Data headings after changing web site

Technical Support and Questions

Thumb img 0715

Sonya Lawrence

Jun 2015

Hi,

Just looked at our student data at http://photosynq.org/projects/photosynthetic-efficiency-project-us15 and we noticed that the downloaded data headings do not match the headings on-line.

  1. We would like to know what should be the headings of the data in rows 125 and below since it does not match the headings in row 1?

  2. I have also noticed that we get a NaN value - why is this?

  3. do we automatically convert all negative values to 0 in the spreadsheet or just the negative values in phiII and SPAD?

  4. we noticed that if can add user ID to the data when you download it (repeats what is there) --can you download it with user ID and not have the whole data with it?

  5. lastly, I am hoping that we will not have the same issue as number 1 with our other students projects -- if we do what should I do to get the right heading information?

Thanks. Sonya Lawrence

Thumb 012409 pbi 160 low low res headshot

Greg Austic

Jun 2015

Comments in-line

  1. We would like to know what should be the headings of the data in rows 125 and below since it does not match the headings in row 1? Yeah, that's a mistake on our part, but it's getting fixed very soon. They changed the protocols at that point, so they are out of alignment. In the future, if you are going to change protocols try not to change the order - just add the protocol to the end. If you must change the order, best to just start a new project.
  2. I have also noticed that we get a NaN value - why is this? NaN happens when you try to divide by zero in an equation and stuff like that. If you get a NaN value you need to investigate and see what's going on. You can't assume it's 0 or something else - you need to investigate.
  3. do we automatically convert all negative values to 0 in the spreadsheet or just the negative values in phiII and SPAD? well... it's not that easy. Again - you have to go through your bad data (like negative phi2 or phi2 > .85) and figure out what went wrong. I can see from looking at the data that there's quite a big of bad data points being submitted. You can only tell bad data by looking at the traces. A good way to get a sense of 'normal' traces is by looking at normal values, and getting a sense of what a normal trace looks like. Then, go check out your wacky values and figure out... is this just a zero phi2 (biologically inactive) or did the user miss the leaf (which should be flagged and excluded). I'd be happy to show you what I mean so you can show the class.
  4. we noticed that if can add user ID to the data when you download it (repeats what is there) --can you download it with user ID and not have the whole data with it? *you can create a series for each user, then download. That way the series name will be the user ID - that's the best way for now.
  5. lastly, I am hoping that we will not have the same issue as number 1 with our other students projects -- if we do what should I do to get the right heading information? *like I said, be careful about changing protocols