Announcement

Collapse
No announcement yet.

Statistical Tool for the MT Geeks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Statistical Tool for the MT Geeks

    Originally posted by "Krafty"
    Order doesn't matter.
    The other teams scores are just used to create the average scores for the rounds, as well as to show who won the ballot.
    I haven't scoured the sheet, so someone correct me if I am wrong. Is there a way to have the sheet calculate differentials on each thing (e.g. dropped open -1)?

    It looks really pretty

    Comment


    • #17
      Statistical Tool for the MT Geeks

      Try the Tabs sheet and look at the Median and Half range averages.

      I believe that is what you are looking for.


      Although that page doesn't compare opening and closing individually, it combines the two scores.

      Comment


      • #18
        Statistical Tool for the MT Geeks

        It appears that you can only place one witness per side per tournament.

        In cases where a student plays two witnesses would you just place a split name (eg Skogan/ Walsh)?
        This is a test to figure out if I can cheat by using a sig.

        Comment


        • #19
          Statistical Tool for the MT Geeks

          Originally posted by "AlphaTim"
          It appears that you can only place one witness per side per tournament.

          In cases where a student plays two witnesses would you just place a split name (eg Skogan/ Walsh)?
          Correct. The sheet doesn't do calculations based upon the actual role (i.e., what witness was played) just whether the person was a witness or attorney. The box to enter the role they played is there only for reference at this point, though in future versions it shouldn't be too hard for someone (including myself) to add in the functions necessary to calculate preformance based on actual role.
          "Call on God, but row away from the rocks." - Hunter S. Thompson

          Comment


          • #20
            Statistical Tool for the MT Geeks

            Yeah, I looked at doing that as well.

            The reason I ask is that I put in data validation for all the forms and was running some test data through it and ran into that instance.

            FTP is blocked at work, but I'll see if I can post the validated version tonight.

            I'm also looking into making a little VBA macro to combine multiple tournaments, but we'll see how busy I am this afternoon.
            This is a test to figure out if I can cheat by using a sig.

            Comment


            • #21
              Statistical Tool for the MT Geeks

              http://www.geocities.com/katerberg9/...plate_v0.3.xls

              Here is a version with validation (try and enter a number besides 1-10 to see what it does).

              I added the tournament info for the future macro to export to a larger file.
              This is a test to figure out if I can cheat by using a sig.

              Comment


              • #22
                Statistical Tool for the MT Geeks

                Nicely done. Two problems:

                1) You forgot to add your name to the credits. =)

                2) The pulldown menu for role includes only witness roles, not attorney roles (i.e. Opener, Mid, Closer) and also doesn't allow for someone that might have multiple roles.

                Aside from that, excellently done. I have a suggestion for someone if they want to undertake it. As of right now the options for roles are limited to the witnesses for this year. My few macro skills are too degraded to undertake this without refreshing them, but if someone wants to add another master sheet to type in the names of the witnesses for the year and then have those entries appear on the pull down menu, that would be useful for making the sheet applicable in future years.

                I'm also going to recommend that this version (and the future version it spawns with macros) be kept along with a simpler, macro-less version since some people may prefer the simpler sheet and some people are inherently fearful of macros in excel.
                "Call on God, but row away from the rocks." - Hunter S. Thompson

                Comment


                • #23
                  Statistical Tool for the MT Geeks

                  Good Catch. That's what I get for doing it quickly without really looking at what the data is representing.

                  1- No I didn't. "feel free to add your name to it if you make substantial changes and want to have your name associated with it". I figure that the little I did is enough to give my school some credit, but I don't think I'll add myself unless I get the macro working.

                  2- The pulldown menus are now corrected. Kinda. I added Split Witness, Opener, Mid, Closer. (The list for validation is under N3:N16). That column just happens to be 2 pixels wide right now. It's very simple to update that for future years (and no macro associated). What I'd recommend people do is create a change split witness to whatever your particular case is (You can also delete the witnesses you don't ever call on P or D). Just expand column N and change those values.

                  I also deleted module1 (which was blank) that was causing the macro warning. So now it looks a bit cleaner and won't scare people who think I'm trying to steal their info (if you want me to add in that "feature for you", let me know ;-)

                  For whoever wants to know how to do validation... Go to Data>Validation.

                  If you select "List" you can either type in the values "W, A" or else refer to the place where the list is located "N3:N16". Doing that will let you have drop down lists.

                  If you just want to give an error message if people type in the wrong number, go to Data>Validation>Whole number. Set it to ignore blanks, min=1 max=10. Nothing but numbers 1-10 will work.


                  For future years... If you create a list of all the roles (and team members for that matter) in a workbook (let's call it "Roster&quot, you can use list validation (Data>Validation>List) and refer to that roster file (=[Roster.xls]Sheet1!$A$1:$A11) That way you can have drop down menus running off one sheet (it will have external links, but as long as you don't move the roster file you'll be fine). This is cool since you can use drop down menus for every tournament and ensure that you don't mistype someone’s name (which becomes important when you start using a person's name as a unique identifier for linking scores across tournaments).

                  And... Here's the cleaned up (and now less incorrect) version 0.4

                  Both of these should work, but I think they have bandwidth restrictions
                  http://www.geocities.com/katerberg9/...plate_v0.4.xls
                  http://us.f13.yahoofs.com/bc/43db237...8nX6DBtPZdqbkM



                  I'm also thinking of making pivot charts for all of the data, but again, we'll see when time comes up. Plus since I'm playing with fake data at this point it's less relevant to me. If you guys think it would be helpful for you I can make the charts fairly easily.

                  I want to do the VBA stuff mainly to refresh my memory of VBA, but I think it would also be helpful for long term tracking. (watch how someone's scores change throughout four years or even across a single year).

                  Given that we have five teams I would also be interested to look at how different witnesses score across the board. I have a feeling that certain (experts) will always score better no matter who is put in them and other witnesses are very hard to score well no matter how we do them. There are obvious biases since more experienced witnesses tend to take harder roles which also score better, but maybe next year at the beginning I'll take a look at the data from practice rounds and see if it makes a difference with our coaches scoring.

                  It could be even better to look at that aggregate data across many teams, but I really doubt that teams would feel comfortable disclosing all that data (and I don't blame you).


                  Anyway, let me know what would be most helpful for you (mockers who consider using this) and I'll see what I can do when I have time.[/url]
                  This is a test to figure out if I can cheat by using a sig.

                  Comment


                  • #24
                    Statistical Tool for the MT Geeks

                    Originally posted by "AlphaTim"
                    but I don't think I'll add myself unless I get the macro working.
                    Me, to a software developer today: "What the hell is a macro?"

                    Bravo, Tim. If you're going to make Excel your bitch, I can imagine no better purpose than simplifying statistical analysis of AMTA rounds.
                    "Luck errs on the side of those who are prepared."

                    Comment


                    • #25
                      Statistical Tool for the MT Geeks

                      The biggest challenge is to make it so that it can be done in Excel.

                      We use SPSS and some proprietary software to do most of the statistical tests. Great for me to run stuff, but it is not stand alone.

                      Thankfully JayZ and his posse took care of writing all those boring functions. Now if I have some time I'll do the boring export functions. They aren't really hard, just boring.

                      Maybe while watching tv tonight...
                      This is a test to figure out if I can cheat by using a sig.

                      Comment


                      • #26
                        Statistical Tool for the MT Geeks

                        Originally posted by "AlphaTim"
                        Now if I have some time I'll do the boring export functions.
                        What the hell is an export function?

                        Just kidding, I know what it is. Or do I?
                        "Luck errs on the side of those who are prepared."

                        Comment


                        • #27
                          Statistical Tool for the MT Geeks

                          *bump*

                          I'm hoping people had a chance to use this thing over regionals and nationals.

                          I'm also hoping people have come up with ideas for or implemented some improvements. =)

                          Come summer I want to get back to work on this thing, so feel free to start pondering changes and improvements. Also, if you haven't tried it yet, please, download the file and give it a shot over gold.
                          "Call on God, but row away from the rocks." - Hunter S. Thompson

                          Comment


                          • #28
                            Statistical Tool for the MT Geeks

                            I used it for our regionals scores. It was really helpful.

                            Comment


                            • #29
                              Re: Statistical Tool for the MT Geeks

                              I went ahead and updated this for the 2008 season. All I did was just change the list of witnesses to reflect the ones in this years case.


                              You can find it here:
                              http://www.olemissmocktrial.org/node/9

                              Comment


                              • #30
                                Re: Statistical Tool for the MT Geeks

                                Omigoodness, this is the most amazing thing ever!! Thank you so much! :gavel:

                                Comment

                                Working...
                                X