Data_Analysis_Processing_Time.Rmd

---
title: "Microplastics Intercalibration Study Data Analysis: Processing Time"
author: "Southern California Coastal Water Research Project, University of Toronto & State Water Resources Control Board of California"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output:
  html_document: 
    toc: true
    toc_depth: 4
    number_sections: true
    toc_float: true
---
![](www/logobanner.png)
```{r Install Packages, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#### Packages ####
library(tidyverse)#A ton of useful functions
library(knitr) #Needed to display table
library(kableExtra)#data tables
library(Steel.Dwass.test) #steeldwass test

```

```{r Data Import, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

Results <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_results_master", guess_max = 100000) #Raw data file

Lab_Info <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_labinformation", guess_max = 100000) #Import .csv file containing all lab information for clean water

Extraction <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_sampleextraction", guess_max = 100000) #Raw data file for other matrices

Microscopy <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_microscopysettings", guess_max = 100000) #Raw data file for other matrices

NileRed <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_nileredsettings", guess_max = 100000) #Raw data file for other matrices

FTIR <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_ftirsettings", guess_max = 100000) #Raw data file for other matrices

Raman <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_ramansettings", guess_max = 100000) #Raw data file for other matrices

```

```{r}



```

```{r Data Tidying - Extraction, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Remove "Lab" from relevant columns in other matrices data frame
Lab_Info$labid <- gsub("Lab","",as.character(Lab_Info$labid))
Extraction$labid <- gsub("Lab","",as.character(Extraction$labid))
Extraction$sampleid <- gsub("Lab","",as.character(Extraction$sampleid))

#Select data relating to expertise for visual microscopy
Expertise <- Lab_Info %>%
  select(labid, matrix, expertiseextraction, expertisevisualmicroscopy, expertisenilered, expertiseftir, expertiseraman, expertisepy_gcms)

#Join expertise data to results data frame
Extraction <- Extraction %>% 
   left_join(Expertise, by = c("labid" = "labid", "matrix" = "matrix"))

Time_Extraction_Sample <- Extraction %>%
  mutate(sampletype = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD")) %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue"))) %>% 
  #drop blank samples
  mutate(blank = ifelse(grepl("4|8", sampleid),"Y", "N")) %>% 
  filter(blank == "N") %>% 
  #drop where time isn't reported
  filter(time != -88) %>% 
  group_by(sampleid) %>% 
  mutate(time_sample = sum(time)) %>% 
  ungroup()
  
Time_Extraction_Lab <- Time_Extraction_Sample %>% 
  ungroup() %>% 
  group_by(matrix, labid, time_sample, expertiseextraction) %>%
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix, labid) %>% 
  mutate(time_lab = mean(time_sample)) %>% 
  ungroup() %>% 
  mutate(matrix = case_when(
    matrix == "CW" ~ "Clean Water",
    matrix == "FT" ~ "Fish Tissue",
    matrix == "SD" ~ "Sediment",
    matrix == "DW" ~ "Dirty Water"))

```

```{r Data Tidying - Microscopy,  echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Remove "Lab" from relevant columns in other matrices data frame
Microscopy$labid <- gsub("Lab","",as.character(Microscopy$labid))
Microscopy$sampleid <- gsub("Lab","",as.character(Microscopy$sampleid))

#Assign matrices to sampleids
Microscopy <- Microscopy %>% 
  mutate(matrix = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD"))

#Join expertise data to results data frame
Microscopy <- Microscopy %>% 
   left_join(Expertise, by = c("labid" = "labid", "matrix" = "matrix"))

Time_Microscopy_Sample <- Microscopy %>%
  mutate(sampletype = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD")) %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue"))) %>% 
  #drop blank samples
  mutate(blank = ifelse(grepl("4|8", sampleid),"Y", "N")) %>% 
  filter(blank == "N") %>% 
  #drop where time isn't reported
  filter(time != -88) %>%
  group_by(sampleid) %>% 
  mutate(time_sample = sum(time))

Time_Microscopy_Lab <- Time_Microscopy_Sample %>% 
  ungroup() %>% 
  group_by(matrix, labid, sampleid, time_sample, expertisevisualmicroscopy) %>%
  summarise() %>%  
  ungroup() %>% 
  group_by(matrix, labid) %>% 
  mutate(time_lab = mean(time_sample)) %>% 
  ungroup() %>% 
  mutate(matrix = case_when(
    matrix == "CW" ~ "Clean Water",
    matrix == "FT" ~ "Fish Tissue",
    matrix == "SD" ~ "Sediment",
    matrix == "DW" ~ "Dirty Water"))

```

```{r Data Tidying - NileRed,  echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Remove "Lab" from relevant columns in other matrices data frame
NileRed$labid <- gsub("Lab","",as.character(NileRed$labid))
NileRed$sampleid <- gsub("Lab","",as.character(NileRed$sampleid))

#Assign matrices to sampleids
NileRed <- NileRed %>% 
  mutate(matrix = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD"))

#Join expertise data to results data frame
NileRed <- NileRed %>% 
   left_join(Expertise, by = c("labid" = "labid", "matrix" = "matrix"))

Time_NileRed_Sample <- NileRed %>%
  #remove fake lab 
  filter(labid != "ACME") %>%
  mutate(sampletype = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD")) %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue"))) %>% 
  #drop blank samples
  mutate(blank = ifelse(grepl("4|8", sampleid),"Y", "N")) %>% 
  filter(blank == "N") %>% 
  #drop where time isn't reported
  filter(time != -88) %>% 
  group_by(sampleid) %>% 
  mutate(time_sample = sum(time)) 

Time_NileRed_Lab <- Time_NileRed_Sample %>%
  ungroup() %>% 
  group_by(matrix, labid, time_sample, expertisenilered) %>%
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix, labid) %>% 
  mutate(time_lab = mean(time_sample)) %>% 
  ungroup() %>% 
  mutate(matrix = case_when(
    matrix == "CW" ~ "Clean Water",
    matrix == "FT" ~ "Fish Tissue",
    matrix == "SD" ~ "Sediment",
    matrix == "DW" ~ "Dirty Water"))

```

```{r Data Tidying - FTIR,  echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Remove "Lab" from relevant columns in other matrices data frame
FTIR$labid <- gsub("Lab","",as.character(FTIR$labid))
FTIR$sampleid <- gsub("Lab","",as.character(FTIR$sampleid))

#Assign matrices to sampleids
FTIR <- FTIR %>% 
  mutate(matrix = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD"))

#Join expertise data to results data frame
FTIR <- FTIR %>% 
   left_join(Expertise, by = c("labid" = "labid", "matrix" = "matrix"))

Time_FTIR_Sample <- FTIR %>%
  #remove fake lab 
  filter(labid != "ACME") %>%
  mutate(sampletype = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD")) %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue"))) %>%
  #drop blank samples
  mutate(blank = ifelse(grepl("4|8", sampleid),"Y", "N")) %>% 
  filter(blank == "N") %>%
  # Where labs have entered 0 for time, this is assumed that time was not recorded
  mutate(time = na_if(time, "0")) %>% 
  #drop where time isn't reported
  filter(time != -88) %>%
  drop_na(time) %>%
  group_by(sampleid) %>% 
  mutate(time_sample = sum(time))  
  
Time_FTIR_Lab <- Time_FTIR_Sample %>% 
  ungroup() %>% 
  group_by(matrix, labid, time_sample, expertiseftir) %>%
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix, labid) %>% 
  mutate(time_lab = mean(time_sample)) %>% 
  ungroup() %>% 
  mutate(matrix = case_when(
    matrix == "CW" ~ "Clean Water",
    matrix == "FT" ~ "Fish Tissue",
    matrix == "SD" ~ "Sediment",
    matrix == "DW" ~ "Dirty Water"))

```

```{r Data Tidying - Raman,  echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Remove "Lab" from relevant columns in other matrices data frame
Raman$labid <- gsub("Lab","",as.character(Raman$labid))
Raman$sampleid <- gsub("Lab","",as.character(Raman$sampleid))

#Assign matrices to sampleids
Raman <- Raman %>% 
  mutate(matrix = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD"))

#Join expertise data to results data frame
Raman <- Raman %>% 
   left_join(Expertise, by = c("labid" = "labid", "matrix" = "matrix"))

Time_Raman_Sample <- Raman %>%
  #remove fake lab 
  filter(labid != "ACME") %>%
  mutate(sampletype = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD")) %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue"))) %>%
  #drop blank samples
  mutate(blank = ifelse(grepl("4|8", sampleid),"Y", "N")) %>% 
  filter(blank == "N") %>% 
  #drop where time isn't reported
  filter(time != -88) %>% 
  
  group_by(sampleid) %>% 
  mutate(time_sample = sum(time)) %>% 
  
  #Correction to time for Raman analysis for Lab BB - submitted incorrectly
  mutate(time_sample = case_when(grepl("BB_FT_5", sampleid) ~ 26.683,
                                 grepl("BB_FT_6", sampleid) ~ 23.133,
                                 grepl("BB_FT_7", sampleid) ~ 16.65,
                                 TRUE ~ as.numeric(time_sample)))

Time_Raman_Lab <- Time_Raman_Sample %>%
  ungroup() %>% 
  group_by(matrix, labid, time_sample, expertiseraman) %>%
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix, labid) %>% 
  mutate(time_lab = mean(time_sample)) %>% 
  ungroup() %>% 
  mutate(matrix = case_when(
    matrix == "CW" ~ "Clean Water",
    matrix == "FT" ~ "Fish Tissue",
    matrix == "SD" ~ "Sediment",
    matrix == "DW" ~ "Dirty Water"))

```

```{r Data Tidying - Images and measurements, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Tidying of microscopy and nile red data to ensure proper merge
Results_M <- Results %>% 
  #remove fake lab 
  filter(labid != "ACME") %>%
  mutate(sampletype = case_when(grepl("CW", sampleid) ~ "CW",
                            grepl("DW", sampleid) ~ "DW",
                            grepl("FT", sampleid) ~ "FT",
                            grepl("SD", sampleid) ~ "SD")) %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue")))

#Select expertise data for clean water
Expertise <- Lab_Info %>%
  select(labid, expertiseextraction, expertisevisualmicroscopy, expertisenilered, expertiseftir, expertiseraman, expertisepy_gcms)

#Join expertise data to results data frame
Results_M <- Results_M %>% 
   left_join(Expertise, by = c("labid" = "labid"))

```

```{r Measurements Tidying, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

Measurement_Time <- Results_M %>%
  filter(!grepl("4", sampleid)) %>% #Remove results for blanks
  filter(!grepl("8", sampleid)) %>% #Remove results for blanks
  mutate(timeimagesmeasurements = na_if(timeimagesmeasurements, "0")) %>% # Where labs have entered 0 for time, this is assumed that time was not recorded
  filter(!grepl("-88", timeimagesmeasurements)) %>%
  filter(!grepl("88", timeimagesmeasurements)) %>%
  drop_na(timeimagesmeasurements) 

Measurement_Time_Sample <- Measurement_Time %>%
  #By Sample
  group_by(sampletype, labid, sampleid, sizefraction) %>%
  #Time per size fraction per sample
  summarise(time_sizefraction = mean(timeimagesmeasurements)) %>% #Time per sizefraction
  ungroup() %>% 
  #Time per sample
  group_by(sampletype, labid, sampleid) %>%
  mutate(time_sample = sum(time_sizefraction)) %>%
  ungroup() %>%
  #By Lab - Mean of time per sample
  group_by(sampletype, labid) %>% #The only difference from above is that I don't group by sample
  mutate(time_lab = mean(time_sample)) %>%  # Mean time per lab for 1 sample
  mutate(matrix = case_when(
    sampletype == "CW" ~ "Clean Water",
    sampletype == "FT" ~ "Fish Tissue",
    sampletype == "SD" ~ "Sediment",
    sampletype == "DW" ~ "Dirty Water")) %>% 
  subset(select = -c(sampletype))
  
Measurement_Time_Lab <- Measurement_Time_Sample %>%   
  ungroup()%>%
  group_by(matrix, labid, time_lab) %>%
  summarise()%>%
  ungroup() %>%
  group_by(matrix) %>%
  mutate(median = median(time_lab)) %>% 
  mutate(mean = mean(time_lab)) %>% #Mean across all labs
  mutate(sd = sd(time_lab))   #SD across all labs

```

```{r Count Particles Analyzed by Each Method, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

Results_Count <- Results %>% 
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  
  rename("microscopy" = stereoscope) %>% 
  rename("nilered" = fluorescencestaining)

#Particles Counted for Microscopy
Results_Count_Microscopy <- Results_Count %>% 
  #Create factor for sampletype
  mutate(sampletype_f = as.factor(sampletype)) %>%
  #Create factor for microscopy
  mutate(microscopy_f = as.factor(microscopy)) %>%
  #Select particles analyzed by microscopy
  # filter(microscopy_f == "Yes") %>%
  group_by(sampletype_f, sampleid) %>% 
  #Total particles per sample
  mutate(total = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  group_by(labid, sampleid, total) %>% 
  summarise()

#Join DF for number of particles to microscopy DF
Time_Microscopy_Particle <- Time_Microscopy_Sample %>% 
  group_by(labid, sampleid, matrix, time_sample) %>% 
  summarise() %>%  
  left_join(Results_Count_Microscopy, by = c("sampleid" = "sampleid", "labid" = "labid")) %>% 
  mutate(time_particle_sample_min = (time_sample/total)*60) %>% 
  group_by(labid, matrix) %>% 
  mutate(time_particle_lab_min = mean(time_particle_sample_min)) %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(mean = mean(time_particle_lab_min)) %>% #Mean across all labs
  mutate(sd = sd(time_particle_lab_min))   #SD across all labs

#Particles Counted for FTIR
Results_Count_FTIR <- Results_Count %>% 
  #Create factor for sampletype
  mutate(sampletype_f = as.factor(sampletype)) %>%
  #Select particles analyzed by FTIR
  filter(ftir == "Yes") %>% 
  group_by(sampletype_f, sampleid) %>% 
  #Total particles per sample
  mutate(total = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  group_by(labid, sampleid, total) %>% 
  summarise()

#Join DF for number of particles to FITR DF
Time_FTIR_Particle <- Time_FTIR_Sample %>% 
  group_by(labid, sampleid, matrix, time_sample) %>% 
  summarise() %>%  
  left_join(Results_Count_FTIR, by = c("sampleid" = "sampleid", "labid" = "labid")) %>% 
  mutate(time_particle_sample_min = (time_sample/total)*60) %>%
  group_by(labid, matrix) %>%
  mutate(time_particle_lab_min = mean(time_particle_sample_min)) %>%
  ungroup() %>%
  # group_by(matrix) %>%
  mutate(mean = mean(time_particle_lab_min)) %>% #Mean across all labs
  mutate(sd = sd(time_particle_lab_min)) %>% #SD across all labs
  mutate(max = max(time_particle_lab_min)) %>% 
  mutate(min = min(time_particle_lab_min)) %>% 
  mutate(n_samples = n_distinct(sampleid)) %>% 
  mutate(n_labs = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(mean, sd, min, max, n_samples, n_labs) %>% 
  summarise()
  
  
#Particles Counted for Raman
Results_Count_Raman <- Results_Count %>% 
  #Count all particles for lab BB as Raman (submission error)
  mutate(raman = if_else(labid == "BB" & sampletype == "FT", "Yes", raman)) %>% 
  #Create factor for sampletype
  mutate(sampletype_f = as.factor(sampletype)) %>%
  #Select particles analyzed by FTIR
  filter(raman == "Yes") %>% 
  group_by(sampletype_f, sampleid) %>% 
  #Total particles per sample
  mutate(total = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  group_by(labid, sampleid, total) %>% 
  summarise()

#Join DF for number of particles to Raman DF
Time_Raman_Particle <- Time_Raman_Sample %>% 
  group_by(labid, sampleid, matrix, time_sample) %>% 
  summarise() %>%  
  left_join(Results_Count_Raman, by = c("sampleid" = "sampleid", "labid" = "labid")) %>% 
  drop_na(total) %>% 
  mutate(time_particle_sample_min = (time_sample/total)*60) %>% 
  group_by(labid, matrix) %>% 
  mutate(time_particle_lab_min = mean(time_particle_sample_min)) %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(mean = mean(time_particle_lab_min)) %>% #Mean across all labs
  mutate(sd = sd(time_particle_lab_min))   #SD across all labs

```

# Extraction Time

```{r Extraction Time, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.height= 7, fig.width= 10}

a <- Time_Extraction_Sample %>% 
  ggplot(aes(x = labid, y = time_sample)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertiseextraction), fill = as.factor(expertiseextraction)), alpha = 0.7, width = .05, size = 3) +
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  facet_wrap(~matrix) +
  labs(x = "Labs", y = "Hours", fill = "Experience", color = "Experience", title = "Sample Extraction Time", caption = "Points represent individual samples spiked with microplastic particles.")

a

b <- Time_Extraction_Sample %>% 
  group_by(matrix, labid, sampleid, time_sample) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Sample ID" = sampleid, "Time (hours)" = time_sample)
  
kable(b, digits = 2, caption = "Extraction Time per Sample") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

c <- Time_Extraction_Lab %>% 
  ggplot(aes(x = matrix, y = time_lab)) +
  theme_test() +
  geom_boxplot(alpha = 0.5,aes(color = matrix, fill = matrix), outlier.shape = NA) +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Matrix", y = "Hours", title = "Sample Extraction Time", caption = "Points represent the mean sample extraction time from individual labs.")

c

d <- Time_Extraction_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Time (hours)" = time_lab)
  
kable(d, digits = 2, caption = "Mean Extraction Time per Lab") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Time_Extraction_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>%   
  ungroup() %>% 
  group_by(matrix) %>%
  mutate(median = median(time_lab)) %>% 
  mutate(mean = mean(time_lab)) %>%
  mutate(sd = sd(time_lab)) %>%
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, median, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Median" = median, "Mean (hours)" = mean, "Standard Deviation (hours)" = sd, "Number of Labs" = n)
  
kable(e, digits = 2, caption = "Mean Extraction Time by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

```{r Extraction Time - Statistical Tests}

Time_Extraction_Lab_Stats <- Time_Extraction_Lab %>% 
  group_by(matrix, time_lab) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(Time_Extraction_Lab_Stats$time_lab) #Not normal
bartlett.test(time_lab ~ matrix, data = Time_Extraction_Lab_Stats) #Unequal variance
kruskal.test(time_lab ~ matrix, data = Time_Extraction_Lab_Stats) #p < 0.05
Steel.Dwass(Time_Extraction_Lab_Stats$time_lab, Time_Extraction_Lab_Stats$matrix)

```

# Microscopy Time

```{r Microscopy Time, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.height= 7, fig.width= 10}

a <- Time_Microscopy_Sample %>% 
  ggplot(aes(x = labid, y = time_sample)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertisevisualmicroscopy), fill = as.factor(expertisevisualmicroscopy)), alpha = 0.7, width = .05, size = 3) +
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  facet_wrap(~matrix) +
  labs(x = "Labs", y = "Hours", fill = "Experience", color = "Experience", title = "Microscopy Time", caption = "Points represent individual samples spiked with microplastic particles.")

a

b <- Time_Microscopy_Sample %>% 
  group_by(matrix, labid, sampleid, time_sample) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Sample ID" = sampleid, "Time (hours)" = time_sample)
  
kable(b, digits = 2, caption = "Microscopy Time per Sample") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

c <- Time_Microscopy_Lab %>% 
  ggplot(aes(x = matrix, y = time_lab)) +
  theme_test() +
  geom_boxplot(alpha = 0.5,aes(color = matrix, fill = matrix), outlier.shape = NA) +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Matrix", y = "Hours", title = "Microscopy Time", caption = "Points represent the mean sample extraction time from individual labs.")

c

d <- Time_Microscopy_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Time (hours)" = time_lab)
  
kable(d, digits = 2, caption = "Mean Microscopy Time per Lab") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Time_Microscopy_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(median = median(time_lab)) %>% 
  mutate(mean = mean(time_lab)) %>%
  mutate(sd = sd(time_lab)) %>%
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, median, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Median" = median, "Mean (hours)" = mean, "Standard Deviation (hours)" = sd, "Number of Labs" = n)
  
kable(e, digits = 2, caption = "Mean Microscopy by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

f <- Time_Microscopy_Particle %>% 
  group_by(matrix, labid, time_particle_lab_min, mean, sd) %>% 
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Mean (min)" = mean, "Standard Deviation (min)" = sd, "Number of Labs" = n)
  
kable(f, digits = 2, caption = "Mean Microscopy by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

```{r Microscopy Time - Statistical Tests}
#Sample
Time_Microscopy_Lab_Stats <- Time_Microscopy_Lab %>% 
  group_by(matrix, time_lab) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(Time_Microscopy_Lab_Stats$time_lab) #Not normal
bartlett.test(time_lab ~ matrix, data = Time_Microscopy_Lab_Stats) #Unequal variance
kruskal.test(time_lab ~ matrix, data = Time_Microscopy_Lab_Stats) #p > 0.05

#Particle
Time_Microscopy_Lab_Stats_Particle <- Time_Microscopy_Particle %>% 
  group_by(matrix, time_particle_lab_min) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(log10(Time_Microscopy_Lab_Stats_Particle$time_particle_lab_min)) #Normal after log transformation
bartlett.test(log10(time_particle_lab_min) ~ matrix, data = Time_Microscopy_Lab_Stats_Particle) #Equal variance after log transformation
x <- aov(log10(time_particle_lab_min) ~ matrix, data = Time_Microscopy_Lab_Stats_Particle) #p < 0.05
summary(x) #No difference p > 0.05

```

# Nile Red Time

```{r NileRed Time, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.height= 7, fig.width= 10}

a <- Time_NileRed_Sample %>% 
  ggplot(aes(x = labid, y = time_sample)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertisenilered), fill = as.factor(expertisenilered)), alpha = 0.7, width = .05, size = 3) +
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  facet_wrap(~matrix) +
  labs(x = "Labs", y = "Hours", fill = "Experience", color = "Experience", title = "Nile Red Time", caption = "Points represent individual samples spiked with microplastic particles.")

a

b <- Time_NileRed_Sample %>% 
  group_by(matrix, labid, sampleid, time_sample) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Sample ID" = sampleid, "Time (hours)" = time_sample)
  
kable(b, digits = 2, caption = "Nile Red Time per Sample") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

c <- Time_NileRed_Lab %>% 
  ggplot(aes(x = matrix, y = time_lab)) +
  theme_test() +
  geom_boxplot(alpha = 0.5,aes(color = matrix, fill = matrix), outlier.shape = NA) +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Matrix", y = "Hours", title = "Nile Red Time", caption = "Points represent the mean sample extraction time from individual labs.")

c

d <- Time_NileRed_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Time (hours)" = time_lab)
  
kable(d, digits = 2, caption = "Mean Nile Red Time per Lab") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Time_NileRed_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(mean = mean(time_lab)) %>%
  mutate(sd = sd(time_lab)) %>%
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Mean (hours)" = mean, "Standard Deviation (hours)" = sd, "Number of Labs" = n)
  
kable(e, digits = 2, caption = "Mean Nile Red by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

# FTIR Time

```{r FTIR Time, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.height= 7, fig.width= 10}

a <- Time_FTIR_Sample %>% 
  ggplot(aes(x = labid, y = time_sample)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertiseftir), fill = as.factor(expertiseftir)), alpha = 0.7, width = .05, size = 3) +
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  facet_wrap(~matrix) +
  labs(x = "Labs", y = "Hours", fill = "Experience", color = "Experience", title = "FTIR Time", caption = "Points represent individual samples spiked with microplastic particles.")

a

b <- Time_FTIR_Sample %>% 
  group_by(matrix, labid, sampleid, time_sample) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Sample ID" = sampleid, "Time (hours)" = time_sample)
  
kable(b, digits = 2, caption = "FTIR Time per Sample") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

c <- Time_FTIR_Lab %>% 
  ggplot(aes(x = matrix, y = time_lab)) +
  theme_test() +
  geom_boxplot(alpha = 0.5,aes(color = matrix, fill = matrix), outlier.shape = NA) +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Matrix", y = "Hours", title = "FTIR Time", caption = "Points represent the mean sample extraction time from individual labs.")

c

d <- Time_FTIR_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Time (hours)" = time_lab)
  
kable(d, digits = 2, caption = "Mean FTIR Time per Lab") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Time_FTIR_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(median = median(time_lab)) %>% 
  mutate(mean = mean(time_lab)) %>%
  mutate(sd = sd(time_lab)) %>%
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, median, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Median" = median, "Mean (hours)" = mean, "Standard Deviation (hours)" = sd, "Number of Labs" = n)
  
kable(e, digits = 2, caption = "Mean FTIR by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

```{r FTIR Time - Statistical Tests}
#Sample
Time_FTIR_Lab_Stats <- Time_FTIR_Lab %>% 
  group_by(matrix, time_lab) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(log10(Time_FTIR_Lab_Stats$time_lab)) #Normal after log transformation
bartlett.test(log10(time_lab) ~ matrix, data = Time_FTIR_Lab_Stats) #Equal variance after log transformation
x <- aov(log10(time_lab) ~ matrix, data = Time_FTIR_Lab_Stats) #p > 0.05
summary(x)
TukeyHSD(x)

#Particle
Time_FTIR_Lab_Stats_Particle <- Time_FTIR_Particle %>% 
  group_by(matrix, time_particle_lab_min) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(log10(Time_FTIR_Lab_Stats_Particle$time_particle_lab_min)) #Normal after log transformation
bartlett.test(log10(time_particle_lab_min) ~ matrix, data = Time_FTIR_Lab_Stats_Particle) #Equal variance after log transformation
x <- aov(log10(time_particle_lab_min) ~ matrix, data = Time_FTIR_Lab_Stats_Particle) #p < 0.05
summary(x) #No difference p > 0.05
TukeyHSD(x)
```

# Raman Time

```{r Raman Time, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.height= 7, fig.width= 10}

a <- Time_Raman_Sample %>% 
  ggplot(aes(x = labid, y = time_sample)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertiseraman), fill = as.factor(expertiseraman)), alpha = 0.7, width = .05, size = 3) +
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  facet_wrap(~matrix) +
  labs(x = "Labs", y = "Hours", fill = "Experience", color = "Experience", title = "Raman Time", caption = "Points represent individual samples spiked with microplastic particles.")

a

b <- Time_Raman_Sample %>% 
  group_by(matrix, labid, sampleid, time_sample) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Sample ID" = sampleid, "Time (hours)" = time_sample)
  
kable(b, digits = 2, caption = "Raman Time per Sample") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

c <- Time_Raman_Lab %>% 
  ggplot(aes(x = matrix, y = time_lab)) +
  theme_test() +
  geom_boxplot(alpha = 0.5,aes(color = matrix, fill = matrix), outlier.shape = NA) +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Matrix", y = "Hours", title = "Raman Time", caption = "Points represent the mean sample extraction time from individual labs.")

c

d <- Time_Raman_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Time (hours)" = time_lab)
  
kable(d, digits = 2, caption = "Mean Raman Time per Lab") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Time_Raman_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(median = median(time_lab)) %>% 
  mutate(mean = mean(time_lab)) %>%
  mutate(sd = sd(time_lab)) %>%
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, median, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Median" = median, "Mean (hours)" = mean, "Standard Deviation (hours)" = sd, "Number of Labs" = n)
  
kable(e, digits = 2, caption = "Mean Raman by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

```{r Raman Time - Statistical Tests}
#Sample
Time_Raman_Lab_Stats <- Time_Raman_Lab %>% 
  filter(!matrix == "Fish Tissue") %>% 
  group_by(matrix, time_lab) %>% 
  summarise() %>% 
  ungroup() 

shapiro.test(log10(Time_Raman_Lab_Stats$time_lab)) #Normal after log transformation
bartlett.test(log10(time_lab) ~ matrix, data = Time_Raman_Lab_Stats) #Equal variance after log transformation
x <- aov(log10(time_lab) ~ matrix, data = Time_Raman_Lab_Stats) #p > 0.05
summary(x)

#Particle
Time_Raman_Lab_Stats_Particle <- Time_Raman_Particle %>%
  filter(!matrix == "FT") %>%   
  group_by(matrix, time_particle_lab_min) %>% 
  summarise() %>% 
  ungroup() 
   
shapiro.test(sqrt(Time_Raman_Lab_Stats_Particle$time_particle_lab_min)) #Normal after log transformation
bartlett.test(sqrt(time_particle_lab_min) ~ matrix, data = Time_Raman_Lab_Stats_Particle) #Equal variance after log transformation
x <- aov(sqrt(time_particle_lab_min) ~ matrix, data = Time_Raman_Lab_Stats_Particle) #p < 0.05
summary(x) #No difference p > 0.05

```

# Images and measurements

```{r Images and measurements, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.height= 7, fig.width= 10}

a <- Measurement_Time_Sample %>% 
  ggplot(aes(x = labid, y = time_sample)) +
  theme_test() +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  facet_wrap(~matrix) +
  labs(x = "Labs", y = "Hours", fill = "Experience", color = "Experience", title = "Images and Measurements Time", caption = "Points represent individual samples spiked with microplastic particles.")

a

b <- Measurement_Time_Sample %>% 
  group_by(matrix, labid, sampleid, time_sample) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Sample ID" = sampleid, "Time (hours)" = time_sample)
  
kable(b, digits = 2, caption = "Images and Measurements Time") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

c <- Measurement_Time_Lab %>% 
  ggplot(aes(x = matrix, y = time_lab)) +
  theme_test() +
  geom_boxplot(alpha = 0.5,aes(color = matrix, fill = matrix), outlier.shape = NA) +
  geom_point(aes(color = matrix, fill = matrix, alpha = 0.7), width = .05, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Matrix", y = "Hours", title = "Images and Measurements Time", caption = "Points represent the mean sample extraction time from individual labs.")

c

d <- Measurement_Time_Lab %>% 
  group_by(matrix, labid, time_lab) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Lab ID" = labid, "Time (hours)" = time_lab)
  
kable(d, digits = 2, caption = "Mean Images and Measurements per Lab") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Measurement_Time_Lab %>% 
  ungroup() %>% 
  group_by(matrix) %>% 
  mutate(n = n_distinct(labid)) %>% 
  ungroup() %>% 
  group_by(matrix, mean, sd, n) %>% 
  summarise() %>% 
  rename("Matrix" = matrix, "Mean (hours)" = mean, "Standard Deviation (hours)" = sd, "Number of Labs" = n)
  
kable(e, digits = 2, caption = "Mean Images and Measurements by Matrix") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

f <- Measurement_Time %>% 
  #By Sample
  group_by(sampletype, labid, sampleid, sizefraction) %>%  
  #Time per size fraction per sample
  summarise(time_sizefraction = mean(timeimagesmeasurements)) %>% #Time per sizefraction
  ungroup() %>% 
  #Time per sample
  group_by(sampletype, labid, sampleid) %>%
  mutate(time_sample = sum(time_sizefraction)) %>%
  ungroup()

Measurement_Time_Particle <- Measurement_Time %>% 
  left_join(f, by = c("labid", "sampleid", "sizefraction", "sampletype")) %>%   
 
  filter(stereoscope == "Yes") %>%
  group_by(sampleid) %>% 
  #Total particles per sample
  mutate(total = n_distinct(sampleid, particleid)) %>% 
  ungroup()%>%
  
  #Divide time by number of particles to get time per particle
  mutate(time_particle = time_sample/total) %>% 
  #Calculate lab average of time per particle
  group_by(sampletype, labid) %>% 
  mutate(time_particle_lab = mean(time_particle)) %>% 
  ungroup() %>%   
  #Calculate mean and SD across labs
  group_by(sampletype, labid, time_particle_lab) %>% 
  summarise() %>% 
  ungroup() %>% 
  group_by(sampletype) %>% 
  mutate(mean_min = 60*mean(time_particle_lab)) %>% #Mean across all labs
  mutate(sd_min = 60*sd(time_particle_lab)) #SD across all labs
  

```

```{r Measurement Time - Statistical Tests}
#Sample
Measurement_Time_Lab_Stats <- Measurement_Time_Lab %>% 
  group_by(matrix, time_lab) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(Measurement_Time_Lab_Stats$time_lab) #Not normal
bartlett.test(time_lab ~ matrix, data = Measurement_Time_Lab_Stats) #Equal variance
kruskal.test(time_lab ~ matrix, data = Measurement_Time_Lab_Stats) #p > 0.05

#Particle
Measurement_Time_Lab_Stats <- Measurement_Time_Particle %>% 
  group_by(sampletype, time_particle_lab) %>% 
  summarise() %>% 
  ungroup()

shapiro.test(Measurement_Time_Lab_Stats$time_particle_lab) #Not normal
bartlett.test(time_particle_lab ~ sampletype, data = Measurement_Time_Lab_Stats) #Equal variance
kruskal.test(time_particle_lab ~ sampletype, data = Measurement_Time_Lab_Stats) #p > 0.05
```

End of Analysis.