Correlation and regression

Question 1

Following stuff im gonna add:

• Histograms & cumulative frequency :D
• Box plots and skewness :D
• Scatter diagrams and regression :D
• Numerical measures ,’:(

Answer 1

Ya

Question 2

What histogram?

Answer 2

The bar graph thing with the frequency density on y axis, and some rando value on x axis

Question 3

What cumulative frequency?

Answer 3

The line graph thing that can be correlated with boxplot.

Question 4

What boxplot?

And how to write/find values?

Answer 4

Is a boxplot with line corresponding to these:

• Lowest value (is just the lowest…)
• LQ (1/4 of whole thing)
• Median ( the middle point)
• UQ (3/4 of whole thing)
• Highest value (is just the highest…)

Correlates with cumulative frequency

Question 5

How do u find the frequency density?

Answer 5

frequency/class width

Question 6

How to find frequency only?

Answer 6

The area of the bar

Question 7

How do u find IQR?

Answer 7

UQ - LQ = IQR
(3/4) - (1/4) = ^^^

Question 8

How do u compare distributions of box plots?

Answer 8

You just talk about literally the values man, compare them

Question 9

What are outliers?
What formula? Not in data booklet
and how to find them?

Answer 9

Data points found below 1.5 x IQR the LQ OR above 1.5 x IQR the UQ

Formula:
(Q1 - (1.5 X IQR))
(Q3 - (1.5 x IQR))

How to find them:
1) You got ur data set
2) Find ur UQ & LQ:
- U either got it ez with boxplot
- Or, find it urself from the given data set by:
Find median (n+1/2)
Then it’s ur 1/4 or 3/4 typ shi

3) Find ur outliers using formula:
- For Q1, it’s numbers below the number u calculated
- For Q3, it’s numbers above “

That bout’ it

Question 10

What say if no outliers found?

Answer 10

No boundaries….

Question 11

How to tell if box plot positively skewed?
whats it mean too?

Answer 11

Median more closer or LQ

When mean > median. Data constitute higher frequency of high valued scores

Question 12

How to tell if box plot negatively skewed?
whats it mean too?

Answer 12

Median more closer to UQ

When mean < median. Data constitute higher frequency of low valued scores

Question 13

How to tell if it’s symmetric skew from some rando ahh bar graph?

Answer 13

It’s just in the middle…
also similarly to box plot for finding if positive/negative skew

Question 14

In a graph, how to find out if it’s a perfect positive correlation or weak?

Answer 14

Depends on how closer each point is towards the “line of best fit”

Question 15

How to answer them regression questions?

Answer 15

1) You are given the regression equation
2) Interpreted with y=mx + c or y = a+ bx
3) How to interpret y or m in context?:
- Look in book, this too big D:
- Check photos
(this probably the hardest part for me for regression)

4) Most of the time tho, u just sub in the certain value within equation to find ur answers.
For finding out whether ur answer is accurate or not from question, based on the data range. Like if it’s within data range or not.

Question 16

Not too sure how to explain these measures of spread. But from what I see:
- Given data set
- Use calculator to find values of these:
1. Mean (x̄)
2. n (wth is that)
3. Sum of x values (∑x)
4. Sum of x² values (∑x²)
5. Population variance (σ²x)
6. Sample variance (s²x)
7. Population standard deviation (σx)
8. Sample standard deviation (sx)
9. Lower quartile (Q1)
10. Upper quartile (Q3)
11. Median (med)

Answer 16

Oki doki?

Question 17

Additionally, there’s in depth detail for mean and standard deviation for when u aren’t given some data table. Like there’s some rando formula for this sooooo… ye look in book?

Answer 17

…

Question 18

I’ll perhaps make another deck exclusively for each definition of the symbols 2 flashcards before

Answer 18

Makes sense