Analysis of internet usage and bioblitz frequency in the Global South

This dataset was used to analyse factors contributing to the number of bioblitzes conducting in countries in the Global South.

This is part of a review into the effectiveness of bioblitz as a method for collecting data on biodiversity.

We modeled population and internet usage with the number of iNaturalist Bioblitzes in a country from our sample (Groom 2021). We only just looked at those global regions where citizen science has tended to have lower prevalence in the past (Africa, Asia,  and Latin America, and the Caribbean) compared to other regions. We identified a total of 254 Bioblitz projects from iNaturalist in 37 countries, in Africa (30 projects in 13 countries); Asia (71 projects in 11 countries); and Latin America and the Caribbean (153 projects in 13 countries).

We took the total population of each country from the mean of 2015-18 values in millions from https://population.un.org/wpp/Download/Standard/Population/ (United Nations, Department of Economic and Social Affairs, Population Division (2019). World Population Prospects 2019, Online Edition. Rev. 1.). Internet usage was taken as the percentage of individuals using the internet in 2017 (http://data.un.org/).

The natural log of the number of iNaturalist projects was modelled against the log of the population in millions and the internet usage using the lm package of R.

SUMMARY_DATA <- read.delim2("summary_data.tsv", row.names=1)
model <- lm(log(projects) ~log(population) + internet, data=SUMMARY_DATA)
summary(model)

Residual standard error: 0.9167 on 34 degrees of freedom

Multiple R-squared:  0.3911, Adjusted R-squared:  0.3553 

F-statistic: 10.92 on 2 and 34 DF,  p-value: 0.0002176

To view properties of the model to ensure it conformed to the assumptions of the model and was a good fit. Plots are included in the attached files.

par(mfrow = c(2,2))
plot(model)

To view other correlations in the data the following code can be used. The output is included in the attached files.

correlations <- cor(SUMMARY_DATA[,c(2,3,4,6,11,13,14)],method = c("spearman"))
install.packages("corrplot")
library("corrplot")
corrplot(correlations, method="square")