Anyone here using a huge amount of important data in one sheet

1

I work for a charity for children with autism and we are doing research where if I were using excel i would need more than 1,000 columns.
I need multiple staff to feed in multiple data, potentially using cross-docs.
I need to export in a way that the data can be plugged into SPSS software (at the moment using one giant tab on excel)
I need it to be safe.
Apparently there is such a thing as a data architect but we have no budget .
If anyone here is doing something similar, i’d love to reach out to you!
Thanks
Kirsty

2

Hi @Kirsty_Hayhoe,
and welcome to Coda Community! :handshake:

I honestly did not implemented something similar and I have the feeling that this is a borderline situation for Coda acting as a “data collection interface”.
If I correctly understood, currently the logic resided in the SPSS platform connected to a single sheet collector.

It really depends on your actual use cases to understand what approach would be better.
Contact me in PM and maybe we can discuss about it.

Cheers,
federico

3

Hi Federico,
Thanks for replying and apologies that i don’t know how to PM!
There are approximately 8 different assessments per child per year (and each assessment has on average 9 different sub sections that need data inputted) there are now 300 children and we haven’t collated data for 4 years (though we have the assessments separately)!.
The output at the end (i’ve been told) needs to be on 1 excel tab only.
I have about 12 staff who could input some results as they go, and 1 volunteer who can input old data and data that is currently on survey monkey.
I’d love to know your thoughts . . .
Kirsty

4

At a high level, Coda seems to me like it would be a good fit for this in terms of data structure, because you can have multiple tables that relate to each other. For example:

  • Children table (shows info about children, and a list of their assessments)
  • Assessments table (each assessment having a date and a child, cumulative score if relevant, and connected to subsections)
  • One or more tables of subsections

The trouble is, what makes Coda a great fit for this is exactly what makes one giant excel tab a bad fit - it may be hard to model the data in the end as just a single giant table in one Excel tab.

Do you have a version of the giant Excel tab already created, or know what kinds of columns SPSS is looking for?

(The above is separate from the “amount of data” question which, as Federico points out, may be a bit borderline for Coda to handle. Another solution that comes to mind is Google Sheets + Google Forms)

5

Thanks Nick,
Here are the first two lines of what was exported from excel to SPSS


This was only a few years worth so it may be bigger now.
I don’t use SPSS myself so not sure if it has to be this big but i think so!
Thanks for looking
Kirsty
6

Hello @Kirsty_Hayhoe :slight_smile:
The doc you shared on Sheets is not visible, if you can show it i can give you also my opinion :slight_smile:

7

Hi Mario, sorry it was too big to paste. I’ll see if I can make the link public. If I can’t please request access thanks for looking kirsty

8

Hi @Kirsty_Hayhoe, i have been able to see it, thanks :slight_smile:

From what i could understand your column are “groupable” in categories, this make me think about dividing your db scheme in the way @Nick_HE suggested,

First of all a little bit of introduction (note: those are my personal ideas, others are more creative and find their own often, so this is just one way :slight_smile:) working with columns in coda is better if done with not so many column, or at least the min necessary that fall into the same category, if yours are groupable you can create a table for each of them, then all rows will have a column that link them to the same “real life person” (that is another table, basically just a list of name, surname or whatever), this is thinking as a database and not as a xls file :slight_smile:

I’m pretty sure that coda can be a game changer tool to manage your datas, the only problem could be performance but to reach an unusable point you will need more than 10000 rows (again, personal interpretation of the things) :smile:

Try to “databasize” your table and see how many different relationship you can find into your data! :grinning:

P.s. you said that in excel you would have needed 1000+ columns, but i’ve seen just about 70, could you explain me this aspect? :slight_smile:

1 Like
9

Thanks All, I’ll have a play around with it, it would so nice to be able to see discrete segments of data!