Chapter 4 Correlation - Year 1 Flashcards
In a scattergraph what are the variables know as?
Are they the IV or the DV
Which axes are they on?
x axis Explanatory variable (IV)
y axis Response variable (DV)
What are the different types of correlation?
Strong/week positive/negative correlation
No correlation
What does causal effect mean?
When 1 axes effects the results of the other
What is a regression line?
A model that approximates the relationship between the variables (in this case a line of a+bx)
What does the line of best fit aim to do?
Be as close to all the points as possible
How do you answer a question that asks you to describe the correlation between to variables?
Week positive correlation between …
How do you justify the use of a regression line?
It is acceptable as the correlation suggests that there is a linear correlation
When is a regression line most suitable?
Why?
When you have data that is exactly in a straight line
It isn’t an estimate any more
What should you only use the regression line to do?
You can only use a regression line to predict the DV within the given range of the data
What is extrapolation?
When we use our data to estimate for a value that is outide our range of data
What is interpolation?
When we use our data to estimate for a value that is inside our range of data
What is a common silly mistake in this topic?
Saying that we can’t do this because it is a graph of x on y when it should be a regression line of x on y
How do you decide which form is a better fit for the data:
What is the ending statement that you should give if y=ab^x is a better fit
You work out the PMCC. If it is at the extreme then it will be a good fit for y=ab^x but if you need to log both sides first then it will be y=kx^n
There is a strong negative linear correlation with the data in the form logM against logE which suggests that a model in the form y=kx^n is a good fit
On which axes will the last letter on a regression line be on?
The x axis
A regression line of y on x
