1. Task (5 points)

Your first task is to start creating the Rmarkdown document, namely to:

• input a reasonable title for the document, e.g., “IDA 2019: Homework 1”
• create a first level head for each task using # Task n
• for this first task create two additional headers:
  • a level-2 header with title “The data set I: Creating a table in RMarkdown”,
  • a level-3 header with title “Overview of variables in the data set”,
• re-create the table depicted below in using Rmarkdown’s own style of writing tables,
• fill in the missing entries (the values are ordered from left to right in the picture; “variable type” refers to nominal, ordinal, metric, etc.),
• take care that the first and second column are centered and the third column is left-oriented and
• that the column names are “bold”.

(Note: Use the provided cheat sheets for RMarkdown from RStudio to look up how to make headers, create a table, ect. in Rmd))

2. Task (4 points)

Before starting to work in R, we should load the relevant packages that we need in order to work with the data.

• create a level-2 header with title: “Loading relevant packages”
• insert in your Rmd-file an R code chunk, in which you can type in the R code
• load the package tidyverse (if you have not already installed the package you have first to install them by using install.packages("tidyverse"))
• as we do not want to have all the warnings and messages (that usually appear while loading a package) in the knitted Rmd file, insert in your R chunk the option that the warnings and messages should not be printed in your knitted Rmd file.

< R CODE HERE >

3.Task (5 points)

Now, we can start to create a tibble in R.

• create a level-2 header: “The data set II: Creating a tibble in R”
• insert in your Rmd-file an R code chunk, in which you can type in the R code
• store your tibble in a variable called brain_data
• as column names we will use: brain_regions, surface_area, thickness and hemisphere,
• as values you should use those from the above Rmd-table,
• make the variable brain_regions a factor (when creating the tibble, use the keyword factor in the creating of this column)
• print the tibble for display in your document

(Hint: your tibble should have 4 columns and 5 rows eventually.)

< R CODE HERE >

4. Task (3 points)

Oh no, we have done a little mistake! The brain region “Transverse temporal” should actually be represented as “Banks superior temporal”. We have to change this level in the factor brain_regions.

Give R Code that

• changes the value “Transverse temporal” of the factor brain_regions into the right value “Banks superior temporal”
• to do this, use the function fct_recode from the purrr package (you can access the factor/colum using brain_data$brain_regions)
• print the brain_data again.

The column brain_regions should now have the value “Banks superior temporal” instead of “Transverse temporal”.

< R CODE HERE >

5. Task (1 point)

Using indexing (like in a matrix), extract the thickness for the brain region “Precentral” from the tibble.

• insert a R code chunk and
• type in a code that prints the answer of the above question.

< R CODE HERE >

6. Task (5 points)

Consider again the data set. Suppose you come across the information, that you can calculate the volume of brain regions with the given variables by the formula: volume = surface_area * thickness.

Therefore, you decide to create a new tibble, named brain_data2 from scratch.

• create a second-level header: “The data set III: Creating a tibble in R”
• insert in your Rmd-file a R code chunk, in which you can type in the R code
• store your tibble in the variable `brain_data2
• as column names and values we will use the same as above but
• you should include a new variable volume whose values are calculated by multiplying surface_area with thickness
• print the created tibble.

< R CODE HERE >

7. Task (1 point)

Why would it not work to use the same code to create a data frame instead of a tibble?

< ANSWER HERE >

8. Task (2 points)

Write R code to

• select only the column brain_regions from the data
• store it in a variable called regions_v1 and
• print this variable

< R CODE HERE >

9. Task (1 point)

Write R code that returns the data type of regions_v1?

< R CODE HERE >

10. Task (2 points)

Suppose you do not know the values of the variable regions_v1 but you want to know how many brain regions are included. Write some R code that returns the length of the vector regions_v1.

< R CODE HERE >

11. Task (1 point)

Let us consider now again the whole data set brain_data2. Give R code which returns the number of different elements in the vector stored in column hemisphere. For this, you want to first call the function unique (which returns a vector with all and only unique elements) and then the function length (to get the length of that vector). Use the pipe operator %>% to do this.

< R CODE HERE >

12. Task (2 points)

Describe in no more than 10 words what the goal of this code is

brain_data2[str_detect(brain_data2$hemisphere, "R"),"brain_regions"]

< ANSWER HERE >

13. Task (7 points)

Let us consider again the variable volume in the data set brain_data2. We can calculate the variable volume also by using a function. Which we will do now.

13.1 First,

• extract column surface_area from the data set brain_data2 and store it as variable x
• do the same for the variabel thickness and store it as y

< R CODE HERE >

13.2 Second,

• create a function that takes as input variables x and y and gives as output the multiplication of both
• store the function under the name volume_calc

< R CODE HERE >

13.3 Third,

• create a list and store it as volume_compare; the two list elements are
• volume1, which is the column volume from the data set brain_data2 and
• volume2, which is the volume calculated by your function
• print the list

< R CODE HERE >

14. Task (5 points)

Suppose that we have the chance to measure the volume of each brain region directly. Here are the directly measured values for volume:

• Rostral middle frontal = 17439
• Precentral = 14351
• Lateral Occipital = 12150
• Banks = 2386
• Temporal pole = 2280.1

14.1 Create a new variable volume_true with the given values.

< R CODE HERE >

14.2 Extract volume1 from your created list volume_compare and store it as volume_calculated

< R CODE HERE >

14.3 Write a new function that takes as

• input two variables x and y and
• returns as output the difference between both input variables.
• store the function as volume_diff

< R CODE HERE >

15. Task (3 points)

Create a tibble with two variables: brain_regions (use regions_v1) and calculated_volume, which indicates the difference between the variables volume_calculated and volume_true (make use of your custom-built funtion). Store the tibble as “brain_data3” and print it.

< R CODE HERE >

16. Task (3 points)

Look at the following formula and the output that it returns. What does this formula do? Can you give a question for which this formula returns the answer?

Hint: Use for example the help-possibility in R in order to understand what which.mindoes.

brain_data3[which.min(brain_data3$calculated_volume),1]

## # A tibble: 1 x 1
##   brain_regions         
##   <chr>                 
## 1 Rostral middle frontal

17. Task (4 points)

Here’s a vector with some interesting names:

family <- c("Gomez", "Morticia", "Pugsley", "Wednesday", "Uncle Fester", "Grandma")

Use the function map_chr (which is an iterator that returns a character vector) and a custom-made anonymous function which uses the function str_c (which concatenates strings) to append the string " Adams" to all family members, except Uncle Fester and Grandma. The output should look like this:

## [1] "Gomez Adams"     "Morticia Adams"  "Pugsley Adams"   "Wednesday Adams"
## [5] "Uncle Fester"    "Grandma"