Exam 1 Scope and Method

R

A programming language and software environment used for statistical computing and graphics, widely used among statisticians and data miners.

R Studio

An integrated development environment (IDE) for R, providing tools for coding, debugging, and visualizing data.

Upper left window is called

R script (input)

Lower left window is called

R console (output)

upper right window is

environment (lists created objects)

##

distinguishes output from code

←

to create an object, left is name right is content (four← 1+3)

what should NOT be in quotes

names of objects, functions, and arguments as well as special values (TRUE, FALSE, NA, NULL)

required arguments

inputs we must specify in order to use a particular function

optional arguments

inputs that we could specify if we wanted to modify the functions default setting

sqrt()

square root

#

to comment on the code so R doesn’t read it

setwd()

set working directory

view()

opens a new tab in the upper left window with contents of data set

head()

shows first six rows of observations in a data set, n=x to display different amounts of lines

observation

the information collected from a particular individual in the study

i

row in the dataframe

variable

captures values of a changing characteristic for the multiple individuals

dim()

provides the dimensions of the data frame

$

access values of a single variable

nonbinary

average in nonprecentage

average causal effect

the average of all the individual causal effects of x on y within a group

randomized experiment

treatment assignment is random, ensuring that treatment and control groups are identical to eachother in all observed and unobserved pre-treatment characteristics

average effect formula

ytreatment-ycontrol

observational data

collected from naturally occurring events

==

measures whether two values are equal to eachother (TRUE or FALSE)

ifelse()

if logical test is true, return this, else, return that

““

around words

[]

to the left, we specify variable we want to subset, inside, specify the criterion of selection

to calculate average reading scores among only students who attended the small class

mean(star$reading[star$small==1])

nonbinary

same unit of measurement as the outcome variable

binary

percentage points (multiply result by 100)

factual outcome

potential outcome under the treatment condition taken in reality

counter factual outcome

potential outcome under the opposite treatment condition as the one taken in reality

reasons one couldn’t do a randomized experiment

ethical, logical, financial

ifelse must have how many arguments

3 arguments, logical test/return if TRUE/ return if FALSE

random sampling

randomly selecting individuals from the population

sampling frame

the complete list of individuals in a population

unit nonresponse

some randomly selected individuals might refuse to participate in the study

item nonresponse

when an individual chooses to not answer a question

misreporting

participants provide false information

Read and store data set

bes← read.csv(“BES.CSV”)

frequency table

shows the values the variable takes and the number of times each value appears in variable, table()

prop.table()

create a table of proportions

exclude=NULL

this will ensure that the function includes the missing values

na.rm

“NA Remove” setting equal to true

na.omit()

remove missing values

two-way frequency table

shows the number of observations that take each combination of values of two specified variables

two-way table of proportions

shows the proportion of observations that take each combination of values of two specified variables

density histogram

uses density instead of frequency as the height

descriptive statistics

numerically summarizes the main characteristics of its distribution

plot()

scatter plot

abline()

add straight line to the graph in R, vertical set argument equal to v, horizontal set argument equal to h

correlation coefficient

summarizes the direction and the linear association between 2 variables (-1 to 1)

z-score

the number of sd’s the observation is above or below the mean

cor()

correlation coefficient

acquiesces bias

select positive response (yes)

social desirability bias

select answer that makes them look good or cool

outcome variables

measures the attitude we are trying to explain

explanatory variables

measures attitude or experience we think explains the outcome

!

flips the value

Wording Pitfalls

Jargon or too much info

▶ Leading descriptions

▶ Double negatives

▶ Compound/double-barreled questions

▶ Ambiguous or vague terms

difference in mean

Δ ̂𝑌 = 𝑌𝑡𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡−𝑌𝑐𝑜𝑛𝑡𝑟𝑜

What is partisan expressive response?

A partisan chooses the survey response option that portrays their ‘team’ in a positive light (or the other team in a negative light), even though that response is inconsistent with their real beliefs or attitudes