Scatter Diagrams And Correlation (4) Flashcards

1
Q

What is association

A

Two variables with a relationship between them

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

What data do scatter diagrams show

A

Bivariate data

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

Where do you plot the independent variable

A

On the horizontal x axis

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

What does the independent variable show

A

It shows the value you change

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

Where do you plot the dependent variable

A

On the y axis

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

What does the dependent variable show

A

The value you do not change

It depends on the explanatory variable

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

What is correlation

A

The link between two variables that shows a trend

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

What is positive correlation

A

When one variable increases another increases

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

What is negative correlation

A

When one increases the other decreases

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

What is linear correlation

A

When they lie on a straight line

(This can be positive or negative)

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

What is non linear correlation

A

Correlation that shows a curved line

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

What is a causal relationship

A

When a change in one variable directly causes a change in another variable.

E.g. The larger a fuel tank the more fuel a car uses

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

Does correlation directly imply causation

A

No
Correlation doesn’t always mean there is a causal relationship.

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

How many factors will cause a change in variables

A

Multiple.
Although a single scatter graph will show how one factor effects another there may be multiple

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

What is a line of best fit

A

A straight line drawn so the plotted points on a scatter diagram are evenly scattered either side of the line

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is a mean point used for

A

Drawing a more accurate line of best fit (than by eye)

17
Q

How do you calculate the mean point

A

The mean x value and mean y value

18
Q

What is interpolation

A

Using the line best fit within the given data values to estimate a value

19
Q

What is extrapolation

A

Extending the line of best fit further than the given values, to estimate a data value

20
Q

How reliable is interpolation or extrapolation

A

Interpolation - reliable as its within the known data values

Extrapolation - unreliable as it is past the known data values

21
Q

What is the equation for a line / line of best fit

A

Y = mx + c

In statistics you are likely to see

Y = ax + b

A = gradient
B = y intercept

22
Q

What is the line of best fit also called

A

The regression line

23
Q

What does the line of best fit of a gradient show

A

The rate of increase of the dependent variable (response variable) to the independent (explanatory) variable

24
Q

What does the y intercept show

A

The response variables value when the explanatory variable is 0

25
What is Spearman's rank coefficient
r↓s Shows the strength of correlation
26
On Spearman's rank what does each significant value between 1 and 0 show
1 - Perfect positive 0.8 - strong positive 0.4 - weak positive 0 - no correlation Negative is the opposite
27
How do you calculate spearman's rank coefficient
First rank your values using the same scale Then subtract the two values for your d value N is the number of groups Next square this r↓s = 1- (6× sum of d^2 /n(n^2-1)
28
What does Pearsons product moment correlation coefficient show
It is similar to spearman's rank correlation coefficient It tests for linear correlation. It tells us how far the data points are from the regression line It is measured between 1 and -1