Question

Correlations please help

Answer 1

@satellite73

Answer 2

were you able to make the scatter plots?

Answer 3

@jim_thompson5910 I have 2 additional questions on top of this one to do in 40 minutes if you are able to help me please.

Answer 4

Yes.

Answer 5

are you able to use a program/calculator to compute the correlation coefficient?

Answer 6

I used SPSS

Answer 7

ok that works

Answer 8

I've never used it, but I know it's a stats program

Answer 9

Yeah.. but I don't know how to read anything to put it into an answer.. this is what I wrote for that one but I dont know if it's right The correlation coefficient tells a person how close the data is to being linear. The number one tells you that you have a line when you connect all the dots and a 0 would be equally spread out. Correlation coefficients tell you if the data is related as 2 variables increase in certain ways. Never assume causation just because you get high correlation. When they are more linear they have a correlation coefficient closer to one. The correlation coefficient is r^2 which is fairly close to 1. Correlation is how well the data relates, when you increase something by 1, the other variable always increase.

Answer 10

I'm to the point of rushing through it since I have 3 questions to do in 30 minutes now...

Answer 11

yeah you pretty much have the general basics of correlation coefficients

if r is near 0, then the data is not really linear (so use another model)
if r is near +1, then the data is pretty likely to be linearly correlated with a positive slope
if r is near -1, then the data is pretty likely to be linearly correlated with a negative slope

of course, correlation does not imply causation

Answer 12

so it sounds like you have it down

Answer 13

Someone else helped me with that one... could you help me with my last 2 questions please

Answer 14

so hopefully I can stop crying and stressing lmao

Answer 15

ok

Answer 16

you'll be fine

Answer 17

Ill post my graph and stuff that I did too

Answer 18

ok go ahead

Answer 19

That's it

Answer 20

so what is the correlation coefficient based on this result SPSS produces?

Answer 21

See that's where I am confused. I'm not sure exactly which number I should be looking at

Answer 22

For some strange reason, they compute the correlation coefficient with a variable and itself (getting a correlation coefficient of 1). I'm not sure why you'd do that since you're always going to get 1 in that case.

Answer 23

The scatter plot shows us we do NOT have a straight line, so the correlation coefficient is clearly not 1.

Answer 24

It turns out that the correlation coefficient is actually 0.14422205

and this is much better since the data is more randomly spread out and not falling near or close to a line

Answer 25

So I look at the GPA one?

Answer 26

and SPSS does produce this result of 0.144

you have to look in the cells which have GPA and hours of study

Answer 27

So I would say my correlation coefficient is 0.144, but how do I explain that?

Answer 28

Well the program produces that value, so you just say that's what SPSS calculated.

And it justifies the scatter plot. The dots don't fit near or on a line, so the data is not linear. The r value is closer to 0 than to 1, so the data is not linear

Answer 29

I don't know what the "sig(two tailed)" is referring to, but ignore it

Answer 30

Ok so I would type...

My correlation coefficient is 0.144 based on the results of SPSS calculation. It justifies the scatter plot because the points don't fit on or near a line, so the data is not linear. The r value is closer to 0 than to 1, so the data is not linear.

Is that right?

Answer 31

yeah, but in your own words that you understand

Answer 32

oK... I have one more thing please

Answer 33

The red is the only answer I think I know

Answer 34

ok one sec

Answer 35

Thank you so much.. I have to leave very soon

Answer 36

sadly I'm not finding much on "predicting the z-score formula", I wonder if they mention that anywhere in your book?

Answer 37

Shoot thank you

Answer 38

yeah I've never heard of it

then again, I haven't taken many stats courses (I've yet to dive into multiple regression and such)

Answer 39

I don't know why I'm so confused lol
thank you for your help on the others though :)

Answer 40

I might have found something here

http://www.learningace.com/doc/7701925/b7621ca500e4a54f7f10a745af4cfa85/chapter-3-correlation-and-prediction

but it'll take me a bit of time to read through it

Answer 41

you're welcome

Answer 42

I found that

Answer 43

ok one sec

Answer 44

I looked more and it says Beta is my regression score.

Answer 45

So what we want is

\[\Large Z_{y} = \beta*Z_{x}\]

but I'm not sure how to find beta

Answer 46

It tells me beta in the first attachmetn

Answer 47

it says the r=.36

Answer 48

are you sure that's beta? it seems too easy

Answer 49

I guess it's the closest thing we have that they give us

Answer 50

I'll attach a couple of steps it says real quick then I have to throw on some make up lol

Answer 51

oh I guess you're right: beta = r

Answer 52

so yeah, it would be

\[\Large Z_{y} = 0.36*Z_{x}\]

Answer 53

So how do you think I would do the prediction

Answer 54

in part B, you

step 1) calculate Zx, the z-score for similarity for x = 5

step 2) plug Zx into the formula above to find Zy

Answer 55

So do I put 5 in for Zx? so .36*5?

Answer 56

no

Answer 57

x = 5, so the z-score Zx is..

Zx = (x - mu)/SD

Zx = (5-3.25)/1.92

Zx = 0.91

Answer 58

Then you plug that into the formula

Zy = 0.36*Zx

Zy = 0.36*0.91

Zy = 0.3276

Zy = 0.33

Answer 59

So my final answer is the 0.33?

Answer 60

yeah for that part

Answer 61

Sorry I'm getting ready for work at the same time lol

Answer 62

oh wait, you'd then solve for y

Zy = (y - M)/SD

0.33 = (y - 6)/2.55

0.33*2.55 = y - 6

0.8415 = y - 6

0.8415+6 = y

y = 6.8415

That's the predicted attraction (so around 7)

Answer 63

So to use the formula Zy = beta*Zx, you have to standardize the score first

then you unstandardize to get it back to a raw score (the predicted raw score)

Answer 64

So that's for part b? the whole thing you just typed? or just this last one?

Answer 65

they all go together

Answer 66

So all 3 parts? and my final prediction is 7?

Answer 67

hopefully you see how i went from x = 5 to Zx = Zx = 0.91

Answer 68

What about part A?

Answer 69

and how I got Zy = 0.33

Answer 70

part A is just the formula itself

Answer 71

so maybe 2 lines at most

Answer 72

and yeah around 7 I'd say

Answer 73

Ok... so what exactly do I write for part A? sorry I'm rushing you I really appreciate it.

Answer 74

just the formula that you attached, the one with beta

then you plug in beta = r = 0.36

Answer 75

Zy = beta*Zx

plug in beta = 0.36

Answer 76

So Zy= .36* what?

Answer 77

Zx

Answer 78

for part b, you are plugging in Zx = 0.91 (you have to show how you got Zx = 0.91)

Answer 79

It shows 4 Zx so how do I know which to put?

Answer 80

4Zx? where are you getting 4?

Answer 81

I have no idea lol

Answer 82

it'd be Zy = 0.36*Zx

Answer 83

I'm just replacing beta with 0.36

Answer 84

they just want you to predict Zy based on Zx

Answer 85

you have to compute Zx first though

Answer 86

I don't know how.

Answer 87

I have 4 minutes lol