Volunteer Accuracy in a Benthic Macroinvertebrate Participatory Science Project - Data and Code Summary

The data and R code provided here are the underpinnings of a manuscript in the journal, Citizen Science: Theory and Practice (Volunteer Accuracy in a Benthic Macroinvertebrate Participatory Science Project). Volunteer-derived aquatic macroinvertebrate identifications and resulting water quality metrics are compared to results from a professional entomologist. The assessment included a total of 357 benthic macroinvertebrate quality control (QC) samples collected by volunteers using leaf packs, kick nets, and visual assessments between 2011 and 2016 for the Environmental Quality Institute (EQI) in North Carolina, USA. 
Of the 357 total samples, 284 were of sufficient quality to be used in the analysis. Data incudes counts of organisms conducted by volunteers on each sample and counts conducted by an entomologist. Macroinvertebrate index values are calculated based on volunteer and entomologist counts and are compared using linear regression and Bray-Curtis dissimilarity methods. 

Software/Hardware
This code was successfully run in February 2024 using R version 4.3.2, with a PC running Windows 10 Enterprise, 64-bit operating system. R packages used in the analysis include: vegan and reshape2.

Contact
W. Adam Sigler
Check my ORCiD for my current email address
https://orcid.org/0000-0002-4815-0580

Rights
The original data presented here are available under CC-BY 4.0 https://creativecommons.org/licenses/by/4.0/legalcode

Folder Structure and File Descriptions

1_Data
1_CompiledData_2023-03-29.csv
This file (48 columns and 715 rows, including column headings) contains data with all taxa group counts for volunteers and biologist for all sites, seasons, years, and collection methods. Columns are taxa groups and each row is a site visit for the volunteer or entomologist counts. This file is an output from the workflow component of the resource where it was compiled from 11 raw data Excel files.  

2_SampleQuality_2023-03-29.csv
This file is an inventory of sample quality (n = 357), which was created by Virginia Hamilton through a manual assessment of the biologist report indicating condition for each sample. A quality of 1 indicates no issues. A quality of 2 was a rough initial accounting of samples with 3 or fewer taxa observed, but this assignment was not consistently applied and was not used in the analysis. A quality of 3 indicates an issue with preservation or sample labeling that precluded use in the analysis. 

3_TaxaList_2023-12-21_1920.csv
This file contains a list of the 43 taxa under consideration with SMIE name, SMIE tolerance score, trophic group, and Order. This input table provided the taxa group names that was the foundation of Table 1 in the manuscript. Observation counts and similarities were added to this template table during analysis to create Table 1. 

2_Code
AllAnalysis_2024-02-03_1237_was.R
This R script contains code to conduct all analysis and generate all tables and figures for the manuscript. This script imports data from the 1_Data folder and exports results to the 3_Results folder. 

RegressionOnPlot_180527_1510_was.R 
This R script contains code for fitting linear regression models to data and adding statistics to regression plots. This script is used as source code for the primary AllAnalysis_2024-02-03_1237_was.R Script. 

3_Results

Plots Folder
BC.boxplot_quality12_2024-02-03.tiff
Figure 3 in the manuscript - boxplot of similarity scores for the 284 samples used in the analysis. 	

BC.boxplot_quality12wStats_2024-02-03.tiff
Same as Figure 3 from the manuscript, with mean similarity printed on plot.

Index3Plots_AllMethods_2024-02-03.png
Figure 1A from the manuscript  regressions of macroinvertebrate index values for volunteer versus entomologist identified samples; includes all samples (including visual assessment). 

Index3Plots_LeafKick_2024-02-03.png
Figure 1B from the manuscript  regressions of macroinvertebrate index values for volunteer versus entomologist identified samples; includes only leaf pack and kick net samples (excludes visual assessment).  

Main_BC_Figure_2024-02-03.png
Figure 2 from the manuscript similarity scores by taxa group; this version omits labels to facilitate manual labeling of points.

Main_BC_Figure_Label_2024-02-03.png
Same as Figure 2 from the manuscript, with labels automatically added in R for reference, but not sufficiently aesthetic for manuscript.

SampleIndexValues_2024-02-03_was.csv
This file is created by the AllAnalysis_2024-02-03_1237_was.R script and has a row for each sample analyzed (284) and has columns for macroinvertebrate index values based on the volunteer counts and the entomologist count as well as the Bray-Curtis similarity for the volunteer versus entomologist counts. 

Table1_Main__2024-02-03_was.csv
This file is created by the AllAnalysis_2024-02-03_1237_was.R script and is the data for Table 1 in the manuscript. 

TaxaCountDetail_Main__2024-02-03.csv
This file is created by the AllAnalysis_2024-02-03_1237_was.R script. It includes a row for each taxa group in Table 1 and has columns with organism counts for each site visit date for each sampling method for volunteers and the entomologist.