Chapter 4/1 Correlation/Regression (AS/A-Level) Flashcards

1
Q

x-axis

A

Independent/explanatory Variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

y-axis

A

Dependent/response Variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Why may some conclusions not be valid?

A

It is possible that although variables are correlated, there is not necessarily a casual relationship.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the form for a regression line

A

y = a + bx

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Interpret the y-intercept (the a value in y = a + bx)

A

Usually initial bla bla bla

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Interpret the gradient (the b value of y = a + bx)

A

e.g. For every 1 hour of revision, the exam mark increase by bla bla

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is Interpolating?

A

Estimating a value inside the data range

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is Extrapolating?

A

Estimating a value outside the data range

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Exponential equation

A

y = ab^x

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Re write y = ab^x in terms of logs

A

logy = loga + xlogb

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is PMCC?

A

denoted by r, describes the linear correlation between two variables. It can take values between -1 and 1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

r = -1

A

Perfect Negative

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

r = 0

A

No correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

r = 1

A

Perfect Positive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Calculate r on calculator

A

Menu -> 6 -> 2 -> Data Entry -> Regression Calc

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Comment on suitability of linear regression model

A

Because r is close to 1, correlation is very strong therefore a subitable model bla bla bla