SeasonalEvaluations.Rmd

---
title: "Seasonal Evaluations"
date: "Last updated: 12/15/2020"
output:
  html_document:
    toc: true
    toc_float: true
    collapsed: true
    number_sections: false
    toc_depth: 2
---

```{css, echo=FALSE}
h1 {
  text-align: center;
}
```

Seasonal evaluations are quarterly internal reports that look back on the releases published by meteorological season in order to gain insights into what the most popular topics were across our programs, across media platforms, which assets performed well and which content was less popular.

If you're unsure about release or graphic names, [click here](https://docs.google.com/spreadsheets/d/1bArJV-oHTp09L4fk1vrAXzKVUX-wzr1ONmv0PGuXiwY/edit#gid=1301634659) to find a guide, under **'Graphics Directory'** tab.

***

# **Winter 2020/21**

Winter 2020/21 content is considered to be all weekly releases and graphics published between '2020WinterPackage' and '2021ReturnToCOP', inclusive. 

```{r Read in Data Fall 2020, echo = FALSE, warning=TRUE, message=FALSE}
#read in tracking data from Master Tracking Google Spreadsheet
data <- read.csv('February2021TrackingData.csv')

#load pre-written evaluation functions
source("monthlyreport_functions.R")

#reformat
dataReformatted <- reformat_googlesheet(data)
dataReformatted$year <- as.numeric(dataReformatted$year)
dataReformatted$date <- as.Date(dataReformatted$date, format = "%m/%d/%Y")

#extract data for season
start_date <- as.POSIXct("2020/11/31")
end_date <- as.Date("01/01/2021", format = "%m/%d/%Y")
start_date < end_date
seasonData <- dataReformatted[which(dataReformatted$date > as.Date("11/31/2020", format = "%m/%d/%Y")),]

#read in csv of season's releases
releases <- read.csv("fall2020releases.csv")
releases <- na.omit(releases)
releases$date <- as.Date(releases$date, format="%m/%d/%Y")
require('tidyr')
releases.sep <- separate_rows(releases, graphic, sep=",")
#remove blank space from release and graphic columns
releases.sep$title <- trimws(releases.sep$title, which="both")
releases.sep$graphic <- trimws(releases.sep$graphic, which="both")
#cast date column as date
releases.sep$date <- as.Date(releases.sep$date, format="%m/%d/%Y")
write.csv(releases.sep, "fall2020releasesandgraphics.csv")
```

## Ranked Releases

```{r Ranked Releases Fall 2020, include = FALSE, echo = FALSE, warning=FALSE, message=FALSE}
releasePopularity <- rankReleases(seasonData)
#format table
library(formattable)
countedSortedReleases.formatted <- formattable(releasePopularity, align=c("r", "c", "c", "c", "c"))
#add rank column
countedSortedReleases.formatted$rank <- seq.int(nrow(countedSortedReleases.formatted))
#reorder columns
countedSortedReleases.formatted <- countedSortedReleases.formatted[,c(6,1,2,3,4,5)]
#rename columns
#names(countedSortedReleases.formatted) <- c("Rank", "Release", "Social", "Stories", "TV", "Total Hits")
```

```{r Releases Fall 2020, echo = FALSE, warning=FALSE, message=FALSE}
#merge tracking data with list of Fall 2020 releases
seasonReleasesHits <- merge(releasePopularity, releases, by.x="program.content", by.y="title", all.y=TRUE)
#reformat
seasonsReleasesByHits <- seasonReleasesHits[,c(6,1,2,3,4,5)]
seasonsReleasesByHits <- seasonsReleasesByHits[order(-seasonsReleasesByHits$total),]
names(seasonsReleasesByHits) <- c("Date Published", "Release", "Social", "Stories", "TV", "Total Hits")
#format using HTML widget
library(DT)
datatable(seasonsReleasesByHits, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```


## Ranked Graphics
 
```{r Ranked Graphics Fall 2020, include=FALSE}
graphicPopularity <- rankGraphics(seasonData)
#format table
library(formattable)
countedSortedGraphics.formatted <- formattable(releasePopularity, align=c("r", "c", "c", "c", "c"))
#add rank column
countedSortedGraphics.formatted$rank <- seq.int(nrow(countedSortedGraphics.formatted))
#reorder columns
countedSortedGraphics.formatted <- countedSortedGraphics.formatted[,c(6,1,2,3,4,5)]
#rename columns
names(countedSortedGraphics.formatted) <- c("Rank", "Graphic", "Social", "Stories", "TV", "Total Hits")
```

```{r Fall Graphics Fall 2020, echo=FALSE}
#merge tracking data with list of Fall 2020 Graphics
seasonGraphicsHits <- merge(graphicPopularity, releases.sep, by.x="graphic.used", by.y="graphic", all.y=TRUE)
#reformat
seasonsGraphicsByHits <- seasonGraphicsHits[,c(6,1,2,3,4,5)]
seasonsGraphicsByHits <- seasonsGraphicsByHits[order(seasonsGraphicsByHits$total, decreasing=TRUE),]
names(seasonsGraphicsByHits) <- c("Date Published", "Graphic Name", "Social", "Stories", "TV", "Total Hits")
#format using HTML widget
library(DT)
datatable(seasonsGraphicsByHits, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```

## Previous Content Used

Content usage that was tracked during the fall but was published before the fall (i.e. content that subscribers have gone back to look for). 

The top three pieces of content that subscribers pulled out of the archives were: [2020DistanceEmissions](https://medialibrary.climatecentral.org/resources/travel-and-air-pollution-during-covid-19), WeatherPower (largely due to A.J Fox's ongoing sponsored segment), and [2020CO2Peak](https://medialibrary.climatecentral.org/resources/co2-and-the-climate-curve).

```{r Previous content used Fall 2020, echo = FALSE, warning=FALSE, message=FALSE}
#remove rows which have hits from current season
seasonData.sep <- separate_rows(seasonData, program.content, sep=", ")
nonSeasonReleases <- seasonData.sep[-which(seasonData.sep$program.content %in% releases$title),]

evergreen.rank <- rankReleases(nonSeasonReleases)
evergreen.rank.tidy <- evergreen.rank[-c(4),]
names(evergreen.rank.tidy) <- c("Release", "Social", "Stories", "TV", "Total Hits")
library(DT)
datatable(evergreen.rank.tidy, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```

***

# **Fall 2020**

Fall content is considered to be all weekly releases and graphics published between '2020FallPackage' and '2020SeasonalWarming', inclusive. 

```{r Read in Data Fall 2020, echo = FALSE, warning=TRUE, message=FALSE}
#read in tracking data from Master Tracking Google Spreadsheet
data <- read.csv("November2020TrackingData.csv")

#load pre-written evaluation functions
source("monthlyreport_functions.R")

#reformat
dataReformatted <- reformat_googlesheet(data)
dataReformatted$year <- as.numeric(dataReformatted$year)

#extract data for season
seasonData <- dataReformatted[which(dataReformatted$month %in% c("September", "October", "November") & dataReformatted$year==2020),]

#read in csv of season's releases
releases <- read.csv("fall2020releases.csv")
releases <- na.omit(releases)
releases$date <- as.Date(releases$date, format="%m/%d/%Y")
require('tidyr')
releases.sep <- separate_rows(releases, graphic, sep=",")
#remove blank space from release and graphic columns
releases.sep$title <- trimws(releases.sep$title, which="both")
releases.sep$graphic <- trimws(releases.sep$graphic, which="both")
#cast date column as date
releases.sep$date <- as.Date(releases.sep$date, format="%m/%d/%Y")
write.csv(releases.sep, "fall2020releasesandgraphics.csv")
```

## Ranked Releases

```{r Ranked Releases Fall 2020, include = FALSE, echo = FALSE, warning=FALSE, message=FALSE}
releasePopularity <- rankReleases(seasonData)
#format table
library(formattable)
countedSortedReleases.formatted <- formattable(releasePopularity, align=c("r", "c", "c", "c", "c"))
#add rank column
countedSortedReleases.formatted$rank <- seq.int(nrow(countedSortedReleases.formatted))
#reorder columns
countedSortedReleases.formatted <- countedSortedReleases.formatted[,c(6,1,2,3,4,5)]
#rename columns
#names(countedSortedReleases.formatted) <- c("Rank", "Release", "Social", "Stories", "TV", "Total Hits")
```

```{r Releases Fall 2020, echo = FALSE, warning=FALSE, message=FALSE}
#merge tracking data with list of Fall 2020 releases
seasonReleasesHits <- merge(releasePopularity, releases, by.x="program.content", by.y="title", all.y=TRUE)
#reformat
seasonsReleasesByHits <- seasonReleasesHits[,c(6,1,2,3,4,5)]
seasonsReleasesByHits <- seasonsReleasesByHits[order(-seasonsReleasesByHits$total),]
names(seasonsReleasesByHits) <- c("Date Published", "Release", "Social", "Stories", "TV", "Total Hits")
#format using HTML widget
library(DT)
datatable(seasonsReleasesByHits, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```


## Ranked Graphics
 
```{r Ranked Graphics Fall 2020, include=FALSE}
graphicPopularity <- rankGraphics(seasonData)
#format table
library(formattable)
countedSortedGraphics.formatted <- formattable(releasePopularity, align=c("r", "c", "c", "c", "c"))
#add rank column
countedSortedGraphics.formatted$rank <- seq.int(nrow(countedSortedGraphics.formatted))
#reorder columns
countedSortedGraphics.formatted <- countedSortedGraphics.formatted[,c(6,1,2,3,4,5)]
#rename columns
names(countedSortedGraphics.formatted) <- c("Rank", "Graphic", "Social", "Stories", "TV", "Total Hits")
```

```{r Fall Graphics Fall 2020, echo=FALSE}
#merge tracking data with list of Fall 2020 Graphics
seasonGraphicsHits <- merge(graphicPopularity, releases.sep, by.x="graphic.used", by.y="graphic", all.y=TRUE)
#reformat
seasonsGraphicsByHits <- seasonGraphicsHits[,c(6,1,2,3,4,5)]
seasonsGraphicsByHits <- seasonsGraphicsByHits[order(seasonsGraphicsByHits$total, decreasing=TRUE),]
names(seasonsGraphicsByHits) <- c("Date Published", "Graphic Name", "Social", "Stories", "TV", "Total Hits")
#format using HTML widget
library(DT)
datatable(seasonsGraphicsByHits, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```

## Previous Content Used

Content usage that was tracked during the fall but was published before the fall (i.e. content that subscribers have gone back to look for). 

The top three pieces of content that subscribers pulled out of the archives were: [2020DistanceEmissions](https://medialibrary.climatecentral.org/resources/travel-and-air-pollution-during-covid-19), WeatherPower (largely due to A.J Fox's ongoing sponsored segment), and [2020CO2Peak](https://medialibrary.climatecentral.org/resources/co2-and-the-climate-curve).

```{r Previous content used Fall 2020, echo = FALSE, warning=FALSE, message=FALSE}
#remove rows which have hits from current season
seasonData.sep <- separate_rows(seasonData, program.content, sep=", ")
nonSeasonReleases <- seasonData.sep[-which(seasonData.sep$program.content %in% releases$title),]

evergreen.rank <- rankReleases(nonSeasonReleases)
evergreen.rank.tidy <- evergreen.rank[-c(4),]
names(evergreen.rank.tidy) <- c("Release", "Social", "Stories", "TV", "Total Hits")
library(DT)
datatable(evergreen.rank.tidy, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```

***

# **Summer 2020**

Summer content is considered to be all weekly releases and graphics published between '2020FirstDayAbove' and '2020Records', inclusive. 

```{r Read in Data Summer 2020, echo = FALSE, warning=FALSE, message=FALSE}
#read in tracking data from Master Tracking Google Spreadsheet
data <- read.csv("Sept2020TrackingData.csv")

#load pre-written evaluation functions
source("monthlyreport_functions.R")

#reformat
dataReformatted <- reformat_googlesheet(data)
#extract data for season
seasonData <- dataReformatted[which(dataReformatted$month %in% c("May", "June", "July", "August", "September") & dataReformatted$year==2020),]

#read in csv of season's releases
releases <- read.csv("summer2020releases.csv")
require('tidyr')
releases.sep <- separate_rows(releases, graphic, sep=",")
#remove blank space from release and graphic columns
releases.sep$title <- trimws(releases.sep$title, which="both")
releases.sep$graphic <- trimws(releases.sep$graphic, which="both")
#cast date column as date
releases.sep$date <- as.Date(releases.sep$date, format="%m/%d/%Y")
write.csv(releases.sep, "summer2020releasesandgraphics.csv")
```

## Ranked Releases

```{r Ranked Releases Summer 2020, include = FALSE, echo = FALSE, warning=FALSE, message=FALSE}
releasePopularity <- rankReleases(seasonData)
#format table
library(formattable)
countedSortedReleases.formatted <- formattable(releasePopularity, align=c("r", "c", "c", "c", "c"))
#add rank column
countedSortedReleases.formatted$rank <- seq.int(nrow(countedSortedReleases.formatted))
#reorder columns
countedSortedReleases.formatted <- countedSortedReleases.formatted[,c(6,1,2,3,4,5)]
#rename columns
names(countedSortedReleases.formatted) <- c("Rank", "Release", "Social", "Stories", "TV", "Total Hits")
```

```{r Releases Summer 2020, echo = FALSE, warning=FALSE, message=FALSE}
#merge tracking data with list of Summer 2020 releases
seasonReleasesHits <- merge(releasePopularity, releases, by.x="program.content", by.y="title", all.y=TRUE)
#reformat
seasonsReleasesByHits <- seasonReleasesHits[,c(6,1,2,3,4,5)]
names(seasonsReleasesByHits) <- c("Date Published", "Release", "Social", "Stories", "TV", "Total Hits")
#format using HTML widget
library(DT)
datatable(seasonsReleasesByHits, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```


## Ranked Graphics
 
```{r Ranked Graphics Summer 2020, include=FALSE}
graphicPopularity <- rankGraphics(seasonData)
#format table
library(formattable)
countedSortedGraphics.formatted <- formattable(releasePopularity, align=c("r", "c", "c", "c", "c"))
#add rank column
countedSortedGraphics.formatted$rank <- seq.int(nrow(countedSortedGraphics.formatted))
#reorder columns
countedSortedGraphics.formatted <- countedSortedGraphics.formatted[,c(6,1,2,3,4,5)]
#rename columns
names(countedSortedGraphics.formatted) <- c("Rank", "Graphic", "Social", "Stories", "TV", "Total Hits")
```

```{r Summer Graphics , echo=FALSE}
#merge tracking data with list of Summer 2020 Graphics
seasonGraphicsHits <- merge(graphicPopularity, releases.sep, by.x="graphic.used", by.y="graphic", all.y=TRUE)
#reformat
seasonsGraphicsByHits <- seasonGraphicsHits[,c(6,1,2,3,4,5)]
seasonsGraphicsByHits <- seasonsGraphicsByHits[order(seasonsGraphicsByHits$total, decreasing=TRUE),]
names(seasonsGraphicsByHits) <- c("Date Published", "Graphic Name", "Social", "Stories", "TV", "Total Hits")
#format using HTML widget
library(DT)
datatable(seasonsGraphicsByHits, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```

## Previous Content Used

Content usage that was tracked during the summer but was published before the summer (i.e. content that subscribers have gone back to look for). 

**WeatherPower** was the top evergreen graphic, largely because of AJ Fox's continued sponsored segment. 

After that, the top three 'Evergreen' releases were [2020DaysAbove](https://medialibrary.climatecentral.org/resources/2020-days-above-thresholds), [2020Hurricanes](https://medialibrary.climatecentral.org/resources/2020-hurricane-season) and [2019Warming](https://medialibrary.climatecentral.org/resources/earth-day-fastest-warming-cities-and-states-2019) [20202019EOYGlobalTemps](https://medialibrary.climatecentral.org/resources/2019-in-review-global-temperature-rankings), [2017SummerHeatPrepPackage	](https://medialibrary.climatecentral.org/resources/the-season-of-extreme-heat), and [2019HeatIndex](https://medialibrary.climatecentral.org/resources/climate-matters-extreme-heat-index-threatening-outdoor-sports).


```{r Previous content used Summer 2020, echo = FALSE, warning=FALSE, message=FALSE}
#remove rows which have hits from current season
seasonData.sep <- separate_rows(seasonData, program.content, sep=", ")
nonSeasonReleases <- seasonData.sep[-which(seasonData.sep$program.content %in% releases$title),]

evergreen.rank <- rankReleases(nonSeasonReleases)
evergreen.rank.tidy <- evergreen.rank[-c(2),]
names(evergreen.rank.tidy) <- c("Release", "Social", "Stories", "TV", "Total Hits")
library(DT)
datatable(evergreen.rank.tidy, fillContainer=TRUE, options = list(scrollX="70px", scrollY="400px", pageLength = 10), rownames=FALSE)
```

***