Experimental Design B/C

[Jim_R]

Discussion for Experimental Design B/C.

[ZekeBud]

Before this gets kicking again, remember to read the rules, look at a rubric and practice, practice, practice.

Odds are, the rules won't change too much, so a similar rubric to this one for Division C would probably be rather important to you: http://www.tufts.edu/as/wright_center/p ... _c_div.pdf

Important notes:
You must include everything. If you leave it out, you're not going to get credit and you won't get those shiny medals.

Be specific and clear. If you don't label things carefully and you don't adequately explain your statements, it's just as bad as not including anything.

Work as a team. This is a timed event, so you need to work in unison to make sure everything gets done.


Go and check the Wiki and the Archives for more info and examples. Those should keep you going for quite some time and bring you a great deal of experience.

[srsvball95]

Here's the link to the wiki. Its the same for Division B and C: http://scioly.org/wiki/Experimental_Design

[Iustitia]

What kind of people do you want for this event? There's like 6 applicants in our school

[eyeball138]

People who are perfectionists and pay attention to lots of detail are usually good for this event.

[haven chuck]

But speed is also important since there is a TON to cover in 50 minutes

[eyeball138]

True. You need perfectionists who can work quickly.

[eyeball138]

Just saw the rules, I didn't see any changes.

[Phenylethylamine]

What kind of people do you want for this event? There's like 6 applicants in our school

There are different sections of the event, which each require different skills.

For the problem statement/hypothesis/materials/procedure part, which is very systematic, yes, you'd want a perfectionist, or at least someone good with detail. One of my partners last year wrote good 25-30 step procedures, including every. single. detail. to the point that someone completely clueless could have followed them easily (yeah, probably overkill, but pretty impressive, and impressive is often worth it in an event where the judging can be so subjective). You need to go in with a general structure in your mind and just fill in the relevant details and actual event-related information.

For the collection of data, making of graphs, and recording of observations, a perfectionist can also be useful, but mainly you just need someone who's not too much of a spaz (when you're throwing that paper airplane, you don't want it to hit the team three tables over; when you're measuring something, you want the end of that ruler to stay at the end of the object). Some knowledge of statistics and graphing is very useful, if not downright necessary. Also, you don't need to be a perfectionist about things like the lines of your data tables being perfectly parallel (*cough* pjgsciorox *cough*); as long as it's legible and doesn't give a serious impression of messiness, you're all good.

For the analysis/conclusion/future experimentation/possible errors, you need someone who can write a lot of intelligent-sounding bull as quickly as possible. If you're running out of time, you may need someone to start writing these sections before the relevant data is actually collected, in which case whoever does it needs to have some idea of the conclusion you expect. Some knowledge of statistics is also helpful here, mainly for the analysis part. The bigger your scientific vocabulary, the better.

Keep in mind that these don't have to be three separate people; those are just the skills needed for each part.

[nejanimb]

Phenylethylamine, is that how you break the sections down? I'm curious, since that's not how we usually do it, and I'm always curious to hear other strategies.

The first section we do exactly the same, where one person does everything from the beginning of the lab up through the procedure. After that though, we have one person who does Data Tables, Stats, Graphs, and Analysis. We've always found that this is a full time job, and that just doing this takes the full time. Usually, this person will be the one who helps the experimenter if there is something that requires an extra hand. The last person does the experimentation, the Errors, the Conclusion, and PracticalApplications/Further Study. Observations are done by whoever feels like they have the time. I've always felt like the analysis section was linked closer to the graphs and stat section than the conclusion part, so we have a math-heavy analysis that the Stat person.

As far as the type of people who should do this event, they key is to have people who simultaneously are meticulous but are cool under pressure. Of course, this seems a bit paradoxical, and really there aren't many people who fit this bill. You also need people who are simultaneously good at following instructions and a plan, but who also are good at thinking on their feet to solve problems - again, these traits are not always found in the same person, so it definitely takes a particular kind of person to do experimental.

As far as breakdown of trying to build a team, besides fitting the general characteristics I just described, you'll want someone who can be very methodical and break down exactly what needs to happen (the intro person), someone who's good with numbers (for the stat section in the middle of the lab), and a calm and collected person who has a good science intuition to do the conclusion and experimentation.

Even though I make it sound very specific, I think just about anyone who takes the time to learn ED can do a pretty good job. Good luck!

[Phenylethylamine]

What I described is basically how we break it down, although it can be kind of fluid depending on who finishes what first (also, clean-up is usually done by the person doing procedure etc., since they're often done first; otherwise, it would be the experimenter. The conclusion person is never the first one done. If they are, they can always write more).

Our system is sort of based on the 'skills' for the various sections:

• Problem-hypothesis-materials-procedure take a very systematic, structured approach, and are substantially similar from one competition to another. However, the basic structure needs to be adapted quickly to the specific situation of each event. The main skills involved are meticulousness and a good memory.
• Analysis-errors-conclusion-recommendation are basically writing. That's what ties them together in our system; they're the sections that basically have to be free response, straight-up mini-essays. That's definitely a skill unto itself: writing a lot, fast, coherently, and making the end product sufficiently eloquent and complex to impress the judges xD. [So even though the analysis is based in the statistics, the way it has to be presented groups it with the conclusion etc. The person does need some understanding of the statistics, but I think there are more people who understand the implications of mean/median/mode/range than who can write with the necessary clarity, complexity, and speed, so that's usually not an issue.]
• Collection of data (and the attendant stats and graphs thereof) needs a steady hand and at least a modicum of organization, and obviously the ability to calculate stats and graph data. In a perfect world, this would be the most straightforward section, but since data in this event doesn't always show what you want it to... this person gets the job of making up what the data "should" be if necessary (yes, bad scientific practice, I know; however, given the time-constrained nature of the event, sometimes things don't go quite the way they would in a real lab setting- e.g., sometimes the conclusion is written before the data collection is actually finished, so if the remaining data doesn't fit the conclusion, it can be easier to change a few numbers than change an entire conclusion). That takes a real ability to both understand the data and think on your feet.

The observations- qualitative and quantitative- are usually left to the experimenter or the procedure etc. person, depending on who finishes first, who's cleaning up, and so on. Nominally they're part of the 'experiment' section in our system, but about half the time all or part of them ends up done by the procedure person.

[sk8lynne]

i wish experimental was an event at regionals for division c. Now i can't compete in it even though its my favorite event and i did really well in it last year. It's not that hard to set up so i don't understand why they chose not to include it in the regional events.

[andrewwski]

That's something you'd have to discuss with your regional coordinator.

[duckiegirl2]

I was wondering if anyone knew of a good way to prepare for experimental design. Would doing a made up science experiment ( following all the rules in the experimental design rules page) be a good way?

[AlphaTauri]

Well, the way our team practiced last year was someone tossed semi-random objects into a box and told the ED people to make an experiment out of it. Apparently, this tactic worked- we did pretty well at Regionals and States. (It's hilarious watching them practice, though- during one of their experiments with a pendulum, we (the rest of the team) weren't allowed within two feet of the thing.)