Experimental Design B/C

brayden box
Member
Member
Posts: 90
Joined: September 21st, 2015, 7:43 am
Division: B
State: NY
Contact:

Re: Experimental Design B/C

Postby brayden box » November 9th, 2015, 6:46 am

Could you do 4 levels? And if you did, would you get more points for it?
BBOX

User avatar
samlan16
Member
Member
Posts: 523
Joined: December 30th, 2013, 2:54 pm
Division: Grad
State: TN
Contact:

Re: Experimental Design B/C

Postby samlan16 » November 9th, 2015, 7:12 am

Could you do 4 levels? And if you did, would you get more points for it?
You can, but they only care about the first three that you list under the IV. So no, you would not get more points.

In case you have similar questions in the future, check the rubric here before asking.
Remember, we are proud of every team that participated and you are all winners.

User avatar
Whiteheat073
Member
Member
Posts: 49
Joined: November 6th, 2015, 8:27 am
Division: B
State: IL
Location: Standing in Wredling Middle School's Hallway, Being The Target For Air Trajectory Practice
Contact:

Re: Experimental Design B/C

Postby Whiteheat073 » November 10th, 2015, 8:36 am

Is there going to be Experimental Design in regionals for Div. B?
CM
Yes I'm a geek who likes TF2. Don't judge me.

User avatar
samlan16
Member
Member
Posts: 523
Joined: December 30th, 2013, 2:54 pm
Division: Grad
State: TN
Contact:

Re: Experimental Design B/C

Postby samlan16 » November 10th, 2015, 12:00 pm

Is there going to be Experimental Design in regionals for Div. B?
I don't know the conventions for your regional tournament, but given that you are in Illinois, it should be run. Their state organization is well established, so you should expect all events to be run.
Remember, we are proud of every team that participated and you are all winners.

Willows
Member
Member
Posts: 5
Joined: May 22nd, 2012, 3:13 pm
Division: C

Re: Experimental Design B/C

Postby Willows » November 15th, 2015, 10:28 am

Have you guys ever used anything other than linear regression in an experiment? Do you think it'd be worth it to know quadratic, logarithmic, and/or sinusoidal regressions? And if a quadratic curve does fit the data better, should it still be sketched on the graph as a "line (or curve) of best fit"?

laneyoung
Member
Member
Posts: 18
Joined: June 20th, 2014, 8:27 am
Division: B
State: IL

Re: Experimental Design B/C

Postby laneyoung » November 15th, 2015, 10:33 am

Is there going to be Experimental Design in regionals for Div. B?
While this isn't an official answer, Experimental Design should be run at the regional you're attending (assuming that your school is Wredling).
Illinois Arbitrator
Illinois Elementary Science Olympiad Committee

User avatar
watermydoing14
Member
Member
Posts: 108
Joined: July 25th, 2013, 8:53 pm
Division: C
State: WA
Location: Interlake
Contact:

Re: Experimental Design B/C

Postby watermydoing14 » November 22nd, 2015, 8:49 pm

Have you guys ever used anything other than linear regression in an experiment? Do you think it'd be worth it to know quadratic, logarithmic, and/or sinusoidal regressions? And if a quadratic curve does fit the data better, should it still be sketched on the graph as a "line (or curve) of best fit"?
I think it's generally best to use a linear regression because then you can interpret the intercepts. Plus, I think it's generally good practice to linearize data when the relationship turns out not to be linear (although I don't know that the event supervisors would expect teams to do that in the time given). Best case scenario is that you design an experiment that you know will turn out to be linear. If it's not, I would still sketch the line/curve of best fit just to be safe. Although technically you don't /need/ it since it's in the statistics section, so as long as you have enough other appropriate statistics to get the 6 points, you should be fine. However, I'm not sure how relevant other statistics would be on a non-linear graph, so I think having a non-linear graph could be a bit of a risk.
2013~Designer Genes~Disease Detectives~Forensics~Remote Sensing
2014~Anatomy~Experimental Design~Mission Possible
2015~Anatomy~Cell Biology~Experimental Design~Mission Possible
2016~Air Trajectory~Anatomy~Cell Biology~Experimental Design~Protein Modeling~Robot Arm~Wright Stuff
Interlake High School
GO SAINTS!

User avatar
coprolite_dipstick
Member
Member
Posts: 91
Joined: March 12th, 2015, 4:19 pm
Division: B
State: CA

Re: Experimental Design B/C

Postby coprolite_dipstick » November 28th, 2015, 4:00 pm

Hi, what exactly are linear/quadratic/logarithmic/sinusoidal regressions? This is my first year doing the event and I've honestly never even heard of those before.

Also, I'm in Div. B, so would knowing those be necessary for my division or is it mostly just for C?

Thanks
2016: CVMC/CV Invite/Mesa Robles/Reg/State
ExpD: 1/1/1/1/9
Foss: 3/1/8/4/1
Green Gen: 2/1/4/1/7
Met: 2/2/3/8/4
the dipstick is an intricate device used to measure the amount of rain in a rain gauge. it can also be used as a derogatory term for your meteorology partners

User avatar
samlan16
Member
Member
Posts: 523
Joined: December 30th, 2013, 2:54 pm
Division: Grad
State: TN
Contact:

Re: Experimental Design B/C

Postby samlan16 » November 29th, 2015, 2:57 pm

Hi, what exactly are linear/quadratic/logarithmic/sinusoidal regressions? This is my first year doing the event and I've honestly never even heard of those before.

Also, I'm in Div. B, so would knowing those be necessary for my division or is it mostly just for C?

Thanks
In simple terms, a regression is a statistical model of a set of bivariable data (i.e. has an independent and dependent variable, each with some numerical value) such that the average distance from the line to each point is a minimum. This link talks about linear regressions, which should be used when the data appears to form a uniform line. Likewise, quadratic regressions are for data that form a parabolic shape; logarithmic regressions for data forming the shape of a logarithmic graph; and sinusoidal regressions for data forming the shape of a sine wave. The derivation involves a ton of linear algebra; if you want to look it up, feel free.

Regressions are not required for either division and do not earn you brownie points, but you definitely will impress the judges if you do a regression.
Remember, we are proud of every team that participated and you are all winners.

Bozongle
Member
Member
Posts: 76
Joined: January 20th, 2013, 12:41 pm
Division: B
State: IN

Re: Experimental Design B/C

Postby Bozongle » December 9th, 2015, 9:55 am

I've been in Experimental Design for a few years now and the same question still comes to mind: When getting to a very competitive level (i.e. the teams competing know the rubric inside and outside, word for word in their head), for example, top 10 at Nationals, how are teams distinguished in rank? When it gets to a point where each Ex Des has the same key information required, how does a proctor decide if between four 120 point Ex Des's, one should be the champion?

Considering Experimental Design is mostly memorization of the rubric with consistent practice, I'm wondering what kind of special flair a team should do to distinguish themselves from the rest of the high-scoring pack.


Return to “2016 Lab Events”

Who is online

Users browsing this forum: No registered users and 1 guest