Skip to content

Commit

Permalink
summary stats
Browse files Browse the repository at this point in the history
  • Loading branch information
@carlylovas
carlylovas committed Jan 24, 2024
1 parent d750d5e commit 122c196
Show file tree
Hide file tree
Showing 2 changed files with 225 additions and 539 deletions.
656 changes: 147 additions & 509 deletions R/license_networks.html
View file

Large diffs are not rendered by default.

108 changes: 78 additions & 30 deletions R/license_networks.qmd
View file
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@ To focus on fishery access and harvester adaptability, we've removed indigenous
#| label: cleaning data
#| echo: false
#| message: false
#| fig-height: 9
# Remove indigenous and non-harvester licenses
Expand Down Expand Up @@ -76,43 +77,59 @@ license_types %>%

## Licenses Issued

Below are two plots depicting the total number of licenses issued per year and the number of unique license types issued per year
Below is a time series depicting the total number of licenses issued per year (line) and the number of licensed individuals per year (column)

```{r}
#| label: total licenses issued
#| echo: false
#| message: false
#| warning: false
#| column: page
#| fig-width: 7
#| fig-height: 5
#| column: page
#| fig-width: 8
#| fig-height: 6
#| fig-align: center
ind_lic_portfolio %>%
select(license_year, license_types$license_type) %>%
select(license_year, license_types$license_type) %>% # landings_number) %>%
pivot_longer(cols = 2:56, names_to = "license_type", values_to = "count") %>%
left_join(license_types) %>%
group_by(license_year) %>%
summarise(total = sum(count)) -> total_licenses
ggplot(data = total_licenses)+
ind_lic_portfolio %>%
select(license_year, landings_number) %>%
distinct() %>%
group_by(license_year) %>%
nest() %>%
mutate(total_ind = map_dbl(data, function(x){nrow(x)})) %>%
select(!data) %>%
full_join(total_licenses) -> total_licenses
ggplot(data = total_licenses) +
geom_col(aes(x = license_year, y = total_ind, fill = total_ind)) +
geom_line(aes(x = license_year, y = total)) +
ggtitle("Total Licenses Issued") +
ylab("Total") + xlab("Year") +
ylim(c(10000,20000)) +
theme_gmri(plot.title = element_text(size = 8),
axis.text = element_text(size = 8),
axis.title = element_text(size = 8))
guides(fill = guide_colorbar(title = "Total Individuals", label = F)) +
ylim(c(NA,20000)) +
theme_gmri(plot.title = element_text(size = 10),
axis.text = element_text(size = 10),
axis.title = element_text(size = 10))
```

Available licenses types change over time as new opportunities and management practices take shape and fishery expand or close. Below is a time series depicting the amount of unique license types available in a given year.

```{r}
#| label: unique license types each year
#| echo: false
#| message: false
#| warning: false
#| column: page
#| fig-width: 7
#| fig-height: 5
#| fig-width: 8
#| fig-height: 6
#| fig-align: center
ind_lic_portfolio %>%
select(license_year, license_types$license_type) %>%
Expand All @@ -129,26 +146,29 @@ ggplot(unique_licenses)+
geom_line(aes(x = license_year, y = unique_license_types),
position = "stack") +
ggtitle("Unique License Types") +
theme_gmri(plot.title = element_text(size = 8),
axis.text = element_text(size = 6),
theme_gmri(plot.title = element_text(size = 10),
axis.text = element_text(size = 10),
axis.title = element_blank())
```

## Time series of license holdings

The total amount of licenses issued per year.

```{r}
#| label: time series of license types
#| echo: false
#| message: false
#| warning: false
#| column: page
#| column: screen
#| fig-width: 9
#| fig-height: 13
#| fig-height: 9
#| fig-align: center
ind_lic_portfolio %>%
select(license_year, license_types$license_type) %>%
pivot_longer(cols = 2:55, names_to = "license_type", values_to = "count") %>%
left_join(license_types) %>%
left_join(license_types) %>%
#mutate(description = tolower(description)) %>%
group_by(license_year, license_type, description) %>%
summarise(total = sum(count)) %>%
Expand All @@ -163,49 +183,73 @@ license_time_series %>%
geom_line(aes(x = license_year, y = total)) +
ylim(c(0, NA)) +
theme_gmri(plot.title = element_text(size = 8),
axis.text = element_text(size = 6),
axis.text = element_text(size = 8),
axis.title = element_blank()) +
ggtitle(description)
return(out)
})) -> license_time_series
gridExtra::marrangeGrob(license_time_series$time_series, layout_matrix = matrix(1:12, ncol = 3, nrow = 4, byrow = TRUE), top = NULL, left = textGrob(
expression(bold("Total Annual")), rot = 90, gp = gpar(col = "black", fontsize = 8)))
gridExtra::marrangeGrob(license_time_series$time_series, layout_matrix = matrix(1:12, ncol = 4, nrow = 3, byrow = TRUE), top = NULL, left = textGrob(
expression(bold("Total Annual")), rot = 90, gp = gpar(col = "black", fontsize = 10)))
```

## Average license holdings
Below are the summary statisitcs of the average number of licenses held by an individual per year.

Below are the summary statisitcs of the average number of licenses held by an individual per year. Individuals who were not assigned a landings number (NA) were removed from the dataset before averages were calculated.

```{r}
#| label: average license holdings
#| echo: false
#| message: false
#| warning: false
#| column: page
#| fig-width: 7
#| fig-height: 5
#| fig-width: 8
#| fig-height: 6
#| fig-align: center
# nrow(ind_lic_portfolio %>%
# select(landings_number) %>%
# distinct())
ind_lic_portfolio %>%
select(license_year, license_types$license_type, landings_number) %>%
pivot_longer(cols = 2:56, names_to = "license_type", values_to = "count") %>%
filter(!count == 0) %>%
filter(!is.na(landings_number)) %>%
left_join(license_types) %>%
group_by(license_year, landings_number) %>%
summarise(total = sum(count)) %>%
drop_na() -> average_licenses
distinct() %>%
mutate(total = sum(count)) -> average_licenses
# average_licenses %>% filter(is.na(landings_number))
average_licenses %>%
group_by(license_year) %>%
nest() %>%
mutate(average = map_dbl(data, function(x){mean(x$total)})) %>%
#total_ind = map_dbl(data, function(x){nrow(x)})) %>%
select(!data) -> average_licenses
# (summary(average_licenses))
# summary(lm(total ~ license_year, data = average_licenses))
gt(broom::glance(lm(total ~ license_year, data = average_licenses)))
# gt(broom::glance(lm(average ~ license_year, data = average_licenses)))
average_licenses %>%
ggplot() +
stat_smooth(aes(x = license_year, y = total), method = "lm") +
geom_point(aes(x = license_year, y = average)) +
geom_line(aes(x = license_year, y = average), linewidth = 0.5) +
stat_smooth(aes(x = license_year, y = average), method = "lm") +
ylim(c(1,3)) +
ggtitle("Average number of licenses held per individual") +
theme_gmri(plot.title = element_text(size = 8),
axis.text = element_text(size = 6),
theme_gmri(plot.title = element_text(size = 10),
axis.text = element_text(size = 10),
axis.title = element_blank())
```

## Annual License Network Metrics

### Degree Centrality

"The *degree* of a node is the number of other nodes that single node is connected to.
Expand All @@ -220,6 +264,7 @@ Important nodes tend to have more connections to other nodes. Highly connected n
#| column: page
#| fig-width: 9
#| fig-height: 13
#| fig-align: center
annual_co_matrix <- ind_lic_portfolio %>%
arrange(license_year) %>%
Expand Down Expand Up @@ -271,6 +316,7 @@ degree_plots <- metrics_series %>%
gridExtra::marrangeGrob(degree_plots$time_series, layout_matrix = matrix(1:15, ncol = 3, nrow = 5, byrow= TRUE), top = NULL, left = textGrob(
expression(bold("Degree")), rot = 90, gp = gpar(col = "black", fontsize = 8)))
```

### Betweeness

"\[*Betweenness*\] ranks the nodes based on the flow of connections through the network. Importance is demonstrated through high frequency of connection with multiple other nodes. Nodes with high levels of betweenness tend to serve as a bridge for multiple sets of other important nodes." [*Source*](https://bookdown.org/jdholster1/idsr/network-analysis.html#advanced-case-study)
Expand All @@ -282,6 +328,7 @@ gridExtra::marrangeGrob(degree_plots$time_series, layout_matrix = matrix(1:15, n
#| column: page
#| fig-width: 9
#| fig-height: 13
#| fig-align: center
betweeness_plots <- metrics_series %>%
mutate(time_series = map2(data, description, function(x, description){
Expand Down Expand Up @@ -312,6 +359,7 @@ Density time series and network plots are available [here](https://carlylovas.gi
#| message: false
#| fig-width: 8
#| fig-height: 8
#| fig-align: center
annual_graphs <- annual_metrics %>%
mutate(ggraph = map2(graph_obj, license_year, function(x, y){
Expand Down

0 comments on commit 122c196

Please sign in to comment.