Data for CRISP

[ooi30627]

Hi Nathanael,

I hope that all is well. Matthew and I are compiling a list of the data that we need to obtain for the analysis. From our understanding of the documents provided by Gina, she used the following variables in the initial correlation analysis:

Census Derived: -Persons below poverty -Civilian age 16+ unemployed -Per capita income -Persons 65+ -Persons 25+ -Minority estimate (all persons except white, non-hispanic) -Single parent households with children <18 -Mobile Homes estimate Households with no vehicle -persons in institutionalized home quarters -civilian, non institutionalized populations with disability

Other: -USDA food access by census tract -CDC SVI -SNAP use

Do you have these data or should we hunt them down? Where did we leave off with in terms of the years to use for the ACS (we discussed averaging across multiple years).

Thanks, John

[ooi30627]

Hi John and Matthew,

For the initial effort the file:

SETX_Tracts_Pounds_Plus__CDC_SVI.csv

Has the CDC SVI variables included. I think they are for each Census Tract in the SETX Food bank area.

I have attached the codebook for the variables and you can also explore this website:

https://svi.cdc.gov/data-and-tools-download.html

The CDC is a good resource. I can also generate a similar file with Block Group data for the region.

Does this answer your question?

For the ACS question. I would suggest that one option would be to average the same years as the ACS.

For example – it looks like the CDC SVI data is 2016. Which is actually surveys collected over the 5 year period of 2012, 2013, 2014, 2015, and 2016. Option 1 would be to compare the 2014 pounds data to the 2016 5-year ACS. Option 2 would be to average 2012-2016 to compare.

We could try both options and see if there is any difference.

[ooi30627]

Hi John and Matthew

The last email got me thinking that it would be great to shift our efforts to github.tamu.edu

I have not used this system before but I would really like to make full use of it.

Dan Goldberg uses the system and he has this course to help explain the work:

https://github.tamu.edu/TAMU-GEOG-215-GeospatialCornerstone/GEOG-215-GeospatialCornerstone/blob/master/homework/01.md

Any chance you guys could help me figure out how to use it?

I think the questions we had on the email below could be better documented and collected through github.

It would also help with version control, transparency and data management.

Thanks

Nathanael

[ooi30627]

Nathanael,

Sounds like good plan. I have used Github a bit in the past, but not much. I could help get you started with it, but I have not used it for collaborations in a long time, so we will have to (re)learn that part together.

Thanks for pointing out the CDC data. As for the ACS, should we go about downloading those datasets anew? Or are they readily available on the architecture server?

[ooi30627]

Hi John

The CDC data is ACS data. Does that make sense?

[ooi30627]

Sorry, I am not being totally clear here. We will go ahead and start with the CDC data, but you presented two options previously (see below). If I understand correctly, in order to do option 2 also, we will need the 2012, 2013, 2014, 2015, and 2016 ACS data. I was just curious if this had been obtained already and/or joined to the food lbs data yet. If not, we will go ahead and start compiling that data as well.

[ooi30627]

Ahh –

To clarify – The 2016 ACS covers 2012-2016. The data that could be averaged is the food pounds data to make it more comparable to the 2016 ACS.

The ACS data would be the same in option 1 and option 2.

I was just reviewing the data in the SETX_Tracts_Pounds_Plus_CDC_SVI.csv file.

It looks like there are food pounds totals by month.

Option 1 = SUM(All months in 2014: Sum_Jan_14 + Sum_Feb_14 + … + Sum_Dec_14)

Option 2 = (Sum all months for each year 2012-2016) then Average the 5 years together

Hopefully this makes more sense,

Nathanael

PS – It would be great if we could transition to GitHub. I think we could keep better track of all of these email decisions.

In fact – I would maybe prioritize getting Github setup over getting the actual analysis done. It could be a huge benefit to the project to figure out a better collaboration tool than emails. I have attached an article that has inspired me to use GitHub more effectively.

[ooi30627]

Thanks Nathanael, this makes much more sense. Sounds like a plan with regards to GitHub, will keep you updated.