HW4_Koenig
Isabella Koenig
2022-11-01
Homework assignment #4
Your full name: Isabella König
Date: 19.10.2022/02.11.2022
TASK 1: 5 points. Short EDA report
The main aim of the task is to import data into R, perform brief explanatory analysis, build at least one plot, export data using R Markdown and ‘knit’ by creating an html.file
1. Go to https://andmed.stat.ee/en/stat and create a data set you want to explore. Import the data into R.
library(readxl)
library(readr)
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.3.6 ✔ dplyr 1.0.10
## ✔ tibble 3.1.8 ✔ stringr 1.4.1
## ✔ tidyr 1.2.1 ✔ forcats 0.5.2
## ✔ purrr 0.3.5
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag() library(rmarkdown)
library(knitr)Importing data3.xlsx which includes data on the registration of marriages by month in Estonia:
marriages_estonia <- read_xlsx("data3.xlsx")## New names:
## • `` -> `...2`
## • `` -> `...3`
## • `` -> `...4`
## • `` -> `...5`
## • `` -> `...6`
## • `` -> `...7`
## • `` -> `...8`
## • `` -> `...9`
## • `` -> `...10`
## • `` -> `...11`
## • `` -> `...12`
## • `` -> `...13`
## • `` -> `...14`
## • `` -> `...15`2. Prepare EDA (Explanatory Data Analysis). Write at least 5 questions and provide your answers during the first stage of EDA.
EDA AND DATA PREPARATION
Creating a data frame for the imported data:
df_marriages <- data.frame(marriages_estonia)Question: Are there any Null / NA values?
Yes, there are and I will remove them.
Removing rows that only have NA values:
new_df_marriages <- df_marriages[-c(1, 2, 9:33), -c(1, 2)]Question: Do the names of columns and rows make sense? Should something be changed about it?
Currently, the names don’t make sense so I will change them.
Changing the name of the columns:
colnames(new_df_marriages) <- as.character(c(Month = "January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December", "Sum per year"))Changing the name of the rows:
rownames(new_df_marriages) <- as.character(c(Year = "2016", "2017", "2018", "2019", "2020", "2021"))Shows how the data frame looks like now:
new_df_marriages## January February March April May June July August September October
## 2016 264 304 311 332 411 603 1184 1014 567 384
## 2017 241 284 340 323 433 601 1242 965 588 373
## 2018 248 262 248 323 415 648 920 1292 458 393
## 2019 276 299 340 348 521 644 887 964 596 395
## 2020 272 548 272 218 277 480 920 1032 499 491
## 2021 251 260 286 279 388 584 1077 920 520 429
## November December Sum per year
## 2016 317 380 6071
## 2017 305 355 6050
## 2018 354 386 5947
## 2019 324 305 5899
## 2020 269 350 5628
## 2021 374 406 5774Question: Which data are we even looking at: what is the observation unit and the reference period?
Answer: As we can see in the above data frame, we look at the number of marriages in Estonia for the years 2016, 2017, 2018, 2019, 2020 and 2021.
Here is a summary of the data:
summary(new_df_marriages)## January February March April
## Length:6 Length:6 Length:6 Length:6
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
## May June July August
## Length:6 Length:6 Length:6 Length:6
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
## September October November December
## Length:6 Length:6 Length:6 Length:6
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
## Sum per year
## Min. :5628
## 1st Qu.:5805
## Median :5923
## Mean :5895
## 3rd Qu.:6024
## Max. :6071Question: What are the data types of the variables? Do we need to change them?
Answer: I will change the data for the marriages per month from character into numeric data and create vectors (will need this later).
Converting the numbers for 2016 into numeric data, calculating the sum and assigning it to marriages_2016:
marriages_2016 <- sum(as.numeric(c(new_df_marriages[1,1:12])))
marriages_2016## [1] 6071Converting the numbers for 2017 into numeric data, calculating the sum and assigning it to marriages_2017:
marriages_2017 <- sum(as.numeric(c(new_df_marriages[2,1:12])))
marriages_2017## [1] 6050Converting the numbers for 2018 into numeric data, calculating the sum and assigning it to marriages_2018:
marriages_2018 <- sum(as.numeric(c(new_df_marriages[3,1:12])))
marriages_2018## [1] 5947Converting the numbers for 2019 into numeric data, calculating the sum and assigning it to marriages_2019:
marriages_2019 <- sum(as.numeric(c(new_df_marriages[4,1:12])))
marriages_2019## [1] 5899Converting the numbers for 2020 into numeric data, calculating the sum and assigning it to marriages_2020:
marriages_2020 <- sum(as.numeric(c(new_df_marriages[5,1:12])))
marriages_2020## [1] 5628Converting the numbers for 2021 into numeric data, calculating the sum and assigning it to marriages_2021:
marriages_2021 <- sum(as.numeric(c(new_df_marriages[6,1:12])))
marriages_2021## [1] 5774Calculating the sum of all marriages from 2016 to 2021 and assigning it to sum_marriages:
sum_marriages <- sum(marriages_2016, marriages_2017, marriages_2018, marriages_2019, marriages_2020, marriages_2021)
sum_marriages## [1] 35369Creating a vector with the numbers of marriages per year and assigning it to ‘marriages_per_year’:
marriages_per_year <- c(marriages_2016, marriages_2017, marriages_2018, marriages_2019, marriages_2020, marriages_2021)
marriages_per_year## [1] 6071 6050 5947 5899 5628 5774Question: How is the data distributed? Is it skewed?
If we look at the years 2016-2021, we do not see a normal distribution, the data is also not skewed. It looks more like a uniform distribution (each year has a similar number of marriages).
Question: Are there any outliers (when it comes to the marriages per year)?
No, there are no outliers (which is already apparent from the barplot but we can also look at it with a scatterplot). The number of marriages for the years 2016 to 2021 ranges from 5628 to 6071.
3. Provide brief descriptive statistical analysis of your data set (like measures of central tendency and dispersion).
It makes sense to look at the data per year (it could also make sense to look at the individual months).
CENTRAL TENDENCY (for the individual years):
2016
Calculating the mean of all marriages in 2016 and assigning it to ‘mean_marriages_2016’:
mean_marriages_2016 <- mean(as.numeric(c(new_df_marriages[1,1:12])))
mean_marriages_2016## [1] 505.9167It does not make sense to have 505.9167 marriages, therefore I convert the number to integer:
mean_marriages_2016 <- as.integer(mean_marriages_2016)
mean_marriages_2016## [1] 5052017
Calculating the mean of all marriages in 2017 and assigning it to ‘mean_marriages_2017’:
mean_marriages_2017 <- mean(as.numeric(c(new_df_marriages[2,1:12])))
mean_marriages_2017## [1] 504.1667It does not make sense to have 504.1667 marriages, therefore I convert the number to integer
mean_marriages_2017 <- as.integer(mean_marriages_2017)
mean_marriages_2017## [1] 5042018
Calculating the mean of all marriages in 2018 and assigning it to ‘mean_marriages_2018’:
mean_marriages_2018 <- mean(as.numeric(c(new_df_marriages[3,1:12])))
mean_marriages_2018## [1] 495.5833It does not make sense to have 495.5833 marriages, therefore I convert the number to integer
mean_marriages_2018 <- as.integer(mean_marriages_2018)
mean_marriages_2018## [1] 4952019
Calculating the mean of all marriages in 2019 and assigning it to ’mean_marriages_2019:
mean_marriages_2019 <- mean(as.numeric(c(new_df_marriages[4,1:12])))
mean_marriages_2019## [1] 491.5833It does not make sense to have 491.5833 marriages, therefore I convert the number to integer:
mean_marriages_2019 <- as.integer(mean_marriages_2019)
mean_marriages_2019## [1] 4912020
Calculating the mean of all marriages in 2020 and assigning it to ‘mean_marriages_2020’:
mean_marriages_2020 <- mean(as.numeric(c(new_df_marriages[5,1:12])))
mean_marriages_2020## [1] 4692021
Calculating the mean of all marriages in 2021 and assigning it to ‘mean_marriages_2021’:
mean_marriages_2021 <- mean(as.numeric(c(new_df_marriages[6,1:12])))
mean_marriages_2021## [1] 481.1667It does not make sense to have 481.1667 marriages, therefore I convert the number to integer:
mean_marriages_2021 <- as.integer(mean_marriages_2021)
mean_marriages_2021## [1] 481Question: How many marriages have happened in Estonia on average per year from 2016 to 2021?
Calculating the mean of marriages per year from 2016 to 2021 and assigning it to ‘mean_total’:
mean_total <- mean(c(marriages_2016, marriages_2017, marriages_2018, marriages_2019, marriages_2020, marriages_2021))
mean_total## [1] 5894.833as.integer(mean_total)## [1] 5894On average, 5894 marriages per year happened in Estonia from 2016 to 2021.
DISPERSION (for the individual years)
2016
Calculating the variance of all marriages in 2016 and assigning it to ‘var_marriages_2016’:
var_marriages_2016 <- var(as.numeric(c(new_df_marriages[1,1:12])))
var_marriages_2016## [1] 88550.27Calculating the standard deviation of all marriages in 2016 and assigns it to ‘sd_marriages_2016’:
sd_marriages_2016 <- sd(as.numeric(c(new_df_marriages[1,1:12])))
sd_marriages_2016## [1] 297.5742017
Calculating the variance of all marriages in 2017 and assigning it to ‘var_marriages_2017’:
var_marriages_2017 <- var(as.numeric(c(new_df_marriages[2,1:12])))
var_marriages_2017## [1] 94078.15calculating the standard deviation of all marriages in 2017 and assigns it to ‘sd_marriages_2017’:
sd_marriages_2017 <- sd(as.numeric(c(new_df_marriages[2,1:12])))
sd_marriages_2017## [1] 306.72162018
Calculating the variance of all marriages in 2018 and assigning it to ‘var_marriages_2018’:
var_marriages_2018 <- var(as.numeric(c(new_df_marriages[3,1:12])))
var_marriages_2018## [1] 99551.36Calculating the standard deviation of all marriages in 2017 and assigns it to ‘sd_marriages_2018’:
sd_marriages_2018 <- sd(as.numeric(c(new_df_marriages[3,1:12])))
sd_marriages_2018## [1] 315.51762019
Calculating the variance of all marriages in 2019 and assigning it to ‘var_marriages_2019’:
var_marriages_2019 <- var(as.numeric(c(new_df_marriages[4,1:12])))
var_marriages_2019## [1] 55810.45Calculating the standard deviation of all marriages in 2017 and assigns it to ‘sd_marriages_2019’:
sd_marriages_2019 <- sd(as.numeric(c(new_df_marriages[4,1:12])))
sd_marriages_2019## [1] 236.24232020
Calculating the variance of all marriages in 2020 and assigning it to ‘var_marriages_2020’:
var_marriages_2020 <- var(as.numeric(c(new_df_marriages[5,1:12])))
var_marriages_2020## [1] 69069.09Calculating the standard deviation of all marriages in 2020 and assigns it to ‘sd_marriages_2020’:
sd_marriages_2020 <- sd(as.numeric(c(new_df_marriages[5,1:12])))
sd_marriages_2020## [1] 262.812021
Calculating the variance of all marriages in 2021 and assigning it to ‘var_marriages_2021’:
var_marriages_2021 <- var(as.numeric(c(new_df_marriages[6,1:12])))
var_marriages_2021## [1] 69914.88Calculating the standard deviation of all marriages in 2021 and assigning it to ‘sd_marriages_2021’:
sd_marriages_2021 <- sd(as.numeric(c(new_df_marriages[6,1:12])))
sd_marriages_2021## [1] 264.4142FURTHER ANALYSIS
Question: From 2016 to 2021, did more people get married in summer (June, July, August) than in spring (March, April, May)?
Assigning the marriages that happened in March, April and May to ‘marriages_spring’ and calculating the sum:
marriages_spring <- as.numeric(c(new_df_marriages[, 3], new_df_marriages[, 4], new_df_marriages[, 5]))
sum(marriages_spring)## [1] 6065Assigning the marriages that happened in June, July and August to ‘marriages_summer’ and calculating the sum:
marriages_summer <- as.numeric(c(new_df_marriages[, 6], new_df_marriages[, 7], new_df_marriages[, 8]))
sum(marriages_summer)## [1] 15977Indeed, a lot more marriages happened in summer than in spring (more than double in summer compared to spring).
Question: Did less people get married in the year the COVID pandemic started in Europe (2020) compared to previous years?
Comparing marriages in 2020 with those in 2016:
marriages_2020 < marriages_2016## [1] TRUEComparing marriages in 2020 with those in 2017:
marriages_2020 < marriages_2017## [1] TRUEComparing marriages in 2020 with those in 2018:
marriages_2020 < marriages_2018## [1] TRUEComparing marriages in 2020 with those in 2019:
marriages_2020 < marriages_2019## [1] TRUEComparing marriages in 2020 with those in 2021:
marriages_2020 < marriages_2021## [1] TRUELess people got married in Estonia in 2020 than in the years of 2016, 2017, 2018, 2019 and 2021, respectively.
Question: Do we see a steady increase or decrease of marriages from 2016 to 2021?
To answer this question, it is easiest to look at a barplot with the numbers of marriages per year:
Apparently, marriages per year have gone down from 2016 to 2020 and increased a little again in 2021.
4. Include at least one plot into your report. If ggplot2 is too complicated for you now, create a plot with R base functions.
See above plots.
5. Create a pdf or html file with short EDA of your data set. Send a pdf file directly to me, send html code through a gist on GitHub.
See above.