In 2015, I realized that I generate a lot of data from skateboarding and I had difficulty figuring out which tricks I was good at and kept losing in skateboarding competitions. The reasons for failure were not only my lack of a variety of tricks but also, how to keep the tricks consistent. I don't trust my gut feeling completely when it comes to figuring out what tricks I am good at. I prefer to follow a set of tricks that I know with some certainty that I will be able to do with ease. Thus, the skate project was born, I decided for a short period of time that I would record the number of times — I'd do a trick consistently. It was in a scale of 0-5, I even made a playlist where I explain some bits of the project which you can follow up in the video below; It is a playlist with 9 videos approximately 2-3 minutes long.
https://www.youtube.com/playlist?list=PLwteCCP_K4M6AinSMeSZL2ULwMnHps4ki
If you don't fancy checking out videos, below is the codebook that explains the experiment as a whole, how I collected the data and the tricks under study.
This involves loading the data in your preferred programming language environment; It is then followed by removing unnecessary columns, removing data inconsistencies or replacing them with something more usable. For example if the columns were in uppercase and changing them to lowercase, maybe there's just an empty space in one of the rows of the dataframe and then you delete that row.
Load a couple of packages that will be useful in the subsequent steps. We import them in one code cell so that in case I accidentally delete one of the import we don't need to go searching through the code to find it besides it's more organized with this formatting, right? If don't know or aren't sure what any of these packages do visit https://www.rdocumentation.org and search for the name of the package to learn what it can enable you to do.
library(psych)
library(tidyverse)
library(magrittr)
library(lubridate)
library(Hmisc)
library(plotly)
library(purrr)
library(M3C)
library(RColorBrewer)
Load the data in R programming language. You can find an updated version here
# specify where the skate_project csv file is
path <- file.path("~/skate_project1.csv")
# use the read_csv which helps parse the types of each column
skate_df <- read_csv(path)
# dimensions of the skate_df. The number of rows and columns of the dataset
dim(skate_df)
# select 100 rows for this experiment
# take a slice of the data just 100 rows
skate_df <- skate_df %>% slice(1:100)
# 100 rows and 26 columns