app.R

# Load Packages
library(shiny)
library(shinydashboard)
library(shinyBS)
library(boastUtils)
library(ggplot2)
library(car)
library(sortable)
library(shinyWidgets)
library(rstatix)
library(lme4)

# Global Constants, Functions, and Data Sets ----
barley1 <- read.csv("blockingValid.csv", header = TRUE)
barley2 <- read.csv("blockingInvalid.csv", header = TRUE)
barley3 <- read.csv("barley3.csv", header = TRUE)
ansOptionAnovaAssumption <- list("Normality of Residuals",
                                 "Homoscedasticity",
                                 "Independence of Observations")
ansOptionAncovaAssumption <- list("Normality of Residuals",
                                  "Homoscedasticity",
                                  "Independence of Observations",
                                  "Linear Relationship covariate and the Response",
                                  "Equality of the covariate's Slope parameter",
                                  "No Potential Outliers")
ansOptionBlockingAssumption <- list("Normality of Residuals",
                                    "Homoscedasticity" ,
                                    "Independence of Observations",
                                    "Interaction of Block and Treatment")
ansOptionRandomAssumption <- list("Normality of Residuals",
                                  "Homoscedasticity",
                                  "Independence of Observations",
                                  "Random Effects")
ansOptionRepeatAssumption <- list("Normality of Residuals",
                                  "Homoscedasticity",
                                  "Independence of Observations",
                                  "Interaction of Block and Treatment",
                                  "Random Effects")
ansOptionPlot <- list("", "Plot A", "Plot B", "Plot C")
honey1 <- data.frame(
  Surplus = c(100, 60, 90, 85, 90, 95, 105, 70, 80),
  Varietal = c(rep("Clover", 3), rep("Orange Blossom", 3), rep("Alfalfa", 3))
)
honey2 <- data.frame(
  Surplus = c(50, 40, 55, 85, 80, 82, 105, 180, 192),
  Varietal = c(rep("Clover", 3), rep("Orange Blossom", 3), rep("Alfalfa", 3))
)
keyboarding1 <- data.frame(
  kbd.type = c(rep("1", 4), rep("2", 4), rep("3", 4)),
  hrs.kbd = c(60, 72, 61, 50, 54, 68, 66, 59, 51, 56, 55, 56),
  hrs.pain = c(85, 95, 69, 58, 41, 74, 71, 52, 34, 40, 41, 40),
  hrs.pain1 = c(85, 95, 69, 58, 41, 74, 71, 52, 34, 68, 41, 56)
)
keyboarding2 <- data.frame(
  kbd.type = c(rep("1", 4), rep("2", 4), rep("3", 4)),
  hrs.kbd = c(60, 72, 61, 50, 54, 68, 66, 59, 56, 56, 55, 29),
  hrs.pain = c(190, 200, 69, 58, 41, 54, 61, 52, 4, 2, 5, 120)
)
apex1 <- data.frame(
  officer = sort(c(rep(LETTERS[1:5], 4))),
  score = c(
    76, 65, 85, 74,
    59, 75, 81, 67,
    49, 63, 61, 46,
    74, 71, 85, 89,
    66, 84, 80, 79
  ),
  score1 = c(
    76, 65, 85, 74,
    59, 75, 81, 67,
    49, 63, 72, 46,
    74, 65, 85, 80,
    66, 84, 80, 79
  )
)
apex2 <- data.frame(
  officer = sort(c(rep(LETTERS[1:5], 4))),
  score = c(
    76, 65, 85, 74,
    5, 75, 81, 67,
    4, 63, 61, 46,
    74, 71, 85, 189,
    66, 84, 80, 79
  )
)
beer1 <- data.frame(
  judge = sort(rep(LETTERS[1:6],4)),
  beer = rep(c("Barnstormer", "King Richard Red",
               "Craftsman", "Red Mo"), 6),
  score = c(50, 60, 70, 70,
            38, 45, 58, 60,
            45, 48, 60, 58,
            65, 65, 75, 75,
            55, 60, 70, 65,
            48, 53, 68, 63)
)
beer2 <- data.frame(
  judge = sort(rep(LETTERS[1:6],4)),
  beer = rep(c("Barnstormer", "King Richard Red",
               "Craftsman", "Red Mo"), 6),
  score = c(50, 60, 70, 70,
            300, 2, 190, 6,
            45, 48, 60, 58,
            65, 65, 75, 75,
            55, 60, 70, 65,
            48, 53, 68, 63)
)

# Define the UI ----
ui <- list(
  dashboardPage(
    skin = "black",
    ## Header ----
    dashboardHeader(
      title = "Assumptions of ANOVA",
      titleWidth = 250,
      tags$li(class = "dropdown", actionLink("info", icon("info"))),
      tags$li(
        class = "dropdown",
        boastUtils::surveyLink(name = "Assumptions_of_ANOVA")
      ),
      tags$li(
        class = "dropdown",
        tags$a(
          id = "home",
          href = 'https://shinyapps.science.psu.edu/',
          icon("house")
          )
        )
    ),
    ## Sidebar menu ----
    dashboardSidebar(
      width = 250,
      sidebarMenu(
        id = "pages",
        menuItem("Overview", tabName = "overview", icon = icon("gauge-high")),
        menuItem("Prerequisites", tabName = "prerequisites", icon = icon("book")),
        menuItem("Explore", tabName = "explore", icon = icon("wpexplorer")),
        menuItem("Model Assumptions", tabName = "assumptiongame", icon = icon("gamepad")),
        menuItem("Assumption Checking", tabName = "checkinggame", icon = icon("gamepad")),
        menuItem("References", tabName = "references", icon = icon("leanpub"))
      ),
      tags$div(
        class = "sidebar-logo",
        boastUtils::sidebarFooter()
      )
    ),
    dashboardBody(
      tabItems(
        ## Overview Page ----
        tabItem(
          tabName = "overview",
          withMathJax(),
          h1("Assumptions of ANOVA Models"),
          p("This app introduces the assumptions for different ANOVA models and
            how to test those assumptions. You can also learn what it looks like when
            assumptions are violated."),
          h2("Instructions"),
          tags$ol(
            tags$li("If needed, check out the Prerequisites page to learn about
                    ANOVA and the general role of assumptions."),
            tags$li("Use the Explore page to dive into different ANOVA models and
                    contexts. You'll view and compare graphics illustrating the
                    when the different assumptions are and are not satisified."),
            tags$li("The Model Assumptions game allows you to test yourself on
                    assumptions pertain to which models."),
            tags$li("The Assumption Checking game provides an opportunity to test
                    yourself on decerning when different assumptions are and are
                    not satisified.")
          ),
          div(
            style = "text-align: center;",
            bsButton(
              inputId = "explore",
              label = "Explore",
              icon = icon("bolt"),
              size = "large")
          ),
          br(),
          br(),
          h2("Acknowledgements"),
          p("This version of the app was developed and coded by Gonghao Liu, Neil J.
            Hatfield, Robert P. Carey, III, Phichchaya Sutaporn and updated by Luqi
            Jiao Emanuele in 2023 and by Nathan Pechulis in 2024.",
            br(),
            br(),
            "Cite this app as:",
            br(),
            boastUtils::citeApp(),
            br(),
            br(),
            div(class = "updated", "Last Update: 07/08/2024 by NP.")
          )
        ),
        ## Prerequisites Page ----
        tabItem(
          tabName = "prerequisites",
          h2("Prerequisites"),
          p("Here you can learn the different ANOVA modeling techniques
            and how each one has assumptions that must be met. Click on the plus
            sign to the right to expand the topic."),
          box(
            title = strong("What is ANOVA?"),
            status = "primary",
            collapsible = TRUE,
            collapsed = TRUE,
            width = 12,
            p("In general, ANOVA refers to a statistical technique that focuses on
            breaking down data values into to components. In a designed study,
            these components reflect different sources of variation. We can analyze
            the variation explained by these sources to learn something about the
            phenomenon we're studying."),
            p("The data values we break down come from the quantiative response
            attribute (a.k.a. dependent variable). Among our sources of variation
            will be one or more qualitative factors (a.k.a. independent variables).
            Each specific model that we use with ANOVA has their own assumptions
            for conducting statistical testing.")
          ),
          box(
            title = tags$strong("Types of ANOVA Models"),
            status = "primary",
            collapsible = TRUE,
            collapsed = TRUE,
            width = 12,
            p("Here are a few of the many different ANOVA models."),
            tags$ul(
              tags$li(tags$strong("Oneway ANOVA:"), "A model where we have a
                      single factor and no other independent variables."),
              tags$li(tags$strong("k-way ANOVA:"), "An extension of the Oneway
                      ANOVA model where we have k different factors."),
              tags$li(tags$strong("ANCOVA:"), "A model where we have at least one
                      factor and at least one quantiative predictor. This model
                      blends elements of ANOVA and linear regression."),
              tags$li(tags$strong("Blocking:"), "A technique used in experimental
                      design to reduce the variability caused by nuisance attribute
                      (i.e., a source of variation that is not of primary interest).
                      This technique will add a block to our ANOVA model."),
              tags$li(tags$strong("Random Effects:"), "Whenever the levels of a
                      factor are themselves randomly sampled and we want to be
                      able to say something about all levels of the factor, we
                      have a random effect. (A fixed effect is when we have chosen
                      specific levels of the factor non-randomly.)"),
              tags$li(tags$strong("Repeated Measures:"), "A family of models that
                      hinge upon repeatedly measuring the response attribute for
                      each case.")
            )
          ),
          box(
            title = strong("Why is assumption testing important to ANOVA?"),
            status = "primary",
            collapsible = TRUE,
            collapsed = TRUE,
            width = 12,
            "ANOVA methods are primarily used for statistical inference. In order
            for the inference to be valid (regardless of whether we reject/fail
            to reject the null hypothesis), the assumptions which underpin that
            inference must be satisfied. If the assumptions aren't, then any
            p-values or confidence intervals we found don't mean what we want
            them to mean."
          ),
          box(
            title = strong("Why should we check the assumptions?"),
            status = "primary",
            collapsible = TRUE,
            collapsed = TRUE,
            width = 12,
            "By checking the assumptions before doing inference, we give ourselves
            the opportunity to shift gears. We can do this in a number of ways
            including but not limited to: using a different model, transforming
            the response, using a robust method, using randomization methods, and
            using nonparametric rank-based methods."
          )
        ),
        ## Explore Page ----
        tabItem(
          tabName = "explore",
          h2("Explore"),
          p("On this page, each ANOVA model tab contains a situation along with
            graphs that depict valid/invalid assumptions for that model. Utilize
            the dropdown box to change the assumption being tested. Note which
            assumptions belong to which model and what it looks like graphically
            when assumptions are violated."
          ),
          tabsetPanel(
            ### Oneway ANOVA ----
            tabPanel(
              title = "Oneway ANOVA",
              br(),
              h3("Scenario"),
              p("An experiment was conducted to determine the relationship between
                  honey output and types of flowers. There were 3 types of
                  flowers, 3 beehives were randomly assigned to each type of
                  flowers with 9 beehives in total."),
              fluidRow(
                column(
                  width = 4,
                  wellPanel(
                    selectInput(
                      "anovaSelect",
                      label = "Select the assumption you want to test",
                      choices = ansOptionAnovaAssumption,
                      width = NULL
                    )
                  )
                ),
                column(
                  width = 8,
                  offset = 0,
                  h4("Valid Example"),
                  textOutput("anovaTextValid"),
                  plotOutput("anovaImageValid"),
                  h4("Invalid Example"),
                  textOutput("anovaTextInvalid"),
                  plotOutput("anovaImageInvalid")
                )
              ),
            ),
            ### ANCOVA ----
            tabPanel(
              "ANCOVA",
              br(),
              h3("Scenario"),
              p("We are wanting to understand the impact of the type of keyboard
                  on how many hours of pain a person experiences in their hands,
                  wrists, and forearms."),
              p("We suspect that the number of hours a person spends keyboarding
                  is related to the number of hours of pain that they feel."),
              p("We have 12 volunteers who will use a specific keyboard we
                  assign them for 2 weeks. During that time, they will record the
                  number of hours they use the keyboard and the number of hours of
                  repetitive motion pain during the study period."),
              fluidRow(
                column(
                  width = 4,
                  wellPanel(
                    selectInput(
                      "ancovaSelect",
                      label = "Select the assumption you want to test",
                      choices = ansOptionAncovaAssumption
                    )
                  )
                ),
                column(
                  width = 8,
                  offset = 0,
                  h4("Valid Example"),
                  textOutput("ancovaTextValid"),
                  plotOutput("ancovaImageValid"),
                  h4("Invalid Example"),
                  textOutput("ancovaTextInvalid"),
                  plotOutput("ancovaImageInvalid")
                )
              )
            ),
            ### Blocking ----
            tabPanel(
              "Blocking",
              br(),
              h3("Scenario"),
              p("A farmer wants to test out four varieties of barley and see if
                there is any difference in yield. He has four fields in which he
                can plant the barley. However, the farmer is aware of differences
                between each field. For example,"),
              tags$ul(
                tags$li("One field has a higher clay content than the others."),
                tags$li("Another field has rockier soil than the others."),
                tags$li("One field has very loose soil while another field has
                        much more compact soil."),
                tags$li("Two fields are in wetter climates; two are in drier
                        climates.")
              ),
              fluidRow(
                column(
                  width = 4,
                  wellPanel(
                    selectInput(
                      "blockingSelect",
                      label = "Select the assumption you want to test",
                      choices = ansOptionBlockingAssumption,
                      width = NULL
                    )
                  )
                ),
                column(
                  width = 8,
                  offset = 0,
                  h4("Valid Example"),
                  textOutput("blockingTextValid"),
                  plotOutput("blockingImageValid"),
                  h4("Invalid Example"),
                  textOutput("blockingTextInvalid"),
                  plotOutput("blockingImageInvalid")
                )
              )
            ),
            ### Random Effects ----
            tabPanel(
              "Random Effects",
              br(),
              h3("Scenario"),
              p("Apex Enterprises is a company that builds roadside restaurants
                carrying one of several promoted trade names, leases franchises
                to individuals to operate the restaurants, and provides management
                services. This company employs a large number of personnel
                officers who interview applicants for jobs in the restaurants.
                At the end of the interview, the personnel officer assigns a
                rating between 0 to 100 to indicate the applicant's potential
                value on the job"),
              p("Apex would like to know two things: How great is the variation
                is in ratings among all personnel officers? What is the mean
                rating given by all personnel officers?"),
              fluidRow(
                column(
                  width = 4,
                  wellPanel(
                    selectInput(
                      "randomEffectSelect",
                      label = "Select the assumption you want to test",
                      choices = ansOptionRandomAssumption,
                      width = NULL
                    )
                  )
                ),
                column(
                  width = 8,
                  offset = 0,
                  h4("Valid Example"),
                  textOutput("randomEffectTextValid"),
                  plotOutput("randomEffectImageValid"),
                  h4("Invalid Example"),
                  textOutput("randomEffectTextInvalid"),
                  plotOutput("randomEffectImageInvalid")
                )
              )
            ),
            ### Repeated Measures ----
            # tabPanel(
            #   "Repeated Measure",
            #   br(),
            #   h3("Scenario"),
            #   p("Beer is big business; the craft brewing industry contributed
            #     $79.1 billion to the US Economy in 2018 and 550,000+ jobs
            #     (PA: $6.335 billion)."),
            #   p("Getting a craft beer scored can be quite the achievement. In
            #     a single blind tasting, judges are given a chilled, properly
            #     poured beer and told the style category. They then judge the
            #     beer on Aroma (24 pts), Appearance (6 pts), Flavor (40 pts),
            #     Mouthfeel (10 pts), and Overall Impression (20 pts)."),
            #   p("We have decided to put several State College beers to the test: "),
            #   tags$ul(
            #     tags$li("Barnstormer (IPA, Happy Valley Brewing Company)"),
            #     tags$li("Craftsman (Brown, Happy Valley Brewing Company)"),
            #     tags$li("Red Mo (Red, Otto's Pub and Brewery)"),
            #     tags$li("King Richard Red (Amber, Robin Hood Brewing Co.)"),
            #   ),
            #   fluidRow(
            #     column(
            #       width = 4,
            #       wellPanel(
            #         selectInput(
            #           "repeatedMeasureSelect",
            #           p("Select the assumption you want to test"),
            #           choices = ansOptionRepeatAssumption,
            #           width = NULL
            #         )
            #       )
            #     ),
            #     column(
            #       width = 8,
            #       offset = 0,
            #       h4("Valid Example"),
            #       textOutput("repeatedMeasureTextValid"),
            #       plotOutput("repeatedMeasureImageValid"),
            #       h4("Invalid Example"),
            #       textOutput("repeatedMeasureTextInvalid"),
            #       plotOutput("repeatedMeasureImageInvalid")
            #     )
            #   )
            # )
          )
        ),
        ## Model Assumptions Game Page ----
        tabItem(
          tabName = "assumptiongame",
          withMathJax(),
          h2("Model Assumptions Matching Game"),
          p("For this game, go through each ANOVA technique and match
            the correct assumptions for said technique using the drag
            and drop menu."),
          tabsetPanel(
            ### ANOVA Level ----
            tabPanel(
              "Oneway ANOVA",
              bucket_list(
                header = "Pick the assumptions for Oneway ANOVA.",
                add_rank_list(
                  text = "Drag assumptions from here",
                  input_id = "dragAnova",
                  labels = c("Normality",
                             "Homoscedasticity",
                             "Independence of Observations",
                             "Linear relationship covariate and the response",
                             "Equality of the covariate's slope parameter",
                             "No statistically significiant potential outliers",
                             "Interaction of block and treatment",
                             "Random effects")
                ),
                add_rank_list(
                  text = "to here",
                  input_id = "dropAnova"
                )
              ),
              fluidRow(
                column(
                  width = 1,
                  bsButton(
                    inputId = 'submitAnova',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 1,
                  offset = 1,
                  uiOutput('markAnova2')
                ),
                column(
                  width = 9,
                  offset = 0,
                  uiOutput('markAnova')
                )

              )
              ),
            ### ANCOVA Level ----
            tabPanel(
              "ANCOVA",
              bucket_list(
                header = "Pick the assumptions for ANCOVA.",
                add_rank_list(
                  text = "Drag assumptions from here",
                  input_id = "dragAncova",
                  labels = c("Normality",
                             "Homoscedasticity",
                             "Independence of Observations",
                             "Linear relationship covariate and the response",
                             "Equality of the covariate's slope parameter",
                             "No statistically significiant potential outliers",
                             "Interaction of block and treatment",
                             "Random effects")
                ),
                add_rank_list(
                  text = "to here",
                  input_id = "dropAncova"
                )
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitAncova',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 1,
                  offset = 1,
                  uiOutput('markAncova2')
                ),
                column(
                  width = 9,
                  offset = 0,
                  uiOutput('markAncova')
                )
              )
            ),
            ### Blocking Level ----
            tabPanel(
              "Blocking",
              bucket_list(
                header = "Pick the assumptions for Blocking.",
                add_rank_list(
                  text = "Drag assumptions from here",
                  input_id = "dragBlocking",
                  labels = c("Normality",
                             "Homoscedasticity",
                             "Independence of Observations",
                             "Linear relationship covariate and the response",
                             "Equality of the covariate's slope parameter",
                             "No statistically significiant potential outliers",
                             "Interaction of block and treatment",
                             "Random effects")
                ),
                add_rank_list(
                  text = "to here",
                  input_id = "dropBlocking"
                )
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitBlocking',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 1,
                  offset = 1,
                  uiOutput('markBlocking2')
                ),
                column(
                  width = 9,
                  offset = 0,
                  uiOutput('markBlocking')
                )
              )
            ),
            ### Random Effects Level ----
            tabPanel(
              "Random Effects",
              bucket_list(
                header = "Pick the assumptions for Random Effects.",
                add_rank_list(
                  text = "Drag assumptions from here",
                  input_id = "dragRandomEffect",
                  labels = c("Normality",
                             "Homoscedasticity",
                             "Independence of Observations",
                             "Linear relationship covariate and the response",
                             "Equality of the covariate's slope parameter",
                             "No statistically significiant potential outliers",
                             "Interaction of block and treatment",
                             "Random effects")
                ),
                add_rank_list(
                  text = "to here",
                  input_id = "dropRandomEffect"
                )
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitRandomEffect',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 1,
                  offset = 1,
                  uiOutput('markRandomEffect2')
                ),
                column(
                  width = 9,
                  offset = 0,
                  uiOutput('markRandomEffect')
                )
              )
            ),
            ### Repeated Measures Level ----
            # tabPanel(
            #   "Repeated Measure",
            #   bucket_list(
            #     header = "Pick the assumptions for Repeated Measures.",
            #     add_rank_list(
            #       text = "Drag assumptions from here",
            #       input_id = "dragRepeatedMeasure",
            #       labels = c("Normality",
            #                  "Homoscedasticity",
            #                  "Independence of Observations",
            #                  "Linear relationship covariate and the response",
            #                  "Equality of the covariate's slope parameter",
            #                  "No statistically significiant potential outliers",
            #                  "Interaction of block and treatment",
            #                  "Random effects")
            #     ),
            #     add_rank_list(
            #       text = "to here",
            #       input_id = "dropRepeatedMeasure"
            #     )
            #   ),
            #   fluidRow(
            #     column(
            #       width = 1,
            #       offset = 0,
            #       bsButton(
            #         inputId = 'submitRepeatedMeasure',
            #         label = "Submit",
            #         size = "large",
            #         style = "default",
            #         disabled = FALSE
            #       )
            #     ),
            #     column(
            #       width = 1,
            #       offset = 1,
            #       uiOutput('markRepeatedMeasure2')
            #     ),
            #     column(
            #       width = 9,
            #       offset = 0,
            #       uiOutput('markRepeatedMeasure')
            #     )
            #   )
            # )
          )
        ),
        ## Assumption Checking Game Page ----
        tabItem(
          tabName = "checkinggame",
          withMathJax(),
          h2("Assumption Checking Game"),
          p("Go through each tab and select the plot that violates
            the assumption based on what you've learned so far.
            Try to get the correct answer for each assumption."),
          tabsetPanel(
            ### Normality Level ----
            tabPanel(
              "Normality",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('normalityGamePlot1')
                ),
                column(
                  width = 4,
                  plotOutput('normalityGamePlot3')
                ),
                column(
                  width = 4,
                  plotOutput('normalityGamePlot2')
                )
              ),
              selectInput(
                "nomalitySelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  bsButton(
                    inputId = 'submitNormality',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markNormality')
                )
              )
            ),
            ### Homoscedasticity Level ----
            tabPanel(
              "Homoscedasticity",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('homoGamePlot1')
                ),
                column(
                  width = 4,
                  plotOutput('homoGamePlot2')
                ),
                column(
                  width = 4,
                  plotOutput('homoGamePlot3')
                )
              ),
              selectInput(
                "homoSelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitHomo',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markHomo')
                )
              )
            ),
            ### Independent Obs. Level ----
            tabPanel(
              "Independence of Observationss",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('indeGamePlot3')
                ),
                column(
                  width = 4,
                  plotOutput('indeGamePlot2')
                ),
                column(
                  width = 4,
                  plotOutput('indeGamePlot1')
                )
              ),
              selectInput(
                "indeSelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitInde',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markInde')
                )
              )
            ),
            ### Linearity Level ----
            tabPanel(
              "Linearity",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('linearGamePlot3')
                ),
                column(
                  width = 4,
                  plotOutput('linearGamePlot2')
                ),
                column(
                  width = 4,
                  plotOutput('linearGamePlot1')
                )
              ),
              selectInput(
                "linearSelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitLinear',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markLinear')
                )
              )
            ),
            ### Common Slope Level ----
            tabPanel(
              "Common Slope",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('slopeGamePlot1')
                ),
                column(
                  width = 4,
                  plotOutput('slopeGamePlot3')
                ),
                column(
                  width = 4,
                  plotOutput('slopeGamePlot2')
                )
              ),
              selectInput(
                "slopeSelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitSlope',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markSlope')
                )
              )
            ),
            ### Potential Outliers Level ----
            tabPanel(
              "No Potential Outliers",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('outGamePlot3')
                ),
                column(
                  width = 4,
                  plotOutput('outGamePlot2')
                ),
                column(
                  width = 4,
                  plotOutput('outGamePlot1')
                )
              ),
              selectInput(
                "outSelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitOut',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markOut')
                )
              )
            ),
            ### Interaction of Block Level ----
            tabPanel(
              "Interaction of Block",
              fluidRow(
                column(
                  width = 4,
                  plotOutput('interGamePlot1')
                ),
                column(
                  width = 4,
                  plotOutput('interGamePlot3')
                ),
                column(
                  width = 4,
                  plotOutput('interGamePlot2')
                )
              ),
              selectInput(
                "interSelected",
                label = 'Your choice',
                choices = ansOptionPlot
              ),
              fluidRow(
                column(
                  width = 1,
                  offset = 0,
                  bsButton(
                    inputId = 'submitInter',
                    label = "Submit",
                    size = "large",
                    style = "default",
                    disabled = FALSE
                  )
                ),
                column(
                  width = 10,
                  offset = 1,
                  uiOutput('markInter')
                )
              )
            )
          )
        ),
        ## References Page ----
        tabItem(
          tabName = "references",
          withMathJax(),
          h2("References"),
          p(class = "hangingindent",
            "Bailey, E. (2015). shinyBS: Twitter bootstrap components for shiny.
            (v0.61). [R package]. Available from
            https://CRAN.R-project.org/package=shinyBS"),
          p(class = "hangingindent",
            "Bates, D., Maechler, M., Bolker, B., Walker, S. (2015). Fitting
            Linear Mixed-Effects Models Using lme4. Journal of Statistical Software,
            67(1), 1-48. doi:10.18637/jss.v067.i01."),
          p(class = "hangingindent",
            "Carey, R. (2019). boastUtils: BOAST Utilities. (v0.1.0).
            [R Package]. Available from
            https://github.com/EducationShinyAppTeam/boastUtils"),
          p(class = "hangingindent",
            "Chang, W. and Borges, R. B. (2018). shinydashboard: Create
            dashboards with 'Shiny'. (v0.7.1) [R Package]. Available from
            https://CRAN.R-project.org/package=shinydashboard"),
          p(class = "hangingindent",
            "Chang, W., Cheng, J., Allaire, J., Xie, Y., and McPherson, J.
            (2019). shiny: Web application framework for R. (v1.4.0)
            [R Package]. Available from https://CRAN.R-project.org/package=shiny"),
          p(class = "hangingindent",
            "de Vries, A., Schloerke, B. and Russell, K. (2019).
            sortable: Drag-and-Drop in 'shiny' Apps with 'SortableJS'. R package
            version 0.4.2. Avaliable from: https://CRAN.R-project.org/package=sortable"),
          p(class = "hangingindent",
            "Fox, J. and Weisberg, S. (2019). An {R} Companion to Applied
            Regression, Third Edition. Thousand Oaks CA: Sage. Avaliable from:
            https://socialsciences.mcmaster.ca/jfox/Books/Companion/"),
          p(class = "hangingindent",
            "Hatfield, N. J. (2020), Stat 461: ANOVA Course Notes [course notes], Spring 2020."),
          p(class = "hangingindent",
            "Kassambara, A. (2020). rstatix: Pipe-Friendly Framework for Basic
            Statistical Tests. R package version 0.6.0.
            https://CRAN.R-project.org/package=rstatix"),
          p(class = "hangingindent",
            "Kutner, M. H., Nachtsheim, C. J., Neter, J., and Li, W. (2005),
            Applied Linear Statistical Models [apex enterprises data set],
            New York: McGraw-Hill Irwin"),
          p(class = "hangingindent",
            "Oehlert, G. W. (2000), A First Course in Design and Analysis of Experiments
            [keyboarding data set], New York: W. H. Freeman."),
          p(class = "hangingindent",
            "Perrier, V., Meyer, F. and Granjon, D. (2020). shinyWidgets:
            Custom Inputs Widgets for Shiny. R package version 0.5.3. Avaliable from:
            https://CRAN.R-project.org/package=shinyWidgets"),
          p(class = "hangingindent",
            "Wickham, W. (2016). ggplot2: Elegant graphics for data analysis.
            [R Package]. Springer-Verlag New York. Available from
            https://ggplot2.tidyverse.org") ,
          br(),
          br(),
          br(),
          boastUtils::copyrightInfo()
          )
        )
      )
    )
  )

# Define the server ----
server <- function(input, output, session) {
  ## Set Up "Explore" Button ----
  observeEvent(
    eventExpr = input$explore,
    handlerExpr = {
      updateTabItems(
        session = session,
        inputId = "pages",
        selected = "explore"
      )
    }
  )

  ## Set up info button ----
  observeEvent(
    eventExpr = input$info,
    handlerExpr = {
      sendSweetAlert(
        session = session,
        title = "Instructions",
        text = "Learn the assumptions for each model and how to exam them. Also,
               test yourself with the drag and drop and multiple choice games.",
        type = "info"
      )
    }
  )

  ## Explore Page Elements ----

  ### ANOVA ----
  #### Text Messages ----
  #### This is a test of a new coding format
  observe(
    x = {
      validMessage <- switch(
        EXPR = input$anovaSelect,
        "Normality of Residuals" = "In this plot, the envelope should contain almost all the points
            in the graph.",
        "Homoscedasticity" = "The points here should have similar variability for each x value.",
        "Independence of Observations" = "The points in this graph should have no pattern."
      )

      invalidMessage <- switch(
        EXPR = input$anovaSelect,
        "Normality of Residuals" = "In this plot, 1/9th (~11%) of the cases are located outside 97% envelope.",
        "Homoscedasticity" = "The points in this graph tend to have different amounts of variation
            for different x values.",
        "Independence of Observations" = "The points in this graph tend to have a pattern."
      )

      output$anovaTextValid <- renderText(expr = validMessage)
      output$anovaTextInvalid <- renderText(expr = invalidMessage)
    }
  ) %>%
    bindEvent(input$anovaSelect)

  #### Valid Plot ----
  observeEvent(
    eventExpr = input$anovaSelect,
    handlerExpr = {
      output$anovaImageValid <- renderPlot(
        expr = {
          if (input$anovaSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = honey1$Surplus,
              distribution = "norm",
              envelope = 0.97,
              ylab = "Surplus Honey (lbs)"
            )
          }
          else if (input$anovaSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              Surplus ~ Varietal,
              vertical = TRUE,
              data = honey1,
              cex.lab = 1.5,
              cex.axis = 1.5
            )
          }
          else if (input$anovaSelect == "Independence of Observations") {
            plot(
              honey1$Surplus,
              type = "b",
              ylab = "Surplus Honey (lbs)",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
        },
        alt = if (input$anovaSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with all the points lay in the
          97% confidence envelope."
          } else if (input$anovaSelect == "Homoscedasticity") {
            "For homoscedasticity, this is a plot of response versus explanatory,all the
            points lay in a random position."
          } else if (input$anovaSelect == "Independence of Observations") {
            "For independence, this is a plot of response versus index, all the points
            lay in a random position."
          }
      )
    }
  )

  #### Invalid Plot ----
  observeEvent(
    eventExpr = input$anovaSelect,
    handlerExpr = {
      output$anovaImageInvalid <- renderPlot(
        expr = {
          if (input$anovaSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = honey2$Surplus,
              distribution = "norm",
              envelope = 0.97,
              ylab = "Surplus Honey (lbs)")
          }
          else if (input$anovaSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              Surplus ~ Varietal,
              vertical = TRUE,
              data = honey2,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$anovaSelect == "Independence of Observations") {
            plot(
              honey2$Surplus,
              type = "b",
              ylab = "Surplus Honey (lbs)",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
        },
        alt = if (input$anovaSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with 2 points lay out side of
          the 97% confidence envelope."
          } else if (input$anovaSelect == "Homoscedasticity") {
           "For homoscedasticity, this is a plot of response versus explanatory,all the
            points lay in a pattern."
          } else if (input$anovaSelect == "Independence of Observations") {
            "For independence, this is a plot of response versus index, all the points
            lay in a pattern."
          }
      )
    }
  )

  ### ANCOVA ----
  #### Text Messages ----
  #### This is a test of a new coding format
  observe(
    x = {
      validMessage <- switch(
        EXPR = input$ancovaSelect,
        "Normality of Residuals" = "In this plot, the envelope should contain almost all the points
            in the graph.",
        "Homoscedasticity" = "The points here should have similar variability for each x value.",
        "Independence of Observations" = "The points in this graph should have no pattern.",
        "Linear Relationship covariate and the Response" = "By the graph, we expect to see a linear relationship between covariate
            and response.",
        "Equality of the covariate's Slope parameter" = "The different covariates represented by different colors all have
            similar slopes.",
        "No Potential Outliers" = "We expected no visualized outliers."
      )

      invalidMessage <- switch(
        EXPR = input$ancovaSelect,
        "Normality of Residuals" = "In this plot, too many points are located outside of the envelope.",
        "Homoscedasticity" = "The points in this graph tend to have different amounts of variation
            for different x values.",
        "Independence of Observations" = "The points in this graph tend to have a pattern.",
        "Linear Relationship covariate and the Response" = "There is no linear relationship between covariate and response in
            the graph.",
        "Equality of the covariate's Slope parameter" = "The different covariates represented by different colors don't
            all have similar slopes.",
        "No Potential Outliers" = "There are some visualized outliers in the plot."
      )

      output$ancovaTextValid <- renderText(expr = validMessage)
      output$ancovaTextInvalid <- renderText(expr = invalidMessage)
    }
  ) %>%
    bindEvent(input$ancovaSelect)

  #### Valid Plot ----
  observeEvent(
    eventExpr = input$ancovaSelect,
    handlerExpr = {
      output$ancovaImageValid <- renderPlot(
        expr = {
          if (input$ancovaSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = keyboarding1$hrs.pain,
              distribution = "norm",
              envelope = 0.97,
              ylab = "Hours of Pain"
            )
          }
          else if (input$ancovaSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              hrs.pain1 ~ kbd.type,
              vertical = TRUE,
              data = keyboarding1,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$ancovaSelect == "Independence of Observations") {
            plot(
              keyboarding1$hrs.pain,
              type = "b",
              ylab = "Hours of Pain",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$ancovaSelect == "Linear Relationship covariate and the Response") {
            ggplot(
              data = keyboarding1,
              mapping = aes(y = hrs.pain, x = hrs.kbd)
            ) +
              geom_point(size = 3) +
              geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Hours Spent Keyboarding") +
              ylab("Hours of Pain")
          }
          else if (input$ancovaSelect == "Equality of the covariate's Slope parameter") {
            ggplot(
              data = keyboarding1,
              mapping = aes(
                y = hrs.pain,
                x = hrs.kbd,
                group = kbd.type,
                color = kbd.type,
                shape = kbd.type
              )
            ) +
              geom_point(size = 3) +
              geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Hours Spent Keyboarding") +
              ylab("Hours of Pain") +
              labs(color = "Keyboard Type")

          }
          else if (input$ancovaSelect == "No Potential Outliers") {
            key2 <- rstatix::mahalanobis_distance(keyboarding1)
            key2 <- cbind(key2, factor = keyboarding1$kbd.type)
            ggplot(
              data = key2,
              mapping = aes(y = hrs.pain, x = hrs.kbd, color = factor)
            ) +
              geom_point(size = 3) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Hours Spent Keyboarding") +
              ylab("Hours of Pain") +
              labs(color = "Keyboard", shape = "Potential Outlier")
          }
        },
        alt = if (input$ancovaSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with all the points lay in the
          97% confidence envelope."
        } else if (input$ancovaSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all the
          points lay in a random position."
        } else if (input$ancovaSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points lay
          in a random position."
        } else if (input$ancovaSelect == "Linear Relationship covariate and the Response") {
          "For linear, the plot shows that response and explantory have a linear relationship."
        } else if (input$ancovaSelect == "Equality of the covariate's Slope parameter") {
          "For slope, points in the different groups have a similar trend."
        }  else if (input$ancovaSelect == "No Potential Outliers") {
          "For outlier, the plot shows that the data don't have obvious outliers."
        }
      )
    }
  )

  #### Invalid Plot ----
  observeEvent(
    eventExpr = input$ancovaSelect,
    handlerExpr = {
      output$ancovaImageInvalid <- renderPlot(
        expr = {
          if (input$ancovaSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = keyboarding2$hrs.pain,
              distribution = "norm",
              envelope = 0.97,
              ylab = "Hours of Pain"
            )
          }
          else if (input$ancovaSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              hrs.pain ~ kbd.type,
              vertical = TRUE,
              data = keyboarding2,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$ancovaSelect == "Independence of Observations") {
            plot(
              keyboarding2$hrs.pain,
              type = "b",
              ylab = "Hours of Pain",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$ancovaSelect == "Linear Relationship covariate and the Response") {
            ggplot(
              data = keyboarding2,
              mapping = aes(y = hrs.pain, x = hrs.kbd)
            ) +
              geom_point(size = 3) +
              geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Hours Spent Keyboarding") +
              ylab("Hours of Pain")
          }
          else if (input$ancovaSelect == "Equality of the covariate's Slope parameter") {
            ggplot(
              data = keyboarding2,
              mapping = aes(
                y = hrs.pain,
                x = hrs.kbd,
                group = kbd.type,
                color = kbd.type,
                shape = kbd.type
              )
            ) +
              geom_point(size = 3) +
              geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Hours Spent Keyboarding") +
              ylab("Hours of Pain") +
              labs(color = "Keyboard Type")
          }
          else if (input$ancovaSelect == "No Potential Outliers") {
            key2 <- rstatix::mahalanobis_distance(keyboarding2)
            key2 <- cbind(key2, factor = keyboarding2$kbd.type)
            ggplot(
              data = key2,
              mapping = aes(y = hrs.pain, x = hrs.kbd, color = factor)
            ) +
              geom_point(size = 3) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Hours Spent Keyboarding") +
              ylab("Hours of Pain") +
              labs(color = "Keyboard", shape = "Potential Outlier")
          }
        },
        alt = if (input$ancovaSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with 2 points lay in the 97%
          confidence envelope."
        } else if (input$ancovaSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all the
          points lay in a pattern."
        } else if (input$ancovaSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points lay
          in a pattern."
        } else if (input$ancovaSelect == "Linear Relationship covariate and the Response") {
          "For linear, the plot shows that response and explantory have no linear relationship."
        } else if (input$ancovaSelect == "Equality of the covariate's Slope parameter") {
          "For slope, points in the different groups have a different trend."
        } else if (input$ancovaSelect == "No Potential Outliers") {
          "For outlier, the plot shows that the data have obvious outliers."
        }
      )
    }
  )

  ### Blocking ----
  #### Text Messages ----
  #### This is a test of a new coding format
  observe(
    x = {
      validMessage <- switch(
        EXPR = input$blockingSelect,
        "Normality of Residuals" = "In this plot, the envelope should contain almost all the points
            in the graph.",
        "Homoscedasticity" = "The points here should have similar variability for each x value.",
        "Independence of Observations" = "The points in this graph should have no pattern.",
        "Interaction of Block and Treatment" = "The data in different groups should have a similar pattern."
      )

      invalidMessage <- switch(
        EXPR = input$blockingSelect,
        "Normality of Residuals" = "In this plot, too many points are located outside of the envelope.",
        "Homoscedasticity" = "The points in this graph tend to have different amounts of variation
            for different x values.",
        "Independence of Observations" = "The points in this graph tend to have a pattern.",
        "Interaction of Block and Treatment" = "The data in different groups here have different patterns."
      )

      output$blockingTextValid <- renderText(expr = validMessage)
      output$blockingTextInvalid <- renderText(expr = invalidMessage)
    }
  ) %>%
    bindEvent(input$blockingSelect)

  #### Valid Plot ----
  observeEvent(
    eventExpr = input$blockingSelect,
    handlerExpr = {
      output$blockingImageValid <- renderPlot(
        expr = {
          barleyModel <- aov(Yield ~ Treatment + Field, data = barley1)
          if (input$blockingSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = barleyModel$residuals,
              distribution = "norm",
              envelope = 0.97,
              ylab = "Yield (bushels per arce)"
            )
          }
          else if (input$blockingSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              Yield ~ Treatment,
              vertical = TRUE,
              data = barley2,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$blockingSelect == "Independence of Observations") {
            plot(
              barley1$Yield,
              type = "b",
              ylab = "Yield (bushels per acre)",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$blockingSelect == "Interaction of Block and Treatment") {
            ggplot(
              data = barley1,
              mapping = aes(
                x = Treatment,
                y = Yield,
                color = Field,
                group = Field
              )
            ) +
              geom_point(size = 2) +
              geom_line(linewidth = 1) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Variety") +
              ylab("Yield (bushels per acre)") +
              labs(color = "Field")
          }
        },
        alt = if (input$blockingSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with all the points lay in
          the 97% confidence envelope."
        } else if (input$blockingSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all
          the points lay in a random position."
        }else if (input$blockingSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points
          lay in a random position."
        } else if (input$blockingSelect == "Interaction of Block and Treatment") {
          "For interaction, points in the different block have a similar trend."
        }
      )
    }
  )

  #### Invalid Plot ----
  observeEvent(
    eventExpr = input$blockingSelect,
    handlerExpr = {
      output$blockingImageInvalid <- renderPlot(
        expr = {
          barleyModel <- aov(Yield ~ Treatment + Field, data = barley2)
          if (input$blockingSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = barleyModel$residuals,
              distribution = "norm",
              envelope = 0.5,
              ylab = "Yield (bushels per arce)"
            )
          }
          else if (input$blockingSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              Yield ~ Treatment,
              vertical = TRUE,
              data = barley1,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$blockingSelect == "Independence of Observations") {
            plot(
              barley2$Yield,
              type = "b",
              ylab = "Yield (bushels per acre)",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$blockingSelect == "Interaction of Block and Treatment") {
            ggplot(
              data = barley3,
              mapping = aes(
                x = Treatment,
                y = Yield,
                color = Field,
                group = Field
              )
            ) +
              geom_point(size = 2) +
              geom_line(linewidth = 1) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              xlab("Variety") +
              ylab("Yield (bushels per acre)") +
              labs(color = "Field")
          }
        },
        alt = if (input$blockingSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with 2 points lay in
          the 97% confidence envelope."
        } else if (input$blockingSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all
          the points lay in a pattern."
        } else if (input$blockingSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points
          lay in a pattern."
        } else if (input$blockingSelect == "Interaction of Block and Treatment") {
          "For interaction, points in the different block have a different trend."
        }
      )
    }
  )

  ### Random Effects ----
  #### Text Messages ----
  #### This is a test of a new coding format
  observe(
    x = {
      validMessage <- switch(
        EXPR = input$randomEffectSelect,
        "Normality of Residuals" = "In this plot, the envelope should contain almost all the points
            in the graph.",
        "Homoscedasticity" = "The points here should have similar variability for each x value.",
        "Independence of Observations" = "The points in this graph should have no pattern.",
        "Random Effects" = "In this plot, the envelope should contain almost all the points
            in the graph."
      )

      invalidMessage <- switch(
        EXPR = input$randomEffectSelect,
        "Normality of Residuals" = "In this plot, too many points are located outside of the envelope.",
        "Homoscedasticity" = "The points in this graph tend to have different amounts of variation
            for different x values.",
        "Independence of Observations" = "The points in this graph tend to have a pattern.",
        "Random Effects" = "In this plot, too many points are located outside of the envelope."
      )

      output$randomEffectTextValid <- renderText(expr = validMessage)
      output$randomEffectTextInvalid <- renderText(expr = invalidMessage)
    }
  ) %>%
    bindEvent(input$randomEffectSelect)

  #### Valid Plot ----
  observeEvent(
    eventExpr = input$randomEffectSelect,
    handlerExpr = {
      output$randomEffectImageValid <- renderPlot(
        expr = {
          options("contrasts" = c("contr.sum","contr.poly"))
          apexFE <- aov(score ~ officer, data = apex1)
          apexRE <- lme4::lmer(
            score ~ (1|officer),
            data = apex1,
            REML = TRUE,
            subset = NULL,
            weights = NULL,
            na.action = NULL,
            offset = NULL)
          if (input$randomEffectSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = resid(apexRE),
              distribution = "norm",
              envelope = 0.92,
              ylab = "Score of Applicant",
              main = "Residuals"
            )
          }
          else if (input$randomEffectSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              score1 ~ officer,
              vertical = TRUE,
              data = apex1,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$randomEffectSelect == "Independence of Observations") {
            plot(
              apex1$score,
              type = "b",
              ylab = "Score of Applicant",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$randomEffectSelect == "Random Effects") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = lme4::ranef(apexRE)$officer[, "(Intercept)"],
              distribution = "norm",
              envelope = 0.8,
              ylab = "Score of Applicant",
              main = "Random Effects"
            )
          }
        },
        alt = if (input$randomEffectSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with all the points lay in
          the 97% confidence envelope."
        } else if (input$randomEffectSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all
          the points lay in a random position."
        } else if (input$randomEffectSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points
          lay in a random position."
        } else if (input$randomEffectSelect == "Random Effects") {
          "For random, this is a plot of norm quantile with all the points lay in the
          80% confidence envelope."
        }
      )
    }
  )

  #### Invalid Plot ----
  observeEvent(
    eventExpr = input$randomEffectSelect,
    handlerExpr = {
      output$randomEffectImageInvalid <- renderPlot(
        expr = {
          options("contrasts" = c("contr.sum","contr.poly"))
          apexFE <- aov(score ~ officer, data = apex2)
          apexRE <- lme4::lmer(
            score ~ (1|officer),
            data = apex2,
            REML = TRUE,
            subset = NULL,
            weights = NULL,
            na.action = NULL,
            offset = NULL)
          if (input$randomEffectSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = resid(apexRE),
              distribution = "norm",
              envelope = 0.92,
              ylab = "Score of Applicant",
              main = "Residuals"
            )
          }
          else if (input$randomEffectSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              score ~ officer,
              vertical = TRUE,
              data = apex2,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$randomEffectSelect == "Independence of Observations") {
            plot(
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              apex2$score,
              type = "b",
              ylab = "Score of Applicant",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$randomEffectSelect == "Random Effects") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = lme4::ranef(apexRE)$officer[, "(Intercept)"],
              distribution = "norm",
              envelope = 0.2,
              ylab = "Score of Applicant",
              main = "Random Effects"
            )
          }
        },
        alt = if (input$randomEffectSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with 2 points lay in
          the 97% confidence envelope."
        } else if (input$randomEffectSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all
          the points lay in a pattern."
        } else if (input$randomEffectSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points
          lay in a pattern."
        } else if (input$randomEffectSelect == "Random Effects") {
          "For random, this is a plot of norm quantile with 1 point lay in the
          80% confidence envelope."
        }
      )
    }
  )

  ### Repeated Measures ----
  #### Text Messages ----
  #### This is a test of a new coding format
  observe(
    x = {
      validMessage <- switch(
        EXPR = input$repeatedMeasureSelect,
        "Normality of Residuals" = "In this plot, the envelope should contain almost all the points
            in the graph.",
        "Homoscedasticity" = "The points here should have similar variability for each x value.",
        "Independence of Observations" = "The points in this graph should have no pattern.",
        "Interaction of Block and Treatment" = "The data in different groups should have a similar pattern."
      )

      invalidMessage <- switch(
        EXPR = input$repeatedMeasureSelect,
        "Normality of Residuals" = "In this plot, too many points are located outside of the envelope.",
        "Homoscedasticity" = "The points in this graph tend to have different amounts of variation
            for different x values.",
        "Independence of Observations" = "The points in this graph tend to have a pattern.",
        "Interaction of Block and Treatment" = "The data in different groups here have different patterns."
      )

      output$repeatedMeasureTextValid <- renderText(expr = validMessage)
      output$repeatedMeasureTextInvalid <- renderText(expr = invalidMessage)
    }
  ) %>%
    bindEvent(input$repeatedMeasureSelect)

  ##### Original valid messages  ----
  output$repeatedMeasureTextValid <- renderText({
    if (input$repeatedMeasureSelect == "Normality of Residuals") {
      paste("In this plot, the boundary line should envelop almost all the points
            in the graph.")
    }
    else if (input$repeatedMeasureSelect == "Homoscedasticity") {
      paste("The points here should have similar variability for each x value.")
    }
    else if (input$repeatedMeasureSelect == "Independence of Observations") {
      paste("The points in this graph should have no pattern.")
    }
    else if (input$repeatedMeasureSelect == "Interaction of Block and Treatment") {
      paste("The data in different group should have same pattern.")
    }
    else if (input$randomEffectSelect == "Random Effects") {
      paste("In this plot, the boundary line should envelop almost all the points
            in the graph.")
    }
  })

  ##### Original invalid messages ----
  output$repeatedMeasureTextInvalid <- renderText({
    if (input$repeatedMeasureSelect == "Normality of Residuals") {
      paste("In this plot, too many points are located outside of the envelope.")
    }
    else if (input$repeatedMeasureSelect == "Homoscedasticity") {
      paste("The points in this graph tend to have different amounts of variation
            for different x values.")
    }
    else if (input$repeatedMeasureSelect == "Independence of Observations") {
      paste("The points in this graph tend to have a pattern.")
    }
    else if (input$repeatedMeasureSelect == "Random Effects") {
      paste("In this plot, too many points are located outside of the envelope.")
    }
    else if (input$repeatedMeasureSelect == "Interaction of Block and Treatment") {
      paste("The data in different groups here have different patterns.")
    }
  })

  #### Valid Plot ----
  observeEvent(
    eventExpr = input$repeatedMeasureSelect,
    handlerExpr = {
      output$repeatedMeasureImageValid <- renderPlot(
        expr = {
          beerM1 <- lme4::lmer(
            score ~ beer + (1|judge),
            data = beer1,
            subset = NULL,
            weights = NULL,
            na.action = NULL,
            offset = NULL)
          if (input$repeatedMeasureSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = residuals(beerM1),
              distribution = "norm",
              envelope = 0.9,
              ylab = "Score"
            )
          }
          else if (input$repeatedMeasureSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              score ~ beer,
              vertical = TRUE,
              data = beer1,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$repeatedMeasureSelect == "Independence of Observations") {
            plot(
              beer1$score,
              type = "b",
              ylab = "Score",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$repeatedMeasureSelect == "Interaction of Block and Treatment") {
            ggplot(data = beer1,
                   mapping = aes(x = beer,
                                 y = score,
                                 color = judge,
                                 group = judge)) +
              geom_point(size = 2) +
              geom_line(linewidth = 1) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              viridis::scale_color_viridis(discrete = TRUE, option = "viridis") +
              xlab("Beer") +
              ylab("Score") +
              labs(color = "Judge")
          }
          else if (input$repeatedMeasureSelect == "Random Effects") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = lme4::ranef(beerM1)$judge[, "(Intercept)"],
              distribution = "norm",
              envelope = 0.9,
              ylab = "score"
            )
          }
        },
        alt = if (input$repeatedMeasureSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with all the points lay in
      the 97% confidence envelope."}
        else if (input$repeatedMeasureSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all
      the points lay in a random position"}
        else if (input$repeatedMeasureSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points
      lay in a random position"}
        else if (input$repeatedMeasureSelect == "Interaction of Block and Treatment") {
          "For interaction, points in the different block have a similar trend."}
        else if (input$repeatedMeasureSelect == "Random Effects") {
          "For random, this is a plot of norm quantile with all the points lay in the
    80% confidence envelope."})
    })

  #### Invalid Plot ----
  observeEvent(
    eventExpr = input$repeatedMeasureSelect,
    handlerExpr = {
      output$repeatedMeasureImageInvalid <- renderPlot(
        expr = {
          beerM1 <- lme4::lmer(
            score ~ beer + (1|judge),
            data = beer2,
            subset = NULL,
            weights = NULL,
            na.action = NULL,
            offset = NULL)
          if (input$repeatedMeasureSelect == "Normality of Residuals") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = residuals(beerM1),
              distribution = "norm",
              envelope = 0.9,
              ylab = "Score"
            )
          }
          else if (input$repeatedMeasureSelect == "Homoscedasticity") {
            stripchart(
              pch = 19,
              cex = 1.5,
              score ~ beer,
              vertical = TRUE,
              data = beer2,
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$repeatedMeasureSelect == "Independence of Observations") {
            plot(
              beer2$score,
              type = "b",
              ylab = "Score",
              pch = 19,
              cex = 1.5,
              xlab = "Row index",
              cex.lab = 1.5,
              cex.axis = 1.5)
          }
          else if (input$repeatedMeasureSelect == "Interaction of Block and Treatment") {
            ggplot(data = beer2,
                   mapping = aes(x = beer,
                                 y = score,
                                 color = judge,
                                 group = judge)) +
              geom_point(size = 2) +
              geom_line(linewidth = 1) +
              theme_bw() +
              theme(axis.title = element_text(size = 18)) +
              viridis::scale_color_viridis(discrete = TRUE, option = "viridis") +
              xlab("Beer") +
              ylab("Score") +
              labs(color = "Judge")
          }
          else if (input$repeatedMeasureSelect == "Random Effects") {
            car::qqPlot(
              pch = 19,
              cex = 1.5,
              id = FALSE,
              x = lme4::ranef(beerM1)$judge[, "(Intercept)"],
              distribution = "norm",
              envelope = 0.3,
              ylab = "score"
            )
          }
        },
        alt = if (input$repeatedMeasureSelect == "Normality of Residuals") {
          "For normality, this is a plot of norm quantile with 2 points lay in
      the 97% confidence envelope."}
        else if (input$repeatedMeasureSelect == "Homoscedasticity") {
          "For homoscedasticity, this is a plot of response versus explanatory, all
      the points lay in a pattern."}
        else if (input$repeatedMeasureSelect == "Independence of Observations") {
          "For independence, this is a plot of response versus index, all the points
      lay in a pattern."}
        else if (input$repeatedMeasureSelect == "Interaction of Block and Treatment") {
          "For interaction, points in the block have a different trend."}
        else if (input$repeatedMeasureSelect == "Random Effects") {
          "For random, this is a plot of norm quantile with 1 point lay in the
    80% confidence envelope."})
    })

  ## Model Assumptions Game ----
  observeEvent(input$submit, {
    updateButton(session, "submitAnova", disabled = TRUE)
  })

  observeEvent(input$submitAnova,{
    output$markAnova <- renderUI({
      if ("Normality" %in% input$dropAnova &&
          "Homoscedasticity" %in% input$dropAnova &&
          "Independence of Observations" %in% input$dropAnova &&
          length(input$dropAnova) == 3) {
        paste("Congratulations! You got it right!")
      }
      else if ("Equality of the covariate's slope parameter" %in% input$dropAnova |
               "No statistically significiant potential outliers" %in% input$dropAnova |
               "Linear relationship covariate and the response" %in% input$dropAnova |
               "Random effects" %in% input$dropAnova |
               "Interaction of block and treatment" %in% input$dropAnova) {
        if ("Normality" %in% input$dropAnova &&
            "Homoscedasticity" %in% input$dropAnova &&
            "Independence of Observations" %in% input$dropAnova) {
          paste("You've included some assumptions not needed in this situation")
        }
        else if ("Normality" %in% input$dropAnova == FALSE |
                 "Homoscedasticity" %in% input$dropAnova == FALSE |
                 "Independence of Observations" %in% input$dropAnova == FALSE) {
          paste("You've included some assumptions not needed in this situation and
                You failed to list some assumptions")
        }
      }
      else if ("Normality" %in% input$dropAnova == FALSE |
               "Homoscedasticity" %in% input$dropAnova == FALSE |
               "Independence of Observations" %in% input$dropAnova == FALSE) {
        paste("You failed to list some assumptions")
      }
    })
  })

  observeEvent(input$submitAnova,{
    output$markAnova2 <- renderUI({
      if ("Normality" %in% input$dropAnova &&
          "Homoscedasticity" %in% input$dropAnova &&
          "Independence of Observations" %in% input$dropAnova &&
          length(input$dropAnova) == 3) {
        renderIcon("correct")
      }
      else{
        renderIcon("incorrect")
      }
    })
  })

  observeEvent(input$submit, {
    updateButton(session, "submitAncova", disabled = TRUE)
  })

  observeEvent(input$submitAncova,{
    output$markAncova <- renderUI({
      if ("Normality" %in% input$dropAncova &&
          "Homoscedasticity" %in% input$dropAncova &&
          "Independence of Observations" %in% input$dropAncova &&
          "Linear relationship covariate and the response" %in% input$dropAncova &&
          "Equality of the covariate's slope parameter" %in% input$dropAncova &&
          "No statistically significiant potential outliers" %in% input$dropAncova &&
          length(input$dropAncova) == 6) {
        paste("Congratulations! You got it right!")
      }
      else if ("Random effects" %in% input$dropAncova |
               "Interaction of block and treatment" %in% input$dropAncova) {
        if ("Normality" %in% input$dropAncova &&
           "Homoscedasticity" %in% input$dropAncova &&
           "Independence of Observations" %in% input$dropAncova &&
           "Linear relationship covariate and the response" %in% input$dropAncova &&
           "Equality of the covariate's slope parameter" %in% input$dropAncova &&
           "No statistically significiant potential outliers" %in% input$dropAncova) {
          paste("You've included some assumptions not needed in this situation")
        }
        else if ("Normality" %in% input$dropAnova == FALSE |
                "Homoscedasticity" %in% input$dropAnova == FALSE |
                "Independence of Observations" %in% input$dropAnova == FALSE |
                "Linear relationship covariate and the response" %in% input$dropAncova == FALSE |
                "Equality of the covariate's slope parameter" %in% input$dropAncova == FALSE |
                "No statistically significiant potential outliers" %in% input$dropAncova == FALSE) {
          paste("You've included some assumptions not needed in this situation and
                You failed to list some assumptions")
        }
      }
      else if ("Normality" %in% input$dropAnova == FALSE |
               "Homoscedasticity" %in% input$dropAnova == FALSE |
               "Independence of Observations" %in% input$dropAnova == FALSE |
               "Linear relationship covariate and the response" %in% input$dropAncova == FALSE |
               "Equality of the covariate's slope parameter" %in% input$dropAncova == FALSE |
               "No statistically significiant potential outliers" %in% input$dropAncova == FALSE) {
        paste("You failed to list some assumptions")
      }
    })
  })

  observeEvent(input$submitAncova,{
    output$markAncova2 <- renderUI({
      if ("Normality" %in% input$dropAncova &&
         "Homoscedasticity" %in% input$dropAncova &&
         "Independence of Observations" %in% input$dropAncova &&
         "Linear relationship covariate and the response" %in% input$dropAncova &&
         "Equality of the covariate's slope parameter" %in% input$dropAncova &&
         "No statistically significiant potential outliers" %in% input$dropAncova &&
         length(input$dropAncova) == 6) {
        renderIcon("correct")
      }
      else{
        renderIcon("incorrect")
      }
    })
  })

  observeEvent(input$submit, {
    updateButton(session, "submitBlocking", disabled = TRUE)
  })

  observeEvent(input$submitBlocking,{
    output$markBlocking <- renderUI({
      if ("Normality" %in% input$dropBlocking &&
          "Homoscedasticity" %in% input$dropBlocking &&
          "Independence of Observations" %in% input$dropBlocking &&
          "Interaction of block and treatment" %in% input$dropBlocking &&
          length(input$dropBlocking) == 4) {
        paste("Congratulations! You got it right!")
      }
      else if ("Equality of the covariate's slope parameter" %in% input$dropBlocking |
               "No statistically significiant potential outliers" %in% input$dropBlocking |
               "Linear relationship covariate and the response" %in% input$dropBlocking |
               "Random effects" %in% input$dropBlocking) {
        if ("Normality" %in% input$dropBlocking &&
           "Homoscedasticity" %in% input$dropBlocking &&
           "Independence of Observations" %in% input$dropBlocking &&
           "Interaction of block and treatment" %in% input$dropBlocking) {
          paste("You've included some assumptions not needed in this situation")
        }
        else if ("Normality" %in% input$dropBlocking == FALSE |
                "Homoscedasticity" %in% input$dropBlocking == FALSE |
                "Independence of Observations" %in% input$dropBlocking == FALSE |
                "Interaction of block and treatment" %in% input$dropBlocking == FALSE) {
          paste("You've included some assumptions not needed in this situation and
                You failed to list some assumptions")
        }
      }
      else if ("Normality" %in% input$dropBlocking |
               "Homoscedasticity" %in% input$dropBlocking |
               "Independence of Observations" %in% input$dropBlocking |
               "Interaction of block and treatment" %in% input$dropBlocking) {
        paste("You failed to list some assumptions")
      }
    })
  })

  observeEvent(input$submitBlocking,{
    output$markBlocking2 <- renderUI({
      if ("Normality" %in% input$dropBlocking &&
          "Homoscedasticity" %in% input$dropBlocking &&
          "Independence of Observations" %in% input$dropBlocking &&
          "Interaction of block and treatment" %in% input$dropBlocking &&
          length(input$dropBlocking) == 4) {
        renderIcon("correct")
      }
      else{
        renderIcon("incorrect")
      }
    })
  })

  observeEvent(input$submit, {
    updateButton(session, "submitRandomEffect", disabled = TRUE)
  })

  observeEvent(input$submitRandomEffect,{
    output$markRandomEffect <- renderUI({
      if ("Normality" %in% input$dropRandomEffect &&
          "Homoscedasticity" %in% input$dropRandomEffect &&
          "Independence of Observations" %in% input$dropRandomEffect &&
          "Random effects" %in% input$dropRandomEffect &&
          length(input$dropRandomEffect) == 4) {
        paste("Congratulations! You got it right!")
      }
      else if ("Equality of the covariate's slope parameter" %in% input$dropRandomEffect |
               "No statistically significiant potential outliers" %in% input$dropRandomEffect |
               "Linear relationship covariate and the response" %in% input$dropRandomEffect |
               "Interaction of block and treatment" %in% input$dropRandomEffect) {
        if ("Normality" %in% input$dropRandomEffect &&
           "Homoscedasticity" %in% input$dropRandomEffect &&
           "Independence of Observations" %in% input$dropRandomEffect &&
           "Random effects" %in% input$dropRandomEffect) {
          paste("You've included some assumptions not needed in this situation")
        }
        else if ("Normality" %in% input$dropRandomEffect == FALSE |
                "Homoscedasticity" %in% input$dropRandomEffect == FALSE |
                "Independence of Observations" %in% input$dropRandomEffect == FALSE |
                "Random effects" %in% input$dropRandomEffect == FALSE) {
          paste("You've included some assumptions not needed in this situation and
                You failed to list some assumptions")
        }
      }
      else if ("Normality" %in% input$dropAnova == FALSE |
               "Homoscedasticity" %in% input$dropAnova == FALSE |
               "Independence of Observations" %in% input$dropAnova == FALSE |
               "Random effects" %in% input$dropRandomEffect == FALSE) {
        paste("You failed to list some assumptions")
      }
    })
  })

  observeEvent(input$submitRandomEffect,{
    output$markRandomEffect2 <- renderUI({
      if ("Normality" %in% input$dropRandomEffect &&
          "Homoscedasticity" %in% input$dropRandomEffect &&
          "Independence of Observations" %in% input$dropRandomEffect &&
          "Random effects" %in% input$dropRandomEffect &&
          length(input$dropRandomEffect) == 4) {
        renderIcon("correct")
      }
      else{
        renderIcon("incorrect")
      }
    })
  })

  observeEvent(input$submit, {
    updateButton(session, "submitRepeatedMeasure", disabled = TRUE)
  })

  observeEvent(input$submitRepeatedMeasure,{
    output$markRepeatedMeasure <- renderUI({
      if ("Normality" %in% input$dropRepeatedMeasure &&
          "Homoscedasticity" %in% input$dropRepeatedMeasure &&
          "Independence of Observations" %in% input$dropRepeatedMeasure &&
          "Random effects" %in% input$dropRepeatedMeasure &&
          "Interaction of block and treatment" %in% input$dropRepeatedMeasure &&
          length(input$dropRepeatedMeasure) == 5) {
        paste("Congratulations! You got it right!")
      }
      else if ("Equality of the covariate's slope parameter" %in% input$dropRepeatedMeasure |
               "No statistically significiant potential outliers" %in% input$dropRepeatedMeasure |
               "Linear relationship covariate and the response" %in% input$dropRepeatedMeasure) {
        if ("Normality" %in% input$dropRepeatedMeasure &&
           "Homoscedasticity" %in% input$dropRepeatedMeasure &&
           "Independence of Observations" %in% input$dropRepeatedMeasure &&
           "Random effects" %in% input$dropRepeatedMeasure &&
           "Interaction of block and treatment" %in% input$dropRepeatedMeasure) {
          paste("You've included some assumptions not needed in this situation")
        }
        else if ("Normality" %in% input$dropRepeatedMeasure == FALSE |
                "Homoscedasticity" %in% input$dropRepeatedMeasure == FALSE |
                "Independence of Observations" %in% input$dropRepeatedMeasure == FALSE |
                "Random effects" %in% input$dropRepeatedMeasure == FALSE |
                "Interaction of block and treatment" %in% input$dropRepeatedMeasure == FALSE) {
          paste("You've included some assumptions not needed in this situation and
                You failed to list some assumptions")
        }
      }
      else if ("Normality" %in% input$dropRepeatedMeasure == FALSE |
               "Homoscedasticity" %in% input$dropRepeatedMeasure == FALSE |
               "Independence of Observations" %in% input$dropRepeatedMeasure == FALSE |
               "Random effects" %in% input$dropRepeatedMeasure == FALSE |
               "Interaction of block and treatment" %in% input$dropRepeatedMeasure == FALSE) {
        paste("You failed to list some assumptions")
      }
    })
  })

  observeEvent(input$submitRepeatedMeasure,{
    output$markRepeatedMeasure2 <- renderUI({
      if ("Normality" %in% input$dropRepeatedMeasure &&
          "Homoscedasticity" %in% input$dropRepeatedMeasure &&
          "Independence of Observations" %in% input$dropRepeatedMeasure &&
          "Random effects" %in% input$dropRepeatedMeasure &&
          "Interaction of block and treatment" %in% input$dropRepeatedMeasure &&
          length(input$dropRepeatedMeasure) == 5) {
        renderIcon("correct")
      }
      else{
        renderIcon("incorrect")
      }
    })
  })

  ### Checking  ----
    #### Normality ----
  normalityData1 <- rnorm(n = 50, mean = 0, sd = 1)
  normalityData2 <- rnorm(n = 50, mean = 0, sd = 1)
  normalityData3 <-  rbeta(50, 0.7, 1.5)

  output$normalityGamePlot1 <- renderPlot(
    expr = {
      car::qqPlot(
      pch = 19,
      cex = 1.5,
      id = FALSE,
      x = normalityData1,
      distribution = "norm",
      envelope = 0.95,
      ylab = "data",
      main = "Plot A"
      )},
    alt = "This is a plot of normal quantiles. There are 50 points in the 95% confidence
        envelope.")

  output$normalityGamePlot2 <- renderPlot(
    expr = {
    car::qqPlot(
      pch = 19,
      cex = 1.5,
      id = FALSE,
      x = normalityData2,
      distribution = "norm",
      envelope = 0.95,
      ylab = "data",
      main = "Plot C"
    )},
    alt = "This is a plot of normal quantiles. There are 50 points, most of the
    points lied in the 95% confidence envelope, while others not."
  )

  output$normalityGamePlot3 <- renderPlot(
    expr = {
    car::qqPlot(
      pch = 19,
      cex = 1.5,
      id = FALSE,
      x = normalityData3,
      distribution = "norm",
      envelope = 0.8,
      ylab = "data",
      main = "Plot B"
    )},
    alt = "This is a plot of normal quantiles. There are 50 points in the 95% confidence
    envelope.")

  observeEvent(input$submitNormality,{
    if (input$nomalitySelected == 'plot B') {
      output$markNormality <- renderIcon(icon = "correct")
      } else{
      output$markNormality <-  renderIcon(icon = "incorrect")
      }
    })

  observeEvent(input$submit, {
    updateButton(session, "submitNormality", disabled = TRUE)
  })
    #### Homo ----
  homoData1 <- data.frame(
    homoData1_1 = c(rep('1',10), rep('2',10), rep('3',10), rep('4',10), rep('5',10)),
    homoData1_2 = c(sample(1:100, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE),
                    sample(1:80, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE)))
  homoData2 <- data.frame(
    homoData2_1 = c(rep('1',10), rep('2',10), rep('3',10), rep('4',10), rep('5',10)),
    homoData2_2 = c(sample(1:100, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE),
                    sample(1:100, 10, replace = FALSE)))
  homoData3 <- data.frame(
    homoData3_1 = c(rep('1',10), rep('2',10), rep('3',10), rep('4',10), rep('5',10)),
    homoData3_2 = c(sample(1:100, 10, replace = FALSE),
                    sample(10:90, 10, replace = FALSE),
                    sample(20:80, 10, replace = FALSE),
                    sample(30:70, 10, replace = FALSE),
                    sample(40:60, 10, replace = FALSE)))

  output$homoGamePlot1 <- renderPlot(
    expr = {
    stripchart(
      pch = 19,
      cex = 1.5,
      homoData1_2 ~ homoData1_1,
      vertical = TRUE,
      data = homoData1,
      xlab = 'Index',
      ylab = 'homoData1',
      main = "Plot A",
      cex.lab = 1.5,
      cex.axis = 1.5)
    },
    alt = "This is a plot of response versus explanatory, all the points lay in
    a random position.")

  output$homoGamePlot2 <- renderPlot(
    expr = {
    stripchart(
      pch = 19,
      cex = 1.5,
      homoData2_2 ~ homoData2_1,
      vertical = TRUE,
      data = homoData2,
      xlab = 'Index',
      ylab = 'homoData2',
      main = "Plot B",
      cex.lab = 1.5,
      cex.axis = 1.5)
  },
   alt = "This is a plot of response versus explanatory, all the points lay in a
    random position.")

  output$homoGamePlot3 <- renderPlot(
    expr = {
    stripchart(
      pch = 19,
      cex = 1.5,
      homoData3_2 ~ homoData3_1,
      vertical = TRUE,
      data = homoData3,
      xlab = 'Index',
      ylab = 'homoData3',
      main = "Plot C",
      cex.lab = 1.5,
      cex.axis = 1.5)
    },
    alt = "This is a plot of response versus explanatory, all the points lay in
    a pattern.")

  observeEvent(input$submitHomo,{
    if (input$homoSelected == 'plot C') {
      output$markHomo <- renderIcon(icon = "correct")
    } else{
      output$markHomo <-  renderIcon(icon = "incorrect")
    }
  })

  observeEvent(input$submit, {
    updateButton(session, "submitHomo", disabled = TRUE)
  })
  #### Indep ----
  indeData1 <- rnorm(n = 50, mean = 0, sd = 1)
  indeData2 <- rnorm(n = 50, mean = 0, sd = 1)
  indeData3 <- ts(1:10, frequency = 4, start = c(1959, 2))

  output$indeGamePlot1 <- renderPlot(
    expr = {
    plot(indeData1,
         type = "b",
         main = "Plot C")
    },
    alt = "This is a plot of response versus index, all the points lay in a pattern.")

  output$indeGamePlot2 <- renderPlot(
    expr = {
    plot(indeData2,
         type = "b",
         main = "Plot B")
    },
    alt = "This is a plot of response versus index, all the points layin a random
    position.")

  output$indeGamePlot3 <- renderPlot(
    expr = {
    plot(indeData3,
         type = "b",
         main = "Plot A",
         ylab = "Index",
         xlab = 'Index')
    },
    alt = "This is a plot of response versus index, all the points lay in a random
    position")

  observeEvent(input$submitInde,{
    if (input$indeSelected == 'plot A') {
      output$markInde <- renderIcon(icon = "correct")
    } else{
      output$markInde <-  renderIcon(icon = "incorrect")
    }
  })

  observeEvent(input$submit, {
    updateButton(session, "submitInde", disabled = TRUE)
  })
  #### Linear ----
  linearData1 <- data.frame(
    linearData1_1 = rep(c('1', '2', '3', '4', '5'), 10),
    linearData1_2 = c(sample(1:20, 10, replace = FALSE),
                      sample(10:30, 10, replace = FALSE),
                      sample(15:35, 10, replace = FALSE),
                      sample(20:40, 10, replace = FALSE),
                      sample(30:50, 10, replace = FALSE)),
    linearData1_3 = c(sample(1:20, 10, replace = FALSE),
                      sample(10:30, 10, replace = FALSE),
                      sample(15:35, 10, replace = FALSE),
                      sample(20:40, 10, replace = FALSE),
                      sample(30:50, 10, replace = FALSE))
  )
  linearData2 <- data.frame(
    linearData2_1 = rep(c('1', '2', '3', '4', '5'), 10),
    linearData2_2 = c(sample(50:30, 10, replace = FALSE),
                      sample(40:20, 10, replace = FALSE),
                      sample(35:15, 10, replace = FALSE),
                      sample(30:10, 10, replace = FALSE),
                      sample(20:1, 10, replace = FALSE)),
    linearData2_3 = c(sample(1:20, 10, replace = FALSE),
                      sample(10:30, 10, replace = FALSE),
                      sample(15:35, 10, replace = FALSE),
                      sample(20:40, 10, replace = FALSE),
                      sample(30:50, 10, replace = FALSE))
  )
  linearData3 <- data.frame(
    linearData3_1 = c(rep("1", 10), rep("2", 10), rep("3", 10), rep("4", 10), rep("5", 10)),
    linearData3_2 = sample(1:100, 50, replace = FALSE),
    linearData3_3 = sample(1:100, 50, replace = FALSE)
  )

  output$linearGamePlot1 <- renderPlot(
    expr = {
    ggplot(data = linearData1,
                    mapping = aes(
                      y = linearData1_2,
                      x = linearData1_3,
                      group = linearData1_1,
                      color = linearData1_1
                    )) +
      geom_point(size = 3) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type") +
      ggtitle("Plot C") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "The plot shows that response and explantory have no linear relationship.")

  output$linearGamePlot2 <- renderPlot(
    expr = {
    ggplot(data = linearData2,
                    mapping = aes(
                      y = linearData2_2,
                      x = linearData2_3,
                      group = linearData2_1,
                      color = linearData2_1
                    )) +
      geom_point(size = 3) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type") +
      ggtitle("Plot B") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "The plot shows that response and explantory have a negative linear relationship.")

  output$linearGamePlot3 <- renderPlot(
    expr = {
    ggplot(data = linearData3,
                    mapping = aes(
                      y = linearData3_2,
                      x = linearData3_3,
                      group = linearData3_1,
                      color = linearData3_1
                    )) +
      geom_point(size = 3) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type") +
      ggtitle("Plot A") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "The plot shows that response and explantory have a postitive linear relationship.")

  observeEvent(input$submitLinear,{
    if (input$linearSelected == 'plot A') {
      output$markLinear <- renderIcon(icon = "correct")
    } else{
      output$markLinear <-  renderIcon(icon = "incorrect")
    }
  })

  observeEvent(input$submit, {
    updateButton(session, "submitLinear", disabled = TRUE)
  })
  #### Slope ----
  slopeData1 <- data.frame(
    slopeData1_1 = rep(c('1', '2', '3', '4', '5'), 10),
    slopeData1_2 = c(sample(1:20, 10, replace = FALSE),
                     sample(10:30, 10, replace = FALSE),
                     sample(15:35, 10, replace = FALSE),
                     sample(20:40, 10, replace = FALSE),
                     sample(30:50, 10, replace = FALSE)),
    slopeData1_3 = c(sample(1:20, 10, replace = FALSE),
                     sample(10:30, 10, replace = FALSE),
                     sample(15:35, 10, replace = FALSE),
                     sample(20:40, 10, replace = FALSE),
                     sample(30:50, 10, replace = FALSE))
  )
  slopeData2 <- data.frame(
    slopeData2_1 = rep(c('1', '2', '3', '4', '5'), 10),
    slopeData2_2 = c(sample(50:30, 10, replace = FALSE),
                     sample(40:20, 10, replace = FALSE),
                     sample(35:15, 10, replace = FALSE),
                     sample(30:10, 10, replace = FALSE),
                     sample(20:1, 10, replace = FALSE)),
    slopeData2_3 = c(sample(1:20, 10, replace = FALSE),
                     sample(10:30, 10, replace = FALSE),
                     sample(15:35, 10, replace = FALSE),
                     sample(20:40, 10, replace = FALSE),
                     sample(30:50, 10, replace = FALSE))
  )
  slopeData3 <- data.frame(
    slopeData3_1 = c(rep("1", 10), rep("2", 10), rep("3", 10), rep("4", 10), rep("5", 10)),
    slopeData3_2 = sample(1:100, 50, replace = FALSE),
    slopeData3_3 = sample(1:100, 50, replace = FALSE)
  )

  output$slopeGamePlot1 <- renderPlot(
    expr = {
    ggplot(data = slopeData1,
                    mapping = aes(
                      y = slopeData1_2,
                      x = slopeData1_3,
                      group = slopeData1_1,
                      color = slopeData1_1
                    )) +
      geom_point(size = 3) +
      geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type") +
      ggtitle("Plot A") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "In the plot, points in the different groups have a similar trend.")

  output$slopeGamePlot2 <- renderPlot(
    expr = {
    ggplot(data = slopeData2,
                    mapping = aes(
                      y = slopeData2_2,
                      x = slopeData2_3,
                      group = slopeData2_1,
                      color = slopeData2_1
                    )) +
      geom_point(size = 3) +
      geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type") +
      ggtitle("Plot C") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "In the plot, points in the different groups have a similar trend")

  output$slopeGamePlot3 <- renderPlot(
    expr = {
    ggplot(data = slopeData3,
                    mapping = aes(
                      y = slopeData3_2,
                      x = slopeData3_3,
                      group = slopeData3_1,
                      color = slopeData3_1
                    )) +
      geom_point(size = 3) +
      geom_smooth(formula = y ~ x, method = "lm", se = FALSE) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type") +
      ggtitle("Plot B") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "In the plot, points in the different groups have a different trend.")

  observeEvent(input$submitSlope,{
    if (input$slopeSelected == 'plot B') {
      output$markSlope <- renderIcon(icon = "correct")
    } else{
      output$markSlope <-  renderIcon(icon = "incorrect")
    }
  })

  observeEvent(input$submit, {
    updateButton(session, "submitSlope", disabled = TRUE)
  })
  #### Outliters ----
  outData1 <- data.frame(
    outData1_1 = rep(c('1', '2', '3', '4', '5'), 10),
    outData1_2 = c(sample(1:10, 10, replace = FALSE),
                   sample(10:20, 10, replace = FALSE),
                   sample(20:30, 10, replace = FALSE),
                   sample(30:40, 10, replace = FALSE),
                   sample(40:50, 10, replace = FALSE)),
    outData1_3 = c(sample(1:10, 10, replace = FALSE),
                   sample(10:20, 10, replace = FALSE),
                   sample(20:30, 10, replace = FALSE),
                   sample(30:40, 10, replace = FALSE),
                   sample(40:50, 10, replace = FALSE))
  )
  outData2 <- data.frame(
    outData2_1 = rep(c('1', '2', '3', '4', '5'), 10),
    outData2_2 = c(sample(1:10, 10, replace = FALSE),
                   sample(10:20, 10, replace = FALSE),
                   sample(20:30, 10, replace = FALSE),
                   sample(30:40, 10, replace = FALSE),
                   sample(40:50, 10, replace = FALSE)),
    outData2_3 = c(sample(1:10, 10, replace = FALSE),
                   sample(10:20, 10, replace = FALSE),
                   sample(20:30, 10, replace = FALSE),
                   sample(30:40, 10, replace = FALSE),
                   sample(40:50, 10, replace = FALSE))
  )
  outData3 <- data.frame(
    outData3_1 = c(rep("1", 10), rep("2", 10), rep("3", 10), rep("4", 10), rep("5", 10)),
    outData3_2 = c(sample(1:100, 48, replace = FALSE),
                   sample(200:210, 2, replace = FALSE)),
    outData3_3 = sample(1:100, 50, replace = FALSE)
  )

  output$outGamePlot1 <- renderPlot(
    expr = {
    realOutData1 <- rstatix::mahalanobis_distance(outData1)
    realOutData1 <- cbind(realOutData1, factor = outData1$outData1_1)
    ggplot(data = realOutData1,
                    mapping = aes(
                      y = outData1_2,
                      x = outData1_3,
                      color = factor
                    )) +
      geom_point(size = 3) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type", shape = "Potential Outlier") +
      ggtitle("Plot C") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "The plot shows that the data do not have obvious outliers.")

  output$outGamePlot2 <- renderPlot(
    expr = {
    realOutData2 <- rstatix::mahalanobis_distance(outData2)
    realOutData2 <- cbind(realOutData2, factor = outData2$outData2_1)
    ggplot(data = realOutData2,
                    mapping = aes(
                      y = outData2_2,
                      x = outData2_3,
                      color = factor
                    )) +
      geom_point(size = 3) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type", shape = "Potential Outlier") +
      ggtitle("Plot B") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "The plot shows that the data do not have obvious outliers.")

  output$outGamePlot3 <- renderPlot(
    expr = {
    realOutData3 <- rstatix::mahalanobis_distance(outData3)
    realOutData3 <- cbind(realOutData3, factor = outData3$outData3_1)
    ggplot(data = realOutData3,
                    mapping = aes(
                      y = outData3_2,
                      x = outData3_3,
                      color = factor
                    )) +
      geom_point(size = 3) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Type", shape = "Potential Outlier") +
      ggtitle("Plot A") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "The plot shows that the data has obvious outliers.")

  observeEvent(input$submitOut,{
    if (input$outSelected == 'plot A') {
      output$markOut <- renderIcon(icon = "correct")
    } else{
      output$markOut <-  renderIcon(icon = "incorrect")
    }
  })

  observeEvent(input$submit, {
    updateButton(session, "submitOut", disabled = TRUE)
  })
  #### Interaction ----
  interData1 <- data.frame(
    interData1_1 = rep(c('1', '2', '3', '4', '5'), 10),
    interData1_2 = c(sample(1:10, 10, replace = FALSE),
                     sample(10:20, 10, replace = FALSE),
                     sample(20:30, 10, replace = FALSE),
                     sample(30:40, 10, replace = FALSE),
                     sample(40:50, 10, replace = FALSE)),
    interData1_3 = c(sample(1:10, 10, replace = FALSE),
                     sample(10:20, 10, replace = FALSE),
                     sample(20:30, 10, replace = FALSE),
                     sample(30:40, 10, replace = FALSE),
                     sample(40:50, 10, replace = FALSE))
  )
  interData2 <- data.frame(
    interData2_1 = rep(c('1', '2', '3', '4', '5'), 10),
    interData2_2 = c(sample(1:10, 10, replace = FALSE),
                     sample(10:20, 10, replace = FALSE),
                     sample(20:30, 10, replace = FALSE),
                     sample(30:40, 10, replace = FALSE),
                     sample(40:50, 10, replace = FALSE)),
    interData2_3 = c(sample(1:10, 10, replace = FALSE),
                     sample(10:20, 10, replace = FALSE),
                     sample(20:30, 10, replace = FALSE),
                     sample(30:40, 10, replace = FALSE),
                     sample(40:50, 10, replace = FALSE))
  )
  interData3 <- data.frame(
    interData3_1 = c(rep("1", 10), rep("2", 10), rep("3", 10), rep("4", 10), rep("5", 10)),
    interData3_2 = sample(1:100, 50, replace = FALSE),
    interData3_3 = sample(1:100, 50, replace = FALSE)
  )

  output$interGamePlot1 <- renderPlot(
    expr = {
    ggplot(data = interData1,
                    mapping = aes(x = interData1_2,
                                  y = interData1_3,
                                  color = interData1_1,
                                  group = interData1_1)) +
      geom_point(size = 2) +
      geom_line(linewidth = 1) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Group") +
      ggtitle("Plot A") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "In the plot, points in the different block have a similar trend.")

  output$interGamePlot2 <- renderPlot(
    expr = {
    ggplot(data = interData2,
                    mapping = aes(x = interData2_2,
                                  y = interData2_3,
                                  color = interData2_1,
                                  group = interData2_1)) +
      geom_point(size = 2) +
      geom_line(linewidth = 1) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Group") +
      ggtitle("Plot C") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "In the plot, points in the different block have a similar trend.")

  output$interGamePlot3 <- renderPlot(
    expr = {
    ggplot(data = interData3,
                    mapping = aes(x = interData3_2,
                                  y = interData3_3,
                                  color = interData3_1,
                                  group = interData3_1)) +
      geom_point(size = 2) +
      geom_line(linewidth = 1) +
      theme_bw() +
      theme(axis.title = element_text(size = 18)) +
      xlab("X") +
      ylab("Y") +
      labs(color = "Group") +
      ggtitle("Plot B") +
      theme(plot.title = element_text(size = 15, hjust = 0.5, vjust = 1, face = "bold"))
    },
    alt = "In the plot, points in the different block have a different trend.")


  observeEvent(input$submitInter,{
    if (input$interSelected == 'plot B') {
      output$markInter <- renderIcon(icon = "correct")
    } else{
      output$markInter <-  renderIcon(icon = "incorrect")
    }
  })

  observeEvent(input$submit, {
    updateButton(session, "submitInter", disabled = TRUE)
  })


  }

# Boast App Call ----
boastUtils::boastApp(ui = ui, server = server)