geospatial-pilot-notes.Rmd

---
title: "Geospatial Pilot Notes"
output: github_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, message = FALSE, fig.align = 'center')
```


## Day One
### Setup
kyngchaos only works for MacOS Sierra. Homebrew also seemed to not work for older MacOS. This is worth exploring more, and perhaps a minimum OS should be added to the setup\

### Intro to R lesson
Intro to R section had some redundancies and took longer than expected to get through. We ended up only briefly going through section 4, since most of this was covered in section 3, and then skipped section 5 (subsetting) and jump straight into dplyr lesson ofr subsetting.

ggsave is not in the `Writing Data` section, should be added.

Overall this section took too much time, it probably needs to be completely by the first half day at a maximum in order to get through most of the core lesson.

### Geospatial concepts

We found that using a slideshow presentation for the Concepts lesson was more helpful that just using the lesson notes or using screen captures. It would be helpful if we put together a Google Presentation that would be accessible across DC, that way we don't have instructors re-making essential the same slideshow over and over again.

## Day Two

### Raster and vector data in R

The rasters are large, which means that they can take a long time to load up when plotting. It can be a big time suck for the instructor if a typo is made and they have to recreate the plot. Perhaps using smaller raster files or down-sample the resolution moving forward. Ditto for the challenges. Especially when stacking the ggplots.

Another problem with large rasters is it makes it difficult to source the script in order to run later functions. This came up when we had an additional instructor leading a later lesson that required an eariler raster.

The challenges took much longer than anticipated, especially in the early sections. This is primarily due to (1) plots taking a long time to load, and (2) side-questions asked by the learners.

We had to skip Lesson 5 (Multiband data)

As is currently stands, it is probably not possible to do this lesson entirely in one day. We tried to do the setup and the first lesson at the end of Day One, but still we're unable t complete all of the lessons.

We also did a recap at the start of Day Two, which took additional time but seemed necessary and well received by the learners.

We may want to add a note that about functions getting masked by different packages

Did not do lesson 12 - 14

## Reviews

### Minute cards

By far, the most common criticism was that too much time was spent on introductory R material. Several learners suggest that we ask participants to review the `Intro to R` section on their own prior to workshop, and then dedicate the full two days to just working with geospatial data.

## Internal survey results ([Data](https://docs.google.com/spreadsheets/d/1VnG3S7yDM5q2AnAPEyqO7cR5cHhY1i2BW9GBT3-pS2k/edit?usp=sharing))

```{r gs, echo=FALSE, message=FALSE, warning=FALSE}
library(googlesheets)
library(tidyverse)

ggthemr::ggthemr("flat")

# First add sheet to your Googlesheet with "Data" link above
survey <- gs_title("Geospatial Workshop Survey (Responses)") %>% 
  gs_read() %>% 
  janitor::clean_names()

```

```{r plots}

p <- ggplot(survey) 

p +
  geom_bar(aes(how_much_experience_with_r_did_you_have_prior_to_this_workshop)) +
  labs(title = "Prior experience with R before this workshop?", 
       subtitle = "(1) Very little  or no experience -- (5) Highly experienced",
       caption = paste("Number of responses:", nrow(survey)),x = "", y = "")

p +
  geom_bar(aes(based_on_your_level_of_experience_would_you_recommend_we_keep_the_introduction_to_r_section_for_future_workshops)) +
  labs(title = 'Keep "Intro. to R" section in future workshops?', 
       subtitle = "",
       caption = paste("Number of responses:", nrow(survey)),x = "", y = "")

p +
  geom_bar(aes(how_much_experience_with_geospatial_data_did_you_have_prior_to_this_workshop)) +
  labs(title = "Prior experience with geospatial data before this workshop?", 
       subtitle = "(1) Very little  or no experience -- (5) Highly experienced",
       caption = paste("Number of responses:", nrow(survey)),x = "", y = "")

p +
  geom_bar(aes(based_on_your_level_of_experience_would_you_recommend_we_keep_the_introduction_to_geospatial_data_section_for_future_workshops)) +
  labs(title = 'Keep "Intro. to Geospatial Data" section in future workshops?', 
       subtitle = "",
       caption = paste("Number of responses:", nrow(survey)),x = "", y = "")

```