Is Kobo the right tool for my needs? Help

User Support
1

Hi Folks,

I write snorkeling guides. When I visit destinations I have a 7"
Android tablet. Instead of typing out the name of each tropical fish
I see at each location I want to create a database of fish that I can
look through and quickly select if I have seen them, and then have
this list of fish names available for each location later.

So I created a list of questions in Kobo. Using the Multiselect, I
asked the question Angelfish, with the different types of Angelfish
listed below. I would have as many questions as there are general
types of fish, with all the options for those below.

This all worked great, but when I convert the file into CSV, the
results are just weird names and numbers. I can't see any of the fish
names that I put in as the questions or options. I don't understand
the value of the data if I have to lookup what each question was about
and how the numbers relate to the answers.

What I want is to output the list of fish for each location. Each
location will be a new survey.

So is Kobo the right tool for this? Can I get it to tell me the
actual fish names instead of the number?

The goal is to have it be formatted so that I can just copy the list
and stick it in the guide for that location.

Thanks for your help.

2

Snorkelnut,

A very interesting use of the KoBo Toolbox, great idea, I just hope you are keeping your tablet out of the water. I am not sure that it will work undersea.

It sounds like you set everything up correctly and are getting an aggregated CSV of your collected data.

In the CSV, you are seeing “weird names and numbers” which are not the labels that you put in there for the names of fish. I imagine what you are seeing are the Node names that are auto-assigned by KoBoForm, names like “A01, A02, A03…” and for the values you see “1,2,3…, etc.” Is that correct?

It’s not an error, KoBoForm gives each question an Alias for it’s data node to keep the size of the final data small, and so that you don’t name a variable in your database with the entire text of a lengthy question. The alias starts with A01 for the first question in your survey.

KoBoForm does the same thing with the values for each option under a question, so you might label an option “Angelfish”, but that label gets a value of “1”, and all the options are given similiar numerical values. Again, this is because the label can be very long and contain strange punctuation that could confuse the database.

These values are editable. You can select “Show Advanced Properties” in the KoBoForm interface, when a question is selected. If you edit the field “XML Node”, that will be reflected in the database.

As for the options, the numerical values are required. You cannot replace them with text using KoBoForm. This is because your final CSV output would be a mess with unlimited size text values. Instead, once you have your CSV data output, you create a Key that tells you what each value means.

Using this method, if you ask the question:

What kind of fish did you see?

and changed the XML Node to read “fish”, when you output the CSV and look at it, you would see a column headed “fish”, showing you the data collected from that question.

If you had Angelfish, Dogfish, and Catfish as options in that question, they would have values 1,2,3 and you would see that under the “fish” column in the CSV. Each record would have some combination of those numbers, depending on what fish were viewed.

Your point about the labels is one we have considered from a usability perspective. It’s one of those “feature vs. flaw” type questions. Allowing the values to be edited can be problematic. The solution that is in the works is a labeling solution, which would provide a separate key file containing all the relevant labels for your aliases. It would tell you that "Node A01 is for question “What kind of fish did you see?” and it would tell you “Value 1 is for Angelfish, Value 2 is for Dogfish, Value 3 is for Catfish.” Of course, this would be in a format that you could import into an analysis program like SPSS.

I hope that answers your questions, and that KoBo can help you with your fish tracking.

~Neil Hendrick
KoBo Developer

···

On Sat, Mar 31, 2012 at 2:56 PM, SnorkelNut galen...@mailhaven.com wrote:

Hi Folks,

I write snorkeling guides. When I visit destinations I have a 7"

Android tablet. Instead of typing out the name of each tropical fish

I see at each location I want to create a database of fish that I can

look through and quickly select if I have seen them, and then have

this list of fish names available for each location later.

So I created a list of questions in Kobo. Using the Multiselect, I

asked the question Angelfish, with the different types of Angelfish

listed below. I would have as many questions as there are general

types of fish, with all the options for those below.

This all worked great, but when I convert the file into CSV, the

results are just weird names and numbers. I can’t see any of the fish

names that I put in as the questions or options. I don’t understand

the value of the data if I have to lookup what each question was about

and how the numbers relate to the answers.

What I want is to output the list of fish for each location. Each

location will be a new survey.

So is Kobo the right tool for this? Can I get it to tell me the

actual fish names instead of the number?

The goal is to have it be formatted so that I can just copy the list

and stick it in the guide for that location.

Thanks for your help.

–

You received this message because you are subscribed to the Google Groups “Kobo Users” group.

To post to this group, send email to kobo-...@googlegroups.com.

To unsubscribe from this group, send email to kobo-users+...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/kobo-users?hl=en.

3

Excellent. Thanks for the detailed reply. It answers all my
questions.

And no, I won't be attempting to use the tablet in the water. It is
used after we get out of the water, reviewing pictures and our
memories of what we saw. This will greatly increase the speed of
recording the data, compared to typing or hand writing each fish (and
corals too).

We have decided to use the ODK build tool because it does allow us to
have the fish names show up in our report, without having to use a
key. We are in the process of building the list now. We will see if
it works once completed.

Thanks again!

···

On Apr 2, 8:16 pm, Neil Hendrick <neil.h...@kobotoolbox.org> wrote:

Snorkelnut,

A very interesting use of the KoBo Toolbox, great idea, I just hope you are
keeping your tablet out of the water. I am not sure that it will work
undersea.

It sounds like you set everything up correctly and are getting an
aggregated CSV of your collected data.
In the CSV, you are seeing "weird names and numbers" which are not the
labels that you put in there for the names of fish. I imagine what you are
seeing are the Node names that are auto-assigned by KoBoForm, names like
"A01, A02, A03..." and for the values you see "1,2,3..., etc." Is that
correct?

It's not an error, KoBoForm gives each question an Alias for it's data node
to keep the size of the final data small, and so that you don't name a
variable in your database with the entire text of a lengthy question. The
alias starts with A01 for the first question in your survey.

KoBoForm does the same thing with the values for each option under a
question, so you might label an option "Angelfish", but that label gets a
value of "1", and all the options are given similiar numerical values.
Again, this is because the label can be very long and contain strange
punctuation that could confuse the database.

These values are editable. You can select "Show Advanced
Properties<http://www.kobotoolbox.org/koboform/#&gt;&quot;
in the KoBoForm interface, when a question is selected. If you edit the
field "XML Node", that will be reflected in the database.

As for the options, the numerical values are required. You cannot replace
them with text using KoBoForm. This is because your final CSV output would
be a mess with unlimited size text values. Instead, once you have your CSV
data output, you create a Key that tells you what each value means.

Using this method, if you ask the question:

What kind of fish did you see?

and changed the XML Node to read "fish", when you output the CSV and look
at it, you would see a column headed "fish", showing you the data collected
from that question.

If you had Angelfish, Dogfish, and Catfish as options in that question,
they would have values 1,2,3 and you would see that under the "fish" column
in the CSV. Each record would have some combination of those numbers,
depending on what fish were viewed.

Your point about the labels is one we have considered from a usability
perspective. It's one of those "feature vs. flaw" type questions. Allowing
the values to be edited can be problematic. The solution that is in the
works is a labeling solution, which would provide a separate key file
containing all the relevant labels for your aliases. It would tell you that
"Node A01 is for question "What kind of fish did you see?" and it would
tell you "Value 1 is for Angelfish, Value 2 is for Dogfish, Value 3 is for
Catfish." Of course, this would be in a format that you could import into
an analysis program like SPSS.

I hope that answers your questions, and that KoBo can help you with your
fish tracking.

~Neil Hendrick
KoBo Developer

On Sat, Mar 31, 2012 at 2:56 PM, SnorkelNut <galen...@mailhaven.com>wrote:

> Hi Folks,

> I write snorkeling guides. When I visit destinations I have a 7"
> Android tablet. Instead of typing out the name of each tropical fish
> I see at each location I want to create a database of fish that I can
> look through and quickly select if I have seen them, and then have
> this list of fish names available for each location later.

> So I created a list of questions in Kobo. Using the Multiselect, I
> asked the question Angelfish, with the different types of Angelfish
> listed below. I would have as many questions as there are general
> types of fish, with all the options for those below.

> This all worked great, but when I convert the file into CSV, the
> results are just weird names and numbers. I can't see any of the fish
> names that I put in as the questions or options. I don't understand
> the value of the data if I have to lookup what each question was about
> and how the numbers relate to the answers.

> What I want is to output the list of fish for each location. Each
> location will be a new survey.

> So is Kobo the right tool for this? Can I get it to tell me the
> actual fish names instead of the number?

> The goal is to have it be formatted so that I can just copy the list
> and stick it in the guide for that location.

> Thanks for your help.

> --
> You received this message because you are subscribed to the Google Groups
> "Kobo Users" group.
> To post to this group, send email to kobo-...@googlegroups.com.
> To unsubscribe from this group, send email to
> kobo-users+...@googlegroups.com.
> For more options, visit this group at
>http://groups.google.com/group/kobo-users?hl=en\.