In this blog, I will analyze OECD Material productivity data.
First, I downloaded data from the OECD webiste: Materials - Material productivity - OECD Data
OECD (2022), Material productivity (indicator). doi: 10.1787/dae52b45-en (Accessed on 12 February 2022)
The downloaded CSV file looks like below.
I use R to analyze the data, so I firstly load tidyverse package and use read_csv() function to load the CSV file.
Let's check each variables. For character variables, how many vales are there, for numeric variables, summary statistics.
There are many locations.
For INDICATOR, there is only one value, MATPROD, so we can delete INDICATOR.
There are two kinds of vakue, TOTMAT and NONNRGMAT. So I keep SUBJECT.
There is only one value in MEASURE, it is USD_KG. I will remove MEASURE.
There is only one value in FRQUENCY, it is A. I will remove FREQUENCY.
For TIME, it starts in 1980 and ends in 2019. There is not NA value.
There is not NA value in Value.
Now, my new dataframe, df has LOCATION, SUBJECT, TIME and Value.
I know SUBJECT has two kinds of value, one is TOTMAT and the oter is NONNRGMAT.
So, I want to reshape my dataframe df like below
LOCATION TIME TOTMAT NONNRGMAT
AUS 1990 XXX XXX
AUS 1991 XXX XXX
AUS 1992 XXX XXX
So, I will use pivot_wider() function.
Let's check summary statistics.
There are several NA's in NONNRGMAT and TOTMAT. So, I will delete NA rows
By the way, I have GDP and per capita GDP data file like below.
So, I will conbine this GDP data and Material productibity data.
let's see summary stats.
I will investigater further in next post.
The next post is