Week 9/10/11- Introduction to Linear Regression

statistical significance

= likelihood of observing this relationship

correlations vs scatterplots

correlations QUANTIFY relationships

scatterplots VISUALISE relationships

correlation coefficent

size of the effect

pipe (>%>) operator

“and then“ in code

indicates the strongest correlation between two variables? 

• r = -1

A correlation coefficient ( r ) of 1 indicates: 

A perfect positive correlation

Symbol that allows you to directly access variable from data frame

• $

reject the null hypothesis

A p-value of .04 means that we should:

True

A correlation between two variables tells us something about the relationship between them, but cannot tell us if one variable causes the other. T/F?

Linear regression

is a modelling technique which allows us to make predictions about an outcome variable based on one or more predictor variables

correlation vs regression

corr= strength of rel between 2 contnoius variables

regr= how much one affects the other, can predict a score

True

Linear regression allows us to make predictions about one variable based on the value of another variable. T/F?

tells us how much of the variance in the outcome variable is explained by the predictor variable

R squared

2

If we run a linear regression and get a coefficient of 4 and a standard error of 2 for our predictor variable what is the value of the t-statistic? 

The value of the outcome variable when the value of the predictor variable(s) is zero

the definition of the intercept? 

model <- lm(Weight ~ Height, data = NHANESraw) 

If we wanted to run a linear regression in RStudio to look at whether Height predicts Weight using the NHANESraw dataset?

predicting values formula

to use the collected sample to make generalisation about population sample was taken from

inferential statistics is for

inferential statistics involves (2)

1) estimation of parameters

2)testing statistical hypothese

continuous data

data with possible range e.g

height

weight

discrete data

set data e.g.

car has 4 wheels

To use a t-test the data type for the dependent variable must be

The variance

A t-test examines the difference between two groups of data, whilst also accounting for???

1. The t-statistic (also known as the t-value) 

2. The degrees of freedom 

3. The p-value 

When we conduct a t-test, we are given three pieces of information in our output. These are: 

code for repeated-measures t-test?

paired_t_test()