R Data Frame Operations & Sampling Bias in Statistics

sampling bias

a sampling method that consistently over or under represents certain data points (always rounding up or down)

measurment Error

Random inaccuracy in data collection

Sampling variation

natural differences b/w samples drawn from the same population

Rows in a data frame

cases/observations

Columns in a data frame

variables

Quotation marks in R

For text/character data

Symbol ! in R

NOT.

Symbol != in R

Not equal to.

Symbol == in R

Equal to.

- in R

Subtracts OR removes a variable in select().

how to Update part of a vector in R

By reassigning it (vec[2] <- 5)

tally() function

Counts frequencies.

select() function

selects multiple variables and makes a data table (select(df,y1,y2))

filter() function

Keeps only rows that meet a condition.

arrange() function

arrange rows by a variable.

str() or glimspe function

Shows structure of a data frame (lists variables).

factor() function

Coverts quant. to cat. variables

head() function

Shows first 6 rows.

tail() function

Shows last 6 rows.

sample() function

Draws a random sample.

Filter out missing data in R

filter(df, !is.na(variable)) or filter(df, is.na(variable)==False)

Tilde ~ in R

Means 'about.' Used in formulas like y ~ x = 'y depends on x.'

as.numeric

converts cat. to quant.

mutate()

adds or modifies varaibles (mutuate(df, variable = "y")

Quick overview of data frame variables

str(df) or glimpse(df).

Concerns when generalizing from data

sample is not random, representative, or lacks variation.

select()

produces a data table with a subset of chosen variables

recode()

recodes values by putting old ones in quoations and equaling to a new value (df, "0" = 2)

Random sampling

when every object has an equal chance of being chosen in a population

independent sampling

when selecting 1 object does not affect the chance of selecting another