R CODE

example() 

example on how to use the function 

?

provides documentation on a specific function or dataset that are part of a package

←

assignment operator

— assigning value to an object / variable

ls()

check how many objects you have created / available

rm()

removes objects that you don’t need

seq()

creates a sequence of numbers

length.out=  argument 

sets a desired length of the sequence

library()

loads packages

every time a new R session is open

when to use the library () function?

int

stands for integers

EX]: 1,2,3

dbl 

stands for doubles or real numbers 

EX]: -1, 1.5, 4/5

continuous

• measured data

can have infinite values within a possible range

EX]: i am 3.1” tall and i weigh 34.16 grams

discrete

observations can only exist at limited values, often counts

EX]: i have 8 legs and 4 spots 

date

stands for dates 

EX]: (01/21/2025)

dttm

stands for date-times, a date + a time

EX]: (01/21/2025 11:00am)

fctr

stands for factors

— R uses to represent categorical variables with fixed possible values

EX]: freshman, sophomore, junior, senior

lgl

stands for logical

— vectors that contain only TRUE or FALSE values

chr

stands for characters

— vectors or strings

EX]: “this is a string”

nominal

unordered descriptions

EX]: “i’m a turtle” and “i’m a butterfly”

ordinal

ordered descriptions

EX]: “i am unhappy” and “i am awesome”

binary

only 2 mutually exclusive outcomes

vector 

the simplest data structure in R which consists of an ordered set of values of the same type (e.g. numeric, charcater, data, etc) 

scalar

a vector of length 1

data frame / tibble / dataset

data structure that organizes data into a 2-dimensional table of rows and columns, much like a spreadsheet.

glimpse()

get a sense of all the columns and their content

str() 

get a more detailed sense of columns and all their contends 

colnames()

know the columns name of your dataset as a list 

functions to get to know your data 

• glimpse()

• str() 

• colnames()

• ?

the 5 + 1 data manipulations

• arrange () 

• filter() 

• select () 

• mutate() 

• summarize()

          — group_by() 

arrange()

reorder / sort observations / rows

filter()

keep observations / rows based on conditions

select()

pick variables / columns

mutate() 

create new variables / columns or update existing ones

summarize()

produce descriptive statistics

group_by()

change the unit of analysis by creating groups based on one of more variables / columns

ends_with()

• select function

matches names that end with whatever is in the ()

starts_with()

• select function

matches names that start with whatever is in the ()

contains()

• select function

matches names that contains whatever is in the ()

transmute()

used to compute the new column only

na.rm=T argument

critical if the column you are using for your average contains missing values

|>

can be used to rewrite multiple operations. think of it like reading “then”

measures of location 

mean() and median()

n()

includes a count 

mean()

the sum divided by the length

median()

a value where 50% of () is above it and 50% is below it

measures of spread

sd()

sd()

the root mean squared deviation

measures of rank 

min() and max() 

min()

identifies the smallest value 

max()

identifies the largest value

data= argument

adds in the dataset to use in the graph, so data is loaded in the background 

mapping=aes()

maps what variables we want to visualize on our axes 

geom

determines the visual structure / shape of the chart

aesthetics

• color / fill

• size

• alpha (transparency)

• shape

choosing the right chart depends on…

• the data type of the columns: is the data numerical or categorial?

• the objective of the chart: what is trying to be conveyed with your chart?

distribution chart

shows how values in a dataset are spread out or clustered.

• highlights the range of data, concentration of data points, and whether data tends to be skewed towards specific values.

EX]: histograms, boxplots, violin, and density plots 

correlation chart 

used to examine the relationship between two (or more) numerical variables. 

! correlation does not imply causation 

EX]: scatter plot, smoothing lines, 2d charts, heatmaps, and correlograms. 

ranking chart

displays how different categories (categorical variables) compare in terms of a certain measure

EX]: bar charts, lollipop charts, dot plots 

evolution chart / time-series chart

shows how a variable changes over time

• highlights trends, patterns, seasonality, or fluctuations over a period

EX]: line chart

geom_col() 

use to show the categorical variable with respect to a numerical variable or if you want to use two variables (x and y) 

geom_bar()

only used to show one variable (x or y) 

static / local

if you put an aesthetic in the geom function, it is… 

• just changes the aesthetic

dynamic / global

if you put an aesthetic in the mapping=aes(), it is… 

• assigns an aesthetic to an x or y variable and automatically includes a legend

facet_wrap()

charts by a single variable

facet_grid()

chart by the combination of two variables

linetype= argument 

adds different line styles (solid, dashed, etc) to differentiate 

se=F argument

removes the gray area in the geom_smooth() function or displays the confidence interval around the smooth 

method=”lm” argument

makes the lines linear in a geom_smooth function