SSA datasets do not return a value

[AnasGhadieh]

Hello--

I have been trying to use this API. I first tried this:

http://labs.data.gov/csv-to-api/?source=http://www.ssa.gov/open/data/enumeration-processing-time.csv&format=html (NO LUCK)

I then tried this: http://labs.data.gov/csv-to-api/?source=http://www.ssa.gov/open/data/Yrly-Spoken-Language-DI-Claims.csv&format=html (NO LUCK)

Has someone had similiar issue in the past like this?

Thanks

[waldoj]

The problem is the SSA CSV files—they're not formatted properly. Here are the first few lines of your second example:

Data for Spoken Language Preferences among Initial Claims – Disability Insurance,,,
,,,
Spoken Language,FY - 2011,FY - 2012,FY - 2013
Includes ENGLISH ,Receipt Count,Receipt Count,Receipt Count
ENGLISH ,3108928,3037170,2871211

For this to be machine-readable, the header row has to the first row, instead of the fourth row. Any software trying to turn this into structured data would be unable to, because it would believe that the first column of data is titled Data for Spoken Language Preferences among Initial Claims – Disability Insurance, and that the other three columns have no titles at all.

[AnasGhadieh]

@waldoj Thanks for your reply ! I am trying to apply your suggested changes, can you please give me more details on how to fix this?

[waldoj]

Well, if you have control over those SSA datasets, they should be reformatted to eliminate the first three rows. If you do not have controls over those datasets, though there's not anything to be done—they're simply formatted incorrectly, in a fashion that this tool is not designed to be able to hand.

That said, you've raised a real problem, and I think I've got an idea of how to fix it. I just opened issue #28, to that end, which proposes a solution to this. I'm glad you brought this up!

[AnasGhadieh]

I am currently doing some testing with a sample csv file. I agree with you that this is an issue not only to SSA datasets but other agency's datasets.

For the meantime, I will keep this issue open and follow up with #28 .

[gbinal]

@AnasGhadieh - what he's saying is that you need clean CSVs in order for the tool to work.

Note the difference between the original files (1st, 2nd) and the cleaned up versions (1st, 2nd)

Here they are using the cleaned up versions being processed by the tool and working:

• enumeration processing time
• spoken languages

[gbinal]

By modifying your CSV files to have the header row first and then consistent rows after that, the tool should be working fine. I'm going to close this issue now but def. keep asking questions and helping to brainstorm future functionality.

[AnasGhadieh]

@gbinal Thanks a lot for your help. I did clean up one version of a CSV file and used and the tool working fine.