Big Five Example - BiFiX

library(rextor)

In this vignette, you will get a complete Walkthrough of how to use rextor for your own data analysis. We will use Open Source data on Big Five personality traits, available from the iScience Konstanz account on GitHub: https://github.com/iscience-kn/BiFiX.

STAY TUNED!

Step 1: Reading in the Data

We will be reading in the data directly from the iScience GitHub Account. Make sure to select the raw data file on GitHub or just copy the code below.

In this dataset, we do not need the validation variable. It facilitates dropout analysis in some cases (especially in one-item-one-screen designs) but here we will focus on other variables to clean the data¹. In the first step, we will read in the WEXTOR based data and assign in to a variable named “raw”, that will become our data frame.

raw <- read_WEXTOR("https://raw.githubusercontent.com/iscience-kn/BiFiX/refs/heads/main/data/BiFiX_data.csv",
                    keep_validation = FALSE)

Step 2: Data Cleaning

Great! We now have our empirical data in R, ready to be worked with. But there are some things which are not yet ideal, so will use some more rextor functions to make our lives easier.

When you open up the data and look at it, you might notice that the first variables in the data, which are automatically recorded server variables have the prefix “.wx.”. In R, we generally want our variable names to start with a letter for easy access, so we will use the namepref function (for variable name prefix) to change that to “v_”. Notice that this function will not add that prefix to the variables that were deliberately named by someone in WEXTOR. It will only change those variable names that actually have the old prefix that you specify as shown.

data <- namepref(raw, pref_old = ".wx.", pref_new = "v_")

We will also perform a “seriousness check” and filter out data of anyone who did not confirm they want to participate seriously. This has proven to improve data quality.

data2 <- serious_check(data)

Pipe it all together

When preparing data for further analyses it makes sense to have an unchanged raw version of the data in R and then save any changes in a new data frame. However, in order to avoid situations like above with data1, data2 etc., you might want to use pipes to perform several data preparation steps in one go. Since version 4.1.0 R offers a native pipe that looks like this: |>. It allows you to take the output from previous functions as the first input of the next function and thus create a workflow pipeline that is easily adjustable and usually very efficient and clear to read.

final_data <- raw |> 
  namepref(pref_old = ".wx.", pref_new = "v_") |> 
  serious_check()