ever fitted the bed exactly! (**Procrustes Fitting**)

* Procrustes continued his reign of terror until he was  "fitted" to his own bed by Theseus!

* Link to source: [click here](https://medium.com/@bradyholmer/the-procrustean-bed-of-science-b499d79b81de)]

---
class: center

background-image: url("figures/efa.png")
background-size: 35%
background-position: 20% 75%

# Fourier-based transformation

In these methods, the **semi-landmarks are converted into curves**, and the **coefficients which quantify these curves** are what we examine. For example, in **Elliptic Fourier Analysis (EFA)** *sensu* Kuhl and Giardina (1982)...

.footnote[ k = the **total number of steps around the outline** n = the **harmonic number** ∆x = the **displacement** between point p and p+1 ∆t = the **length of the step** between point p and p+1 tp = **accumulated length** of step segments at point pT = **sum of lengths** of all steps around outline ]

---
class: left

## Stage 6: Some Exploratory & Analytical Procedures

Method | Data Input
-------|-----------
**Visualise Shape Change** | Procrustes coord. / Fourier coeff.
**Principal Component Analysis (PCA)**| Procrustes coord. / Fourier coeff.
**Discriminant Analysis (DA/DFA/LDA/CVA)** | Procrustes coord. / Fourier coeff. / PC Scores (+ Factor)
**Procrustes ANOVA / MANOVA** | Procrustes coord. / Fourier coeff. / PC Scores (+ factor) 
**Correlation & Regression** | Procrustes coord. / Fourier coeff. / PC Scores (+ quant. variables)
**Tree-building / Cluster Exercises** | Procrustes coord. / Fourier coeff. / PC Scores (+ optional factor)

---
class: left

background-image: url("figures/claude_2008_2.png")
background-size: 50%
background-position: 95% 60%

# Visualising Shape Change

* Useful for comparing **individual** and **mean** shapes

* Shape change can be represented as:

* Deformation grids
  
  * Contours
  
  * Outlines
  
  * Lollipop sticks
  
  * Vectors

.footnote[Claude, J. (2008). *Morphometrics with R*. Springer Publishing.]

---
class: left

background-image: url("figures/scholtz_et_al_2020.PNG")
background-position: contain

.footnote[Scholtz et al. (2020). D'Arcy W. Thompson's Cartesian transformations: a critical evaluation. *Zoomorphology*. (https://doi.org/10.1007/s00435-020-00494-1)]

---
class: left

# Principal Component Analysis (PCA)

* Often the **first exploratory procedure**

* **Ordination (multi-dimensional)** method

* Each coordinate configuration = a shape

* **Principal components = sources of shape variation**

* Axis origin (0,0) = **mean shape**

* Scree plots (for examining PC contributions)

* **Important: total within-group variance is examined and not between-group variance! (factors overlay the plot!)**

.footnote[Hoggard et al. (2019). The Potential of Geometric Morphometrics for Danish Archaeology: Two Case Studies. *Arkæologisk Forum*, 40: 30-42. (http://www.archaeology.dk/16738/Nr.%2040%20-%202019). OSF: https://osf.io/en5d2/ ]

---
class: left

background-image: url("figures/hoggard_et_al_2019.png") 
background-position: contain
background-size: 70%

---
class: left

background-image: url("figures/vestergaard_and_hoggard_2019_2.jpg") 
background-position: 95% 45%
background-size: 30%

## Discriminant Analysis (DA)

* Otherwise known as **Discriminant Function Analysis (DFA)** or...

* **Linear Discriminant Analysis** (LDA)
 
 * **Canonical Variate Analysis** (CVA)
 
* Investigates *a priori* classification through **maximum between-group classification**

* Percentage = success of classifier (as based on group data)

* **Jacknifed percentage** 👈👈👈

* **Leave-one-out cross-validation**
  
  * **Success with which a random shape can be correctly classified**

Vestergaard, C. and Hoggard, C.S. (2019). A Novel Geometric Morphometric (GMM) 
Application to the Study of Bronze Age Tutuli. *Danish Journal of Archaeology*, 8: 5-28.

---
class: left

# Procrustes ANOVA / MANOVA

* PCA and DA act as **exploratory devices** for looking at shape difference

* MANOVA and Procrustes ANOVA provide a **statistical framework** for examining shape

* Procrustes ANOVA: performed in `R::Geomorph`

* MANOVA: performed in `R::Momocs`

* **Null Hypothesis: same populations / same shape**

---
class: left

background-image: url("figures/hoggard_et_al_2019b.jpg") 
background-position: 95% 50%
background-size: 35%

# Correlation and Regression

* Useful for shape vs. quantitative data.

* **Measure of association (impact of unit change)**

* Possible variables:

* **Centroid size** 
  
  * **Symmetry** 
  
  * **Latitude**
  
  * **Shape score**

* Example methods: `base:: lm ()` or  `tidyverse`

.footnote[Hoggard et al. (2019). The application of elliptic Fourier analysis in understanding biface shape and symmetry through the British Acheulean. *Journal of Paleolithic Archaeology*, 2 (2): 115-133.]

---
class: left

background-image: url("figures/ivanovaite_et_al_2020.jpg") 
background-position: 105% 45%
background-size: 50%

# Cluster & Tree Exercises

* Useful for examining *a posteriori* classifications

* A variety of cluster analyses are available:

* `Momocs::CLUST` (Hierarchical Clustering)
  
  * `Rphylip::contml` (Maximum Likelihood)

.footnote[Ivanovaite et al. (2020). All these fantastic cultures? Research history and regionalisation in the Late Palaeolithic tanged point cultures of Eastern Europe. *European Journal of Archaeology*, 23 (2): 162-185.]

---
class: left

## <center> Technical Notes and Suggestions <center/>

Principal Component Analysis: consider **between-groups PCA (bgPCA)**

Discriminant Analysis: consider **Kovarovic et al. 2011** (notes on sample size!)

Other cool methods:

1) **Machine Learning** (Unsupervised vs. Supervised classificatory methods)

2) **Bayesian Approaches** to GMM

3) **Partial Least Squares (PLS) and Modularity Studies**

.footnote[Kovarovic et al. (2011). Discriminant function analyses in archaeology: are classification rates too good to be true? *Journal of Archaeological Science*, 38: 3008-3018. doi: 10.1016/j.jas.2011.06.028]

---
class: center, middle, inverse

# Concluding Remarks

## Geometric Morphometrics = 😍😍😍

### In the future we will see...

#### a wider application of GMM in Archaeological Science!

#### a more powerful GMM analyses in Archaeological Science!

#### a more open, replicable and reproducible Archaeological Science!

---
class: center, middle

# Thank you for listening!

Slides created via the R package [**xaringan**](https://github.com/yihui/xaringan).

The chakra comes from [remark.js](https://remarkjs.com), [**knitr**](http://yihui.name/knitr), and [R Markdown](https://rmarkdown.rstudio.com).

Licensing: Presentation: [CC-BY-3.0](http://creativecommons.org/licenses/by/3.0/us/)