R_data <- read.csv("~/R_data.csv")Data Cleaning - Practical
For this practical we will use the data set R_data you can find on Canvas. This is a .csv file. Load the data in R. Make sure you specify the correct file path!
What functions are useful for the first exploration of the data? How many observations and variables are in the data set? What type of variables are there?
There is a typo in one of the variable names (Stauts instead of Status). Change this.
Round the variable folicacid_erys to two decimals.
Verify by evaluating the first 20 values of this new variables (there are several ways to do this).
Make a new variable birhtweight_kg that gives the birth weight in kilo’s What did you choose for the number of decimals to round the variable?
Verify by evaluating the first 10 values of this new variables (there are several ways to do this).
The variables pregnancy_length_weeks and pregnancy_length_days together denote the total length of the pregnancy. For example: pregnancy_length_weeks = 38 and pregnancy_length_days = 4, means this patient is pregnant for 38 weeks plus 4 days. Combine the variables to obtain the length of the pregnancy in days.
Verify by evaluating the first 12 values of this new variables (there are several ways to do this).
Divide the variable BMI into categories: <18.5 (“Underweight”), 18.5 - 24.9 (“Healthy weight”), 25 - 29.9 (“Overweight”), and >30 (Obesity). How many patients (and %) are in each category?
For a current analysis, I am only interested in the patients with “Healthy weight”. Additionally, I only want to look at the relation between Status and birthweight. Make a data set with only these two variables and patientnumber, for a subset of the data with the patients with “Healthy weight”.
What are the dimensions of this data set? First try to think yourself and then check with R code.
Make a third data set containing the variables: patientnumber, Status and BMI. Then merge the data set you created in Question 7 with this data set.
Merging these two data sets can be done several different ways. Describe two ways. What are the dimensions of these data sets?
There are several biomarkers collected in the data set. Investigate whether there are outliers in the biomarkers: cholesterol, triglycerides, and vitaminB12. Which functions did you use?
In question 9 we found an outlier. How do you deal with this outlier?
Fill in the table with summary statistics below
| Intellectual disability | Normal brain development | ||
|---|---|---|---|
| (n = …) | (n = …) | ||
| BMI, median [IQR] | |||
| missing (n = ) | |||
| Educational level | Low, n(%) | ||
| Intermediate, n(%) | |||
| High, n(%) | |||
| missing (n = ) | |||
| Smoking | No, n(%) | ||
| Yes, n(%) | |||
| missing (n = ) | |||
| SAM, mean (SD) | |||
| missing (n = ) | |||
| SAH, median [IQR] | |||
| missing (n = ) | |||
| Vitamin B12, median [IQR] | |||
| missing (n = ) |
You can change the order of factors
R_data$educational_level <- factor(R_data$educational_level, levels = c('low', 'intermediate', 'high'))