Sciencemadness Discussion Board
Not logged in [Login ]
Go To Bottom

Printable Version  
Author: Subject: Graph info to find correlation
Yttrium2
Flammable Solid
*****




Posts: 660
Registered: 7-2-2015
Member Is Online

Mood: Snarky like a shark!

[*] posted on 10-10-2019 at 18:33
Graph info to find correlation


Lets say I want to make a graph

x axis is amount of water I drink


y axis is the hours I study



this will show a "correlation" between water drank & hours studied





how can I throw in a 3rd element, such as miles walked, to see how it affects hours studied?--


Is there a way to collect data on several "elements" to see how they affect hours studied?

(seems they'd have to be done one at a time...)


--


--

--


I'm trying to narrow down what affects how many hours I study.
Seems like I'd have to keep things constant
And manipulate 1 "element" (variable?) - at a time.
I've heard "experiments" can be done when 2 variables get manipulated.
But then how is there "control", or a determination of what causes what?


I'm thinking more exercise, less food, losing weight, drinking plenty of water will all contribute to hours studied.





Any info on the topic appreciated


p.s. maybe I've gotten ahead of myself by trying to throw in a 3rd variable, perhaps I should be more concerned with how to design the graph, and how much of a distance there is between increments for water drank. (I'm bad at excel, one day I'll take a course but It'll probably come more handy when I'm finished with math)





[Edited on 10/11/2019 by Yttrium2]
View user's profile View All Posts By User
j_sum1
Administrator
********




Posts: 5174
Registered: 4-10-2014
Location: Oz
Member Is Online

Mood: Metastable, and that's good enough.

[*] posted on 10-10-2019 at 20:08


The word you are after is multivariable analysis.
Unless you have the mathematical understanding to support your investigation, it is a lot simpler to consider just two variables at a time.

For the three variables you have chosen, I suspect any correlations you find will be pretty meaningless anyway.
The dependent variable, hours studied, is likely to be dependent on many many variables including motivational factors that are difficult to quantify.
You are also going to have a rather small data set -- there is a rule of thumb that suggests a minimum n of 30 is needed to make any meaningful conclusions. If you consider a data set larger than 30 you also need to include time dependence as one of yoru variables.
What is the point of all this? Quit procrastinating and just do your work. (Even if it is your 31st birthday. <cough>)

[Edited on 11-10-2019 by j_sum1]
View user's profile View All Posts By User
SWIM
International Hazard
*****




Posts: 690
Registered: 3-9-2017
Location: 26 Ancho street
Member Is Offline

Mood: fractious

[*] posted on 11-10-2019 at 10:50


Try keeping your schoolbooks in the bathroom.

This will encourage a direct relationship between your frequency of study and your fluid intake.

You can probably get in a lot more bathroom study time if you give up all roughage in your diet, but this practice may prove to be a bit of a strain:mad::mad::mad::




Ebay says they need to get their hands on my bank account if I want to keep selling there.
This sounds like the best idea since putting ortho tricresyl phosphate in Ginger Jake.
I'm walking while I can still walk straight.




View user's profile View All Posts By User
sodium_stearate
Hazard to Others
***




Posts: 234
Registered: 22-4-2011
Location: guard duty at the checkpoint
Member Is Offline

Mood: "may we see your papers, please?"

[*] posted on 12-10-2019 at 07:47
a more realistic graph


Why don't you try making a graph that shows
the correlation between how many inane posts
you make here, vs how pissed off everyone else
here is getting from reading them?





"Opportunity is missed by most people
because it is dressed in overalls and it
looks like work" T.A. Edison
View user's profile View All Posts By User
unionised
International Hazard
*****




Posts: 4244
Registered: 1-11-2003
Location: UK
Member Is Offline

Mood: No Mood

[*] posted on 12-10-2019 at 09:21


Quote: Originally posted by Yttrium2  
Lets say I want to make a graph

x axis is amount of water I drink


y axis is the hours I study



this will show a "correlation" between water drank & hours studied





how can I throw in a 3rd element, such as miles walked, to see how it affects hours studied?--


Is there a way to collect data on several "elements" to see how they affect hours studied?

(seems they'd have to be done one at a time...)


--


--

--


I'm trying to narrow down what affects how many hours I study.
Seems like I'd have to keep things constant
And manipulate 1 "element" (variable?) - at a time.
I've heard "experiments" can be done when 2 variables get manipulated.
But then how is there "control", or a determination of what causes what?


I'm thinking more exercise, less food, losing weight, drinking plenty of water will all contribute to hours studied.





Any info on the topic appreciated


p.s. maybe I've gotten ahead of myself by trying to throw in a 3rd variable, perhaps I should be more concerned with how to design the graph, and how much of a distance there is between increments for water drank. (I'm bad at excel, one day I'll take a course but It'll probably come more handy when I'm finished with math)





[Edited on 10/11/2019 by Yttrium2]


This may help
https://en.wikipedia.org/wiki/Analysis_of_variance
View user's profile View All Posts By User
elementcollector1
International Hazard
*****




Posts: 2666
Registered: 28-12-2011
Location: The Known Universe
Member Is Offline

Mood: Molten

[*] posted on 12-10-2019 at 14:00


In practice, this is generally done by setting your third variable to a constant (e.g. '5 miles walked', '10 miles walked', etc.) and graphing your first two variables given that constant. Multiple lines for multiple values of the third variable are a very good indicator of its relationship to the other two.



Elements Collected:52/87
Latest Acquired: Cl
Next in Line: Ge
View user's profile View All Posts By User

  Go To Top