Experimental Design B/C

Jim_R
Admin Emeritus
Admin Emeritus
Posts: 283
Joined: May 13th, 2001, 4:22 pm
Division: Grad
Location: Pittsburgh, Pa
Contact:

Experimental Design B/C

Postby Jim_R » August 10th, 2014, 10:49 am

-The path of the Administrator is beset on all sides by the inequities of the selfish and the tyranny of evil men.

-Nothing\'s gonna get deleted. We\'re gonna be like three little Fonzies here. And what\'s Fonzie like? Come on, what\'s Fonzie like?
-Cool?
-Correctamundo. And that\'s what we\'re gonna be. We\'re gonna be cool. Now, I\'m gonna count to three, and when I count three, you let go of your mouse, and back away from the keyboard. But when you do it, you do it cool. Ready? One... two... three.

HeavyHitter406
Member
Member
Posts: 45
Joined: April 13th, 2014, 5:50 pm
Division: C
State: IL

Re: Experimental Design B/C

Postby HeavyHitter406 » September 25th, 2014, 3:44 pm

I'm new to division C...how do you do Standard of Deviation? That is part of the Statistics in C, right?
2014 (Wright State/Lisle/Grayslake/Regionals/State/Nationals)

Water Quality: 8/1/2/1/2/3
Experimental Design: 12/1/5/5/8/37
Road Scholar: 12/2/2/3/3/11
Wheeled Vehicle: 32/6/6/2/11/5

actionpotential
Member
Member
Posts: 12
Joined: September 25th, 2014, 5:15 pm
Division: C
State: -

Re: Experimental Design B/C

Postby actionpotential » September 25th, 2014, 5:19 pm

Hey Heavyhitter!

So I've always been confused by that big complicated expression myself. Standard deviation is basically measure of how spread out your data values are.

Analyzed step by step:

1. Take the mean of the data. (Symbol is x with bar over it)
2. For each data point, subtract mean from the data point. Then square this difference.
3. Add up the squares.
4. Divide this sum by number of terms minus 1. (Symbol of number of terms is n.)
5. Take square root.

For example:

Data set: 1, 3, 5, 7, 9

1. Mean = 5
2 and 3. (1 minus 5)^2 + (3 minus 5)^2 + ... = 40
4. 40 divided by (5 minus 1) = 8
5. Square root of 8 = 2.828

Standard Deviation = 2.828


Note:If your data includes possible data points for a population, then in step 4 you would instead divide by the number of terms. But if you're working with a sample of the data, which is probably the case in XPD, you divide by number of terms minus 1.

Ex. If you're analyzing the height of ALL students in your grade level, then you divide by number of data values. If you analyze the height of a randomly selected, representative sample of students in your grade level, you divide by number of data values minus 1.

Alternatively, since you can calculate Standard Deviation on your Scientific/Graphing calculator. Plug data in Stat, and use 1-var stats. Standard deviation is round-greek symbol with line out.



Good luck :D

Phys1cs
Member
Member
Posts: 129
Joined: November 10th, 2013, 6:53 pm
Division: Grad
State: MD

Re: Experimental Design B/C

Postby Phys1cs » September 25th, 2014, 7:07 pm

Hey Heavyhitter!

So I've always been confused by that big complicated expression myself. Standard deviation is basically measure of how spread out your data values are.

Analyzed step by step:

1. Take the mean of the data. (Symbol is x with bar over it)
2. For each data point, subtract mean from the data point. Then square this difference.
3. Add up the squares.
4. Divide this sum by number of terms minus 1. (Symbol of number of terms is n.)
5. Take square root.

For example:

Data set: 1, 3, 5, 7, 9

1. Mean = 5
2 and 3. (1 minus 5)^2 + (3 minus 5)^2 + ... = 40
4. 40 divided by (5 minus 1) = 8
5. Square root of 8 = 2.828

Standard Deviation = 2.828


Note:If your data includes possible data points for a population, then in step 4 you would instead divide by the number of terms. But if you're working with a sample of the data, which is probably the case in XPD, you divide by number of terms minus 1.

Ex. If you're analyzing the height of ALL students in your grade level, then you divide by number of data values. If you analyze the height of a randomly selected, representative sample of students in your grade level, you divide by number of data values minus 1.

Alternatively, since you can calculate Standard Deviation on your Scientific/Graphing calculator. Plug data in Stat, and use 1-var stats. Standard deviation is round-greek symbol with line out.



Good luck :D
Standard deviation is somewhat different from other statistics needed for this event. Unlike the mean, which is an average (a reference number), or the range, (a specific reference point for data values), standard deviation is how much your data should deviate from the mean. Within +/- 3 standard deviations of the mean, (if your mean is 5 and your deviation 1, that would be from points 2 through 8) around 97% of your data should be there. It makes a nice bell curve, if your data is good. Something like 70 ish % should be one standard deviation away.

So be careful when explaining the significance of this particular stat

User avatar
Unome
Moderator
Moderator
Posts: 4084
Joined: January 26th, 2014, 12:48 pm
Division: Grad
State: GA
Location: somewhere in the sciolyverse

Re: Experimental Design B/C

Postby Unome » September 26th, 2014, 4:18 am

Oh... I always did standard deviation without the squares.
Userpage
Chattahoochee High School Class of 2018
Georgia Tech Class of 2022

Opinions expressed on this site are not official; the only place for official rules changes and FAQs is soinc.org.

awesome90220
Member
Member
Posts: 158
Joined: March 10th, 2012, 5:19 pm
Division: B
State: AL
Location: somewhere on this cruel, harsh planet

Re: Experimental Design B/C

Postby awesome90220 » September 30th, 2014, 11:28 am

Apart from the usual mean, median, mode, standard deviation, and line of best fit, what are some other statistics that might come in handy at the competition(divison b)
2016 Season: BISOT/Reg/State/Nats
Wind Power:9/1/1/11
Experimental Design:5/1/1/16
It's About Time:-/1/1/20

HeavyHitter406
Member
Member
Posts: 45
Joined: April 13th, 2014, 5:50 pm
Division: C
State: IL

Re: Experimental Design B/C

Postby HeavyHitter406 » October 7th, 2014, 8:04 pm

Apart from the usual mean, median, mode, standard deviation, and line of best fit, what are some other statistics that might come in handy at the competition(divison b)
I think they look for range (highest data value-lowest data value).
2014 (Wright State/Lisle/Grayslake/Regionals/State/Nationals)

Water Quality: 8/1/2/1/2/3
Experimental Design: 12/1/5/5/8/37
Road Scholar: 12/2/2/3/3/11
Wheeled Vehicle: 32/6/6/2/11/5

HeavyHitter406
Member
Member
Posts: 45
Joined: April 13th, 2014, 5:50 pm
Division: C
State: IL

Re: Experimental Design B/C

Postby HeavyHitter406 » October 7th, 2014, 8:05 pm

Hey Heavyhitter!

So I've always been confused by that big complicated expression myself. Standard deviation is basically measure of how spread out your data values are.

Analyzed step by step:

1. Take the mean of the data. (Symbol is x with bar over it)
2. For each data point, subtract mean from the data point. Then square this difference.
3. Add up the squares.
4. Divide this sum by number of terms minus 1. (Symbol of number of terms is n.)
5. Take square root.

For example:

Data set: 1, 3, 5, 7, 9

1. Mean = 5
2 and 3. (1 minus 5)^2 + (3 minus 5)^2 + ... = 40
4. 40 divided by (5 minus 1) = 8
5. Square root of 8 = 2.828

Standard Deviation = 2.828


Note:If your data includes possible data points for a population, then in step 4 you would instead divide by the number of terms. But if you're working with a sample of the data, which is probably the case in XPD, you divide by number of terms minus 1.

Ex. If you're analyzing the height of ALL students in your grade level, then you divide by number of data values. If you analyze the height of a randomly selected, representative sample of students in your grade level, you divide by number of data values minus 1.

Alternatively, since you can calculate Standard Deviation on your Scientific/Graphing calculator. Plug data in Stat, and use 1-var stats. Standard deviation is round-greek symbol with line out.



Good luck :D
Thanks! I will have to look through a couple times to make sure I've got it, but seems legit! :]
2014 (Wright State/Lisle/Grayslake/Regionals/State/Nationals)

Water Quality: 8/1/2/1/2/3
Experimental Design: 12/1/5/5/8/37
Road Scholar: 12/2/2/3/3/11
Wheeled Vehicle: 32/6/6/2/11/5

User avatar
Panda Weasley
Member
Member
Posts: 133
Joined: September 27th, 2014, 6:24 am
Division: C
State: -
Location: Ravenclaw Tower

Re: Experimental Design B/C

Postby Panda Weasley » November 5th, 2014, 4:26 pm

Does anyone know if Division B people get bonus points for doing Standard Deviation? Also, does anyone have any pointers on how to get maximum points (and or bonus points)?
DFTBA!
Events 2019: Forensics and Fossils
Proud member of Teh Ento Cult.
:ugeek:

User avatar
XJcwolfyX
Member
Member
Posts: 340
Joined: October 22nd, 2010, 7:57 am
Division: C
State: -

Re: Experimental Design B/C

Postby XJcwolfyX » November 5th, 2014, 5:53 pm

Hi Panda! No, you do not receive extra credit or extra points for having Standard Deviation for Division B.

If you would like tips and hints about the event, please join us on the IRC Chat (Chat button on the top bar) or PM me! :)
Medal Counter: 73

User avatar
illusionofconfusion
Member
Member
Posts: 45
Joined: September 18th, 2011, 1:20 pm
Division: C
State: AL

Re: Experimental Design B/C

Postby illusionofconfusion » November 7th, 2014, 12:25 pm

Does anyone have a good way to distinguish between what goes in the conclusion and what goes in the analysis? Most of the time it seems to me like they're the same. :?
~illusionofconfusion ;)

User avatar
Panda Weasley
Member
Member
Posts: 133
Joined: September 27th, 2014, 6:24 am
Division: C
State: -
Location: Ravenclaw Tower

Re: Experimental Design B/C

Postby Panda Weasley » November 7th, 2014, 1:32 pm

The Analysis is a summery of what happened during the test. You are basically summarizing your results and discussing the data points. If there are any unusual data points or trends you noticed this is where you would write about them. You DO NOT discuss your hypothesis here!
The Conclusion is whether your hypothesis was supported or unsupported based on your data, and explaining why.

They do seem pretty similar, when I first started I was very confused by it. I don't really have a good method of remembering which is which though. Something that does help me though is the Analysis is analyzing the data while the Conclusion is concluding the report. Does that help at all? Maybe someone else here will have a better method.
DFTBA!
Events 2019: Forensics and Fossils
Proud member of Teh Ento Cult.
:ugeek:

User avatar
XJcwolfyX
Member
Member
Posts: 340
Joined: October 22nd, 2010, 7:57 am
Division: C
State: -

Re: Experimental Design B/C

Postby XJcwolfyX » November 7th, 2014, 2:33 pm

Analysis:
Say all of your data. You have to mention every data point, including your averages. Then, discuss unusual data points. Make sure you're specific...such as "The time of 2.67 seconds in the 20 cm tall ramp height data was an unusual data point..." and then explain it: "...this could have been caused by...X, Y, Z." If there are no unusual data points, say that. Do not emit that from your analysis. Then, discuss the trends. "As the ramp height increased, the amount of time it took the ball to roll down it and across the floor 1 meter decreased." Then, support this with data. "This is shown because the average times for ramp heights 10 cm, 20 cm, and 30 cm were 2.90, 2.63, and 2.11, respectively. Clearly, these numbers are increasing."

Conclusion:
For your conclusion, you restate and evaluate your hypothesis. "Our hypothesis that said ________ (restate the whole thing here) was correct/incorrect (choose one). This is shown in the data because the average for the 10 cm, 20 cm, and 30 cm ramps were X, Y, Z respectively (yes, you just said this above), which shows a clear increase/decrease in time." Then, you should give a reason why this happened. "This is most likely because (scientific explanation here, such as "as you increase the ramp height, the ball begins with more potential energy and therefore has more kinetic energy when it comes to the bottom of the ramp")......

Hope this helped.
Medal Counter: 73

User avatar
illusionofconfusion
Member
Member
Posts: 45
Joined: September 18th, 2011, 1:20 pm
Division: C
State: AL

Re: Experimental Design B/C

Postby illusionofconfusion » November 8th, 2014, 12:06 pm

Okay, that really helps! I had noticed that I was writing almost the same thing for both, and I was confused, but now I'm not! :D
~illusionofconfusion ;)

User avatar
CulturallyScientific
Member
Member
Posts: 176
Joined: December 29th, 2011, 7:13 pm
Division: Grad
State: CA
Location: Berkeley, CA

Re: Experimental Design B/C

Postby CulturallyScientific » November 16th, 2014, 10:59 pm

Has anyone thought of including "analyze quantitative data with statistical tools, such as T-Test, Chi-Square test, etc" in the recommendations? I only relatively recently learned more about statistics and ExCel, so just wondering if this crossed anyone else's mind.
'16, she/her, environmental-scientist-to-be: green gen, invasives, disease, ex. design, widi.

"…everything flows in an eternal present." (James Joyce)


Return to “2015 Lab Events”

Who is online

Users browsing this forum: No registered users and 3 guests