Lecture 16: Scatter Plots And Correlations Flashcards

(21 cards)

1
Q

Scatterplots

A

A graph used to illustrate the relationship between two quantitative variables

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

Scatterplots measure the relationship between 2 ______ variables

A

Quantitative

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

Measurements in scatter plot samples come in _____

A

pairs (usually from the same subject)

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

Each point on the scatterplot graph represents a _______

A

subject

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

Correlation

A

A number between -1 and 1 indicating the strength of the linear relationship between two variables

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

The more nearly the points “line up” in the scatter plot, the _____ the correlation (in absolute value)

A

Larger

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

Zero correlation

A

No linear relationship

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

Positive correlation

A

Points lean to the right

When one variable’s value increases the other variable’s value increases

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

Negative correlation

A

Points lean to the left

When one variable’s value increases the other variable’s value decreases

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

Most common way to compute a correlation is using the ______ correlation

A

Pearson

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

The symbol for the population correlation is

A

p (rho)

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

The symbol for the sample correlation is

A

r (an estimate of p)

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

Does a correlation have units? If so what are they?

A

They have no units it’s a number silly

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

Will a correlation be affected by the multiplication/division/subtraction/addition of some constant value from all the measurements?

A

Nope it’s not affected by linearly rescaling the units eg. correlation between height and weight is same regardless of whether height is measured in inches or meters

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

The sample correlation is just an estimate of the ________ _______

A

population correlation

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

Where should we compute a CI?

A

Around the sample correlation (the estimate)

17
Q

Could we conduct a p-value for a null hypothesis test?

18
Q

CI for population proportion/population correlation isn’t perfectly symmetric around the estimate due to the hard _____ and ______ on possible values, so unlikely when estimating a population mean, there won’t be exactly the same margin ____ the estimate as _____ the estimate

A

Floor
ceiling

(proportion must be between 0-1 and correlation must be between -1-1)

Above
Below

19
Q

Correlation means Causation true or false?

20
Q

Why does correlation not equal causation

A

Some third variable could be the cause

21
Q

If we need strong evidence of causation that calls for an ______, not a correlational study

A

Experiment! –> no confounding variables/limits them