Spreadsheet For Test Summation

[Alex-RCHS]

A question for event supervisors: Is it possible to input test answers into a spreadsheet in order to expedite scoring?

Basically, it would be a spreadsheet where each team has a column. Each row is a question, and in each box is the score that the team got on that question. For many questions, like MC or T/F, their "score" will be either 1 (if they got it right) or 0 (if they got it wrong). For free response, it will be however many points they got out of the maximum for that question. And then, of course, a sum function would add up each team's total score, and another sheet would automatically rank the teams in order based on score and would break ties however you choose, etc. It would also allow the event supervisor to see which questions had the highest or lowest scores, which can aid in making future tests.

Is this feasible? Do you think it would be too slow? Certainly it would aid in accuracy and speed of summation, but I'm worried that entering scores into the spreadsheet would negate the time saved on summation.

What do you all think?

[Things2do]

If you're fast with a numpad, it would probably save time. Especially if you're typing in a calculator anyways. You're increasing the accuracy of the sum, but you're introducing the potential for erroneously entering a score...

[BrownieInMotion]

I think beyond just time and accuracy, if supervisors used Google Sheets it would be possible to see what questions everyone got right and wrong and use that to help spot poorly worded or bad questions. It would also be possible to share the sheet with teams immediately afterwards so everyone can see what questions other teams got correct. Of course it would require some setup and tech savvy, but I definitely think using spreadsheets to score tests has a lot of potential benefits. Though, full disclosure, I've never run a tournament before, so I'd love to hear what everyone else thinks.

[Unome]

Personally I wouldn't do this in most cases, since I find mental summation more accurate.

I think beyond just time and accuracy, if supervisors used Google Sheets it would be possible to see what questions everyone got right and wrong and use that to help spot poorly worded or bad questions. It would also be possible to share the sheet with teams immediately afterwards so everyone can see what questions other teams got correct. Of course it would require some setup and tech savvy, but I definitely think using spreadsheets to score tests has a lot of potential benefits. Though, full disclosure, I've never run a tournament before, so I'd love to hear what everyone else thinks.

I've thought about this, but as for question analysis, usually you can see that pretty easily while grading. Sharing with teams is a possibility, but I doubt enough people consider that a significant benefit to be worth the effort.

(don't use Google Sheets, it's terrible vs. Excel for anything that doesn't require immediate collaboration)

[drcubbin]

Personally I wouldn't do this in most cases, since I find mental summation more accurate.

I think beyond just time and accuracy, if supervisors used Google Sheets it would be possible to see what questions everyone got right and wrong and use that to help spot poorly worded or bad questions. It would also be possible to share the sheet with teams immediately afterwards so everyone can see what questions other teams got correct. Of course it would require some setup and tech savvy, but I definitely think using spreadsheets to score tests has a lot of potential benefits. Though, full disclosure, I've never run a tournament before, so I'd love to hear what everyone else thinks.

I've thought about this, but as for question analysis, usually you can see that pretty easily while grading. Sharing with teams is a possibility, but I doubt enough people consider that a significant benefit to be worth the effort.

(don't use Google Sheets, it's terrible vs. Excel for anything that doesn't require immediate collaboration)

I could be wrong , but I am guessing the #1 reason this is being proposed is to reduce the wait time after competition while waiting for the scores. If so, I remember one event we went to a few years back (I believe it was Blue Dragon on Long Island) they showed a movie to let the students relax after a very competitive day. Even the teachers enjoyed it. There are many scientific documentaries that could be put up there to take the scores off their minds for a short while. Regarding the speed of grading, I believe if ESs knew that the hosting site had a ScanTron or similar (as Bayard Rustin told us they had), then the test could easily be divided into "scan-able" responses and the rest done by hand. This would make a difference in grading time. Regarding question analysis, I agree with Unome that this is very easy and accurately done visually while grading. I would hate to see SciOly become like "regular school" and begin require us to perform "item analysis" of all tests. Personally, I find nothing more useless or annoying. Ok, there are other more useless and annoying things!

[pb5754]

Personally I wouldn't do this in most cases, since I find mental summation more accurate.

I think beyond just time and accuracy, if supervisors used Google Sheets it would be possible to see what questions everyone got right and wrong and use that to help spot poorly worded or bad questions. It would also be possible to share the sheet with teams immediately afterwards so everyone can see what questions other teams got correct. Of course it would require some setup and tech savvy, but I definitely think using spreadsheets to score tests has a lot of potential benefits. Though, full disclosure, I've never run a tournament before, so I'd love to hear what everyone else thinks.

I've thought about this, but as for question analysis, usually you can see that pretty easily while grading. Sharing with teams is a possibility, but I doubt enough people consider that a significant benefit to be worth the effort.

(don't use Google Sheets, it's terrible vs. Excel for anything that doesn't require immediate collaboration)

I could be wrong , but I am guessing the #1 reason this is being proposed is to reduce the wait time after competition while waiting for the scores. If so, I remember one event we went to a few years back (I believe it was Blue Dragon on Long Island) they showed a movie to let the students relax after a very competitive day. Even the teachers enjoyed it. There are many scientific documentaries that could be put up there to take the scores off their minds for a short while. Regarding the speed of grading, I believe if ESs knew that the hosting site had a ScanTron or similar (as Bayard Rustin told us they had), then the test could easily be divided into "scan-able" responses and the rest done by hand. This would make a difference in grading time. Regarding question analysis, I agree with Unome that this is very easy and accurately done visually while grading. I would hate to see SciOly become like "regular school" and begin require us to perform "item analysis" of all tests. Personally, I find nothing more useless or annoying. Ok, there are other more useless and annoying things!

My understanding was that the purpose behind this is to reduce the likelihood of making errors while adding up points.

[Alex-RCHS]

Thanks for the comments!

Personally I wouldn't do this in most cases, since I find mental summation more accurate.

I've thought about this, but as for question analysis, usually you can see that pretty easily while grading. Sharing with teams is a possibility, but I doubt enough people consider that a significant benefit to be worth the effort.

Regarding the speed of grading, I believe if ESs knew that the hosting site had a ScanTron or similar (as Bayard Rustin told us they had), then the test could easily be divided into "scan-able" responses and the rest done by hand. This would make a difference in grading time. Regarding question analysis, I agree with Unome that this is very easy and accurately done visually while grading. I would hate to see SciOly become like "regular school" and begin require us to perform "item analysis" of all tests. Personally, I find nothing more useless or annoying. Ok, there are other more useless and annoying things!

My understanding was that the purpose behind this is to reduce the likelihood of making errors while adding up points.

To clarify -- the purpose of the idea is to save time, increase accuracy, and aid myself in writing future tests. I won't share any data with teams, except maybe general observations (i.e., "only 32% of teams got Question X right").

Drcubbin, I'm not sure what you're worried about exactly but I'm just floating the idea of using this for myself to aid me in scoring and also to analyze the test questions. I agree that scantrons are a great idea, and I'll look into that in the future. This is mostly for situations in which they're unavailable or unnecessary.

Certainly, my main goal for this is too increase the speed and accuracy of grading. I know that summation errors and awards ceremony-delays are both very common, so I wanted to look into the possibility of reducing both of them.

I've found that for tests with large MC/TF sections, I won't be the one grading that. For me, that was the case for DG at Duke, where I never had a chance to even look over those scores, and will also be the case for DG at states. If I'm able to see which questions are too hard and which are too easy, I can better adjust future tests.

I appreciate the heads up about sheets. It may be necessary to allow simultaneous work (and I personally don't think it's *that* bad), but I also prefer excel.

I think I'll discuss this with the other people involved in DG to see if there's support. If it works out well and speeds up the process, I'll share a template for others to use.

[bearasauras]

It's so nice to read through what everyone has pointed out in this thread. We discussed almost all of the various aspects above when we started making scoring spreadsheets for local SoCal tournaments back in 2006 - the precursor to the current tech event spreadsheets.

The original tech event spreadsheets and other event scoresheets were created to check against summation errors and to check for ties. They were not meant to replace Event Sup doing the calculations, but merely as an independent summation verification. In 2007, Dr. Putz saw these scoresheets and since then, we've been developing the Tech Event Scoresheet every year for public use.

From my experience, the Event Sup calculation error rate is about the same as the error rate for score entry onto the spreadsheet. But to reach the same wrong sum both ways would require quite a bit of luck. The tradeoff for accuracy, however, is needing more time.

Google Sheet is probably the solution for accuracy without sacrificing time since multiple Event Sup can input at the same time. Google Sheet also has almost all of the functionalities required by the current tech event scoresheets so I don't see an issue there. The only thing on the current tech event scoresheets that Google Sheet can't do is running the macro to see if you're using an outdated spreadsheet, but there is no such need on Google Sheet since you can always push update onto it.

PM me if you want to make these spreadsheets for your event. It sounds like what you're looking for is pretty easy to do and shouldn't take too long.

[Unome]

It's so nice to read through what everyone has pointed out in this thread. We discussed almost all of the various aspects above when we started making scoring spreadsheets for local SoCal tournaments back in 2006 - the precursor to the current tech event spreadsheets.

The original tech event spreadsheets and other event scoresheets were created to check against summation errors and to check for ties. They were not meant to replace Event Sup doing the calculations, but merely as an independent summation verification. In 2007, Dr. Putz saw these scoresheets and since then, we've been developing the Tech Event Scoresheet every year for public use.

From my experience, the Event Sup calculation error rate is about the same as the error rate for score entry onto the spreadsheet. But to reach the same wrong sum both ways would require quite a bit of luck. The tradeoff for accuracy, however, is needing more time.

Google Sheet is probably the solution for accuracy without sacrificing time since multiple Event Sup can input at the same time. Google Sheet also has almost all of the functionalities required by the current tech event scoresheets so I don't see an issue there. The only thing on the current tech event scoresheets that Google Sheet can't do is running the macro to see if you're using an outdated spreadsheet, but there is no such need on Google Sheet since you can always push update onto it.

PM me if you want to make these spreadsheets for your event. It sounds like what you're looking for is pretty easy to do and shouldn't take too long.

Oh I hadn't thought about multiple people being able to enter scores... it's not something I'm used to I guess.

[venules]

I actually attempted this for an invitational and it was more time-consuming than I anticipated. It may not be cost effective for exams that have a lot of questions, but it also could just be a matter of how you divide up the scoring/questions on the spreadsheet. I think Google Sheets is really useful though, especially for collaboration like others have mentioned, so I think you could definitely achieve your goal with some trial and error.