Exercises Week 1

Go to basic R

Go to import data

Go to making figures

Exercise 1. Import and Select subject, timepoint and sj.max from the cycling study data set. Print your results. What does printing do?

Solution Ex. 1

# load packages
library(tidyverse)
library(readxl)

cyclingStudy <- read_excel("./data/cyclingStudy.xlsx", na = "NA") # remember to use the na argument

cyclingStudy %>%
        select(subject, timepoint, sj.max) %>%
        print()

Alternatively, everything can be written in one pipe:

cyclingStudy <- read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
 select(subject, timepoint, sj.max) %>%
        print()

Exercise 2. Filter the cycling data set based on time-point keep only observations from time-point pre.

Solution Ex. 2

cyclingStudy <- read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
 filter(timepoint == "pre") %>%
        print()

Exercise 3. Select only columns subject and sj.max. Filter the cycling data set based on time-point keep only observations from time-point pre and meso1. . In what order must you do the operations in order to get it to work?

Solution Ex. 3

cyclingStudy <- read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
        filter(timepoint %in% c("pre", "meso1")) %>%
        select(subject, sj.max) %>%
        print()

Exercise 4. weight.T1 is the weight in kg and VO2.max is the \(\dot VO_{2max}\) in \(ml \times min^{-1}\) from each test. Select the needed variables, filter needed observations, create a new variable \(\dot VO_{2max} / kg\) and calculate the percentage change from time-point “pre” to “meso3”. This operation should include mutate(), filter(), pivot_wider.

Solution Ex. 4

read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
        filter(timepoint %in% c("pre", "meso3")) %>%
        select(subject, timepoint, VO2.max, weight.T1) %>%
        mutate(relative.vo2max = VO2.max / weight.T1) %>%
        select(subject, timepoint, relative.vo2max) %>%
        pivot_wider(names_from = timepoint, values_from = relative.vo2max) %>%
        mutate(change = ((meso3 / pre)-1)*100) %>%
        print()

Exercise 5. Do the same as above but keep the group variable. Calculate the mean of each group in percentage change from pre- to meso3. In the data set there are three groups: “INCR” increased intensity through the training period, “DECR” decreased intensity and “MIX” performed a mix of intensities each week.

Solution Ex. 5

read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
        filter(timepoint %in% c("pre", "meso3")) %>%
        select(subject, group, timepoint, VO2.max, weight.T1) %>%
        mutate(relative.vo2max = VO2.max / weight.T1) %>%
        select(subject, group, timepoint, relative.vo2max) %>%
        pivot_wider(names_from = timepoint, values_from = relative.vo2max) %>%
        mutate(change = ((meso3 / pre)-1)*100) %>%
        group_by(group) %>%
        summarise(m = mean(change, na.rm = TRUE)) %>%
        print()

The n()function can be used with summarise(). n() returns the number of observations including NAs. We can include n() to count the number of observations. If we are only interested in values that are not NA we could use the sum() function and count the number of values that are not NA. is.na() says TRUE if it finds a NA. if we inverse it, we will get a counter of non NAs

read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
        filter(timepoint %in% c("pre", "meso3")) %>%
        select(subject, group, timepoint, VO2.max, weight.T1) %>%
        mutate(relative.vo2max = VO2.max / weight.T1) %>%
        select(subject, group, timepoint, relative.vo2max) %>%
        pivot_wider(names_from = timepoint, values_from = relative.vo2max) %>%
        mutate(change = ((meso3 / pre)-1)*100) %>%
        group_by(group) %>%
        summarise(m = mean(change, na.rm = TRUE), 
                  n = n(), 
                  non.na = sum(!is.na(change))) %>%
        print()

Exercise 6. \(\dot VO_2\) meassures were collected during submaximal exercise, variables VO2.125:VO2.375 are \(\dot VO_2\) measurements from each watt. Transform the data to get this to tidy format.

Solution Ex. 6

read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
       select(subject, timepoint, VO2.125:VO2.375) %>%
        pivot_longer(names_to = "watt", 
                     names_prefix = "VO2.",
                     names_ptypes = list(watt = numeric()),
                     values_to = "VO2", 
                     cols = VO2.125:VO2.375) %>%
        print()

Exercise 7. \(\dot VO_2\) meassures were collected during submaximal exercise, variables VO2.125:VO2.375 are \(\dot VO_2\) measurements from each watt. Transform the data to get this to tidy format. Calculate the mean \(\dot VO_2\) for each group, timepoint and watt.

Solution Ex. 7

read_excel("./data/cyclingStudy.xlsx", na = "NA") %>% # remember to use the na argument
       select(subject, group, timepoint, VO2.125:VO2.375) %>%
        pivot_longer(names_to = "watt", 
                     names_prefix = "VO2.",
                     names_ptypes = list(watt = numeric()),
                     values_to = "VO2", 
                     cols = VO2.125:VO2.375) %>%
        group_by(group, timepoint, watt) %>%
        summarise(vo2 = mean(VO2, na.rm = TRUE)) %>%
        print()