Replicating Table 1

---

This script replicates Figure 5 (Development of the mean of running attributes).

Getting started

clean up

rm (list = ls( ))

custom functions

• fpackage.check: Check if packages are installed (and install if not) in R (source)

fpackage.check <- function(packages) {
    lapply(packages, FUN = function(x) {
        if (!require(x, character.only = TRUE)) {
            install.packages(x, dependencies = TRUE)
            library(x, character.only = TRUE)
        }
    })
}

necessary packages

We install and load the packages we need later on: - moments: for calculating statistics - dplyr: for data manipulation - ggplot2: for data visualization

packages = c("moments", "dplyr", "ggplot2")
fpackage.check(packages)

load club data

load("clubdata.RData")

data wrangling

# data males (mean run freq./vol. and mean freq./vol. other activities: cycling swimming)
mean_freq <- mean_vol <- matrix(NA, nrow=12, ncol=5)
for ( i in 1:5) {
  for ( t in 1:12) {
    mean_freq[t,i] <- mean(clubdata[[i]]$freq_run[,,t][which(clubdata[[i]]$male==1)], na.rm=T)
    mean_vol[t,i] <- mean(clubdata[[i]]$time_run[,,t][which(clubdata[[i]]$male==1)], na.rm=T)
  }
}

sd_freq <- sd_vol <- matrix(NA, nrow=12, ncol=5)
for ( i in 1:5) {
  for ( t in 1:12) {
    sd_freq[t,i] <- sd(clubdata[[i]]$freq_run[,,t][which(clubdata[[i]]$male==1)], na.rm=T)
    sd_vol[t,i] <- sd(clubdata[[i]]$time_run[,,t][which(clubdata[[i]]$male==1)], na.rm=T)
  }
}

se_freq <- se_vol <- matrix(NA, nrow=12, ncol=5) 
for ( i in 1:5) {
  for ( t in 1:12) {
    se_freq[t,i] <- sd_freq[t,i]/sqrt(sum(clubdata[[i]]$male==1))
    se_vol[t,i] <- sd_vol[t,i]/sqrt(sum(clubdata[[i]]$male==1))
  }
}

data_male <- data.frame(
  mean = as.vector(c(mean_freq, mean_vol)),
  sd = as.vector(c(sd_freq, sd_vol)),
  se =  as.vector(c(se_freq, se_vol)),
  club = as.character(sort(replicate(12, c("Club 1 (N=27)", "Club 2 (N=58)", "Club 3 (N=159)", "Club 4 (N=9)", "Club 5 (N=76)")))),
  time = rep(c(1:12)),
  Attribute = c(rep("Frequency", 12*5), rep("Volume", 12*5)),
  Group = "Male")


# data females and others
mean_freq <- matrix(NA, nrow=12, ncol=5)
mean_vol <- matrix(NA, nrow=12, ncol=5)
for ( i in 1:5) {
  for ( t in 1:12) {
    mean_freq[t,i] <- mean(clubdata[[i]]$freq_run[,,t][which(!clubdata[[i]]$male==1)],na.rm=T)
    mean_vol[t,i] <- mean(clubdata[[i]]$time_run[,,t][which(!clubdata[[i]]$male==1)],na.rm=T)
  }
}

sd_freq <- matrix(NA, nrow=12, ncol=5)
sd_vol <- matrix(NA, nrow=12, ncol=5)
for ( i in 1:5) {
  for ( t in 1:12) {
    sd_freq[t,i] <- sd(clubdata[[i]]$freq_run[,,t][which(!clubdata[[i]]$male==1)],na.rm=T)
    sd_vol[t,i] <- sd(clubdata[[i]]$time_run[,,t][which(!clubdata[[i]]$male==1)],na.rm=T)
  }
}

se_freq <- matrix(NA, nrow=12, ncol=5)  
se_vol <- matrix(NA, nrow=12, ncol=5)  
for ( i in 1:5) {
  for ( t in 1:12) {
    se_freq[t,i] <- sd_freq[t,i]/sqrt(sum(!clubdata[[1]]$male==1))
    se_vol[t,i] <- sd_vol[t,i]/sqrt(sum(!clubdata[[1]]$male==1))
  }
}

data_female <- data.frame(
  mean = as.vector(c(mean_freq, mean_vol)),
  sd = as.vector(c(sd_freq, sd_vol)),
  se =  as.vector(c(se_freq, se_vol)),
  club = as.character(sort(replicate(12, c("Club 1 (N=27)", "Club 2 (N=58)", "Club 3 (N=159)", "Club 4 (N=9)", "Club 5 (N=76)")))),
  time = rep(c(1:12)),
  Attribute = c(rep("Frequency", 12*5), rep("Volume", 12*5)),
  Group = "Female and others"
)

---

Plotting

plot <- rbind(data_male, data_female) #, data_all)

pd <- position_dodge(width = .8) # dodge between points to prevent overlap of CIs

plot$Group <- factor(plot$Group,
                     levels = c("Male", "Female and others"))

ggplot(plot, aes(x=time, y=mean, colour=Group, #shape = Attribute,
                 group=Group)) +
  #group=interaction(Group, Attribute))) + 
  geom_point(size = 1.5, position = pd) + 
  geom_line(size = .75, position = pd) +
  geom_errorbar(aes(time, mean, ymin = mean - se, ymax = mean + se), 
                width = 0.6, position = pd) +
  labs(title="Development of running activity attribute means over time", 
       subtitle="Disaggregated by running measure (frequency and volume), club and gender") +
  scale_colour_manual(values=c("#56B4E9", "#E69F00")) +
  labs(x = "", y = "Hours per week        Times per week") +
  scale_y_continuous(breaks = seq(0, 5, by = 1)) +
  scale_x_discrete(limits=c("Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec")) +
  facet_grid(Attribute~club) +
  theme(axis.text.x = element_text(size = 8, angle = 70, vjust = -.2, hjust=-.1), 
        legend.position = "top",
        legend.box.just = "right",
        legend.direction = "horizontal",
        legend.background = element_rect(fill="lightgray", size=.5, linetype="dotted"),
        legend.title = element_text(face = "bold"),
        #legend.background = element_rect(fill ="gray90", size=.5),
        strip.text.y = element_blank())

---

Copyright © 2021 Rob Franken