The following page provides examples for various coding and quantitative techniques I’ve used over the years that may be useful to others. This page is a work in progress. I will be adding more vignettes when available!

Data visualization tutorial

Link: Dataviz workshop

Call me lazy, but when I read academic papers I like to zoom straight to the figures. Well made figures can tell you practically everything you need to know about a study. Of course, I’ll read the main text too (usually). But good figures help me better disentangle the often disjointed nature of scientific writing. There is a reason that supplementary graphical abstracts are growing in popularity! And with the growing visualization toolbox for data scientists, making high quality figures is easier than ever.

Here is a little workshop I put together for a data science collective in Galapagos in 2019. This is by no means an attempt to reinvent the wheel, which I’m still learning how to maneuver myself. These slides constitute a brief introduction to data visualization, focusing on the Grammar of Graphics philosophy as implemented in R with ggplot2.

Mapping Flickr data

Link: Flickr mapping vignette

If you thought your social media data were only being used for marketing and other profitable ventures, well think again! There has been a boom of environmental research in the last decade using big data from social media platforms including Twitter, Instagram, and Flickr to better understand a myriad of modern ecological and conservation challenges.

In this vignette, I demonstrate some basic tasks with spatial point data using the spatstat package. The dataset consists of Flickr images from the northeastern United States from 2012-2016. This synthesis is part of an ongoing project at SUNY ESF to understand patterns of nature-based tourism and recreation in the Northern Forest region from New York to Maine.

Future topics

I’ll continue dusting off some old analyses to put on here. Some potential additions to this website might include:

  • estimating survival in the survival package
  • estimating survival and other demographic parameters in RMark
  • using occupancy, abundance, and distance sampling models in unmarked
  • randomization tests in base R
  • classification/regression trees and random forests with rpart and randomForest
  • analyzing precipitation data series with SPEI




 

Compiled in R with RMarkdown and Github Pages.
Copyright © Harrison Goldspiel.

hbgoldspiel@gmail.com