publicdata/j40-cejst-2
1
0
Fork 0
mirror of https://github.com/DOI-DO/j40-cejst-2.git synced 2025-07-28 07:41:16 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

Download column order completed (#1077)

Browse source 
* Download column order completed

* Kameron changes

* Lucas and Beth column order changes

* cdc_places update

* passing score

* pandas error

* checkpoint

* score passing

* rounding complete - percentages still showing one decimal

* fixing tests

* fixing percentages

* updating comment

* int percentages! 🎉🎉

* forgot to pass back to df

* passing tests

Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>
This commit is contained in:
Jorge Escobar 2022-01-13 15:04:16 -05:00 • committed by GitHub
commit d686bb856e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
13 changed files with 232 additions and 133 deletions
Download patch file Download diff file Expand all files Collapse all files

131
data/data-pipeline/data_pipeline/score/field_names.py
View file

@ -6,7 +6,7 @@ TOP_25_PERCENTILE_SUFFIX = " (top 25th percentile)"
# Geographic field names
GEOID_TRACT_FIELD = "GEOID10_TRACT"
STATE_FIELD = "State Name"
STATE_FIELD = "State/Territory"
COUNTY_FIELD = "County Name"
# Score file field names
@ -45,7 +45,7 @@ MEDIAN_HOUSE_VALUE_PERCENTILE = 90
# Poverty / Income
POVERTY_FIELD = "Poverty (Less than 200% of federal poverty line)"
POVERTY_LESS_THAN_200_FPL_FIELD = (
"Percent of individuals < 200% Federal Poverty Line"
"Percent of individuals below 200% Federal Poverty Line"
)
POVERTY_LESS_THAN_150_FPL_FIELD = (
"Percent of individuals < 150% Federal Poverty Line"
@ -83,15 +83,15 @@ EXPECTED_POPULATION_LOSS_RATE_FIELD = (
)
# Environment
DIESEL_FIELD = "Diesel particulate matter"
PM25_FIELD = "Particulate matter (PM2.5)"
DIESEL_FIELD = "Diesel particulate matter exposure"
PM25_FIELD = "PM2.5 in the air"
OZONE_FIELD = "Ozone"
TRAFFIC_FIELD = "Traffic proximity and volume"
LEAD_PAINT_FIELD = "Percent pre-1960s housing (lead paint indicator)"
WASTEWATER_FIELD = "Wastewater discharge"
AGGREGATION_POLLUTION_FIELD = "Pollution Burden"
RMP_FIELD = "Proximity to Risk Management Plan (RMP) facilities"
TSDF_FIELD = "Proximity to TSDF sites"
TSDF_FIELD = "Proximity to hazardous waste sites"
NPL_FIELD = "Proximity to NPL sites"
AIR_TOXICS_CANCER_RISK_FIELD = "Air toxics cancer risk"
RESPIRATORY_HAZARD_FIELD = "Respiratory hazard index"
@ -106,22 +106,26 @@ HT_INDEX_FIELD = (
ENERGY_BURDEN_FIELD = "Energy burden"
# Health
DIABETES_FIELD = "Diagnosed diabetes among adults aged >=18 years"
ASTHMA_FIELD = "Current asthma among adults aged >=18 years"
HEART_DISEASE_FIELD = "Coronary heart disease among adults aged >=18 years"
CANCER_FIELD = "Cancer (excluding skin cancer) among adults aged >=18 years"
DIABETES_FIELD = (
"Diagnosed diabetes among adults aged greater than or equal to 18 years"
)
ASTHMA_FIELD = (
"Current asthma among adults aged greater than or equal to 18 years"
)
HEART_DISEASE_FIELD = (
"Coronary heart disease among adults aged greater than or equal to 18 years"
)
CANCER_FIELD = "Cancer (excluding skin cancer) among adults aged greater than or equal to 18 years"
HEALTH_INSURANCE_FIELD = (
"Current lack of health insurance among adults aged 18-64 years"
)
PHYS_HEALTH_NOT_GOOD_FIELD = (
"Physical health not good for >=14 days among adults aged >=18 years"
)
PHYS_HEALTH_NOT_GOOD_FIELD = "Physical health not good for greater than or equal to 14 days among adults aged greater than or equal to 18 years"
LIFE_EXPECTANCY_FIELD = "Life expectancy (years)"
LOW_LIFE_EXPECTANCY_FIELD = "Low life expectancy"
# Other Demographics
TOTAL_POP_FIELD = "Total population"
UNEMPLOYMENT_FIELD = "Unemployed civilians (percent)"
UNEMPLOYMENT_FIELD = "Unemployment (percent)"
LINGUISTIC_ISO_FIELD = "Linguistic isolation (percent)"
HOUSEHOLDS_LINGUISTIC_ISO_FIELD = (
"Percent of households in linguistic isolation"
@ -139,25 +143,25 @@ CENSUS_DECENNIAL_POVERTY_LESS_THAN_100_FPL_FIELD_2009 = (
"Percentage households below 100% of federal poverty line in 2009"
)
CENSUS_DECENNIAL_HIGH_SCHOOL_ED_FIELD_2009 = "Percent individuals age 25 or over with less than high school degree in 2009"
CENSUS_DECENNIAL_UNEMPLOYMENT_FIELD_2009 = (
"Unemployed civilians (percent) in 2009"
)
CENSUS_DECENNIAL_UNEMPLOYMENT_FIELD_2009 = "Unemployment (percent) in 2009"
CENSUS_DECENNIAL_TOTAL_POPULATION_FIELD_2009 = "Total population in 2009"
CENSUS_DECENNIAL_AREA_MEDIAN_INCOME_PERCENT_FIELD_2009 = (
"Median household income as a percent of territory median income in 2009"
)
LOW_CENSUS_DECENNIAL_AREA_MEDIAN_INCOME_PERCENT_FIELD_2009 = "Low median household income as a percent of territory median income in 2009"
# Fields from 2010 ACS (loaded for comparison with the territories)
CENSUS_UNEMPLOYMENT_FIELD_2010 = "Unemployed civilians (percent) in 2010"
CENSUS_UNEMPLOYMENT_FIELD_2010 = "Unemployment (percent) in 2010"
CENSUS_POVERTY_LESS_THAN_100_FPL_FIELD_2010 = (
"Percent of individuals < 100% Federal Poverty Line in 2010"
"Percent of individuals less than 100% Federal Poverty Line in 2010"
)
# Combined fields that merge island areas and states data
COMBINED_CENSUS_TOTAL_POPULATION_2010 = (
"Total population in 2009 (island areas) and 2019 (states and PR)"
)
COMBINED_UNEMPLOYMENT_2010 = "Unemployed civilians (percent) in 2009 (island areas) and 2010 (states and PR)"
COMBINED_UNEMPLOYMENT_2010 = (
"Unemployment (percent) in 2009 (island areas) and 2010 (states and PR)"
)
COMBINED_POVERTY_LESS_THAN_100_FPL_FIELD_2010 = (
"Percentage households below 100% of federal poverty line in 2009 (island areas) "
"and 2010 (states and PR)"
@ -259,66 +263,77 @@ TRANSPORTATION_COSTS = "Transportation Costs"
#####
# Names for individual factors being exceeded
# Climate Change
EXPECTED_POPULATION_LOSS_RATE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for expected population loss rate and is low income"
EXPECTED_AGRICULTURE_LOSS_RATE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for expected agriculture loss rate and is low income"
EXPECTED_BUILDING_LOSS_RATE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for expected building loss rate and is low income"
EXPECTED_POPULATION_LOSS_RATE_LOW_INCOME_FIELD = (
f"Greater than or equal to the {PERCENTILE}th percentile"
f" for expected population loss rate and is low income?"
)
EXPECTED_AGRICULTURE_LOSS_RATE_LOW_INCOME_FIELD = (
f"Greater than or equal to the {PERCENTILE}th percentile"
f" for expected agriculture loss rate and is low income?"
)
EXPECTED_BUILDING_LOSS_RATE_LOW_INCOME_FIELD = (
f"Greater than or equal to the {PERCENTILE}th percentile"
f" for expected building loss rate and is low income?"
)
# Clean energy and efficiency
PM25_EXPOSURE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for PM2.5 exposure and is low income"
ENERGY_BURDEN_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for energy burden and is low income"
PM25_EXPOSURE_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for PM2.5 exposure and is low income?"
ENERGY_BURDEN_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for energy burden and is low income?"
# Clean transportation
DIESEL_PARTICULATE_MATTER_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for diesel particulate matter and is low income"
TRAFFIC_PROXIMITY_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for traffic proximity and is low income"
DIESEL_PARTICULATE_MATTER_LOW_INCOME_FIELD = (
f"Greater than or equal to the {PERCENTILE}th percentile for "
"diesel particulate matter and is low income?"
)
TRAFFIC_PROXIMITY_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for traffic proximity and is low income?"
# Affordable and Sustainable Housing
LEAD_PAINT_MEDIAN_HOUSE_VALUE_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile for lead paint and"
f"Greater than or equal to the {PERCENTILE}th percentile for lead paint and"
f" the median house value is less than {MEDIAN_HOUSE_VALUE_PERCENTILE}th "
f"percentile and is low income"
f"percentile and is low income?"
)
HOUSING_BURDEN_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for housing burden and is low income"
HOUSING_BURDEN_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for housing burden and is low income?"
# Remediation and Reduction of Legacy Pollution
RMP_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for proximity to RMP sites and is low income"
SUPERFUND_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for proximity to superfund sites and is low income"
HAZARDOUS_WASTE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for proximity to hazardous waste facilities and is low income"
RMP_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for proximity to RMP sites and is low income?"
SUPERFUND_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for proximity to superfund sites and is low income?"
HAZARDOUS_WASTE_LOW_INCOME_FIELD = (
f"Greater than or equal to the {PERCENTILE}th percentile"
f" for proximity to hazardous waste facilities and is low income?"
)
# Critical Clean Water and Waste Infrastructure
WASTEWATER_DISCHARGE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for wastewater discharge and is low income"
WASTEWATER_DISCHARGE_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for wastewater discharge and is low income?"
# Health Burdens
DIABETES_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile for diabetes and is low income"
)
ASTHMA_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile for asthma and is low income"
)
HEART_DISEASE_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for heart disease and is low income"
DIABETES_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for diabetes and is low income?"
ASTHMA_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for asthma and is low income?"
HEART_DISEASE_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for heart disease and is low income?"
LOW_LIFE_EXPECTANCY_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile "
f"for low life expectancy and is low income"
f"Greater than or equal to the {PERCENTILE}th percentile "
f"for low life expectancy and is low income?"
)
# Workforce
UNEMPLOYMENT_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for unemployment"
f"Greater than or equal to the {PERCENTILE}th percentile for unemployment"
" and has low HS education"
)
LINGUISTIC_ISOLATION_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for households in linguistic isolation"
f"Greater than or equal to the {PERCENTILE}th percentile for households in linguistic isolation"
" and has low HS education"
)
POVERTY_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for households at or below 100% federal poverty level"
f"Greater than or equal to the {PERCENTILE}th percentile for households at or below 100% federal poverty level"
" and has low HS education"
)
LOW_MEDIAN_INCOME_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for low median household income as a "
f"Greater than or equal to the {PERCENTILE}th percentile for low median household income as a "
f"percent of area median income and has low HS education"
)
@ -327,17 +342,17 @@ LOW_HS_EDUCATION_FIELD = "Low high school education"
# Workforce for island areas
ISLAND_AREAS_SUFFIX = " in 2009 (island areas)"
ISLAND_AREAS_UNEMPLOYMENT_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for unemployment"
f"Greater than or equal to the {PERCENTILE}th percentile for unemployment"
f" and has low HS education{ISLAND_AREAS_SUFFIX}"
)
ISLAND_AREAS_POVERTY_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for households at or below 100% federal poverty level"
f"Greater than or equal to the {PERCENTILE}th percentile for households at or below 100% federal poverty level"
f" and has low HS education{ISLAND_AREAS_SUFFIX}"
)
ISLAND_AREAS_LOW_MEDIAN_INCOME_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for low median household income as a "
f"Greater than or equal to the {PERCENTILE}th percentile for low median household income as a "
f"percent of area median income"
f" and has low HS education{ISLAND_AREAS_SUFFIX}"
)
@ -348,27 +363,27 @@ ISLAND_AREAS_LOW_HS_EDUCATION_FIELD = (
# Not currently used in a factor
EXTREME_HEAT_MEDIAN_HOUSE_VALUE_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile for summer days above 90F and "
f"Greater than or equal to the {PERCENTILE}th percentile for summer days above 90F and "
f"the median house value is less than {MEDIAN_HOUSE_VALUE_PERCENTILE}th "
f"percentile and is low income"
f"percentile and is low income?"
)
IMPENETRABLE_SURFACES_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile for impenetrable surfaces and is low "
f"Greater than or equal to the {PERCENTILE}th percentile for impenetrable surfaces and is low "
f"income"
)
AIR_TOXICS_CANCER_RISK_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for air toxics cancer risk and is low income"
RESPIRATORY_HAZARD_LOW_INCOME_FIELD = f"At or above the {PERCENTILE}th percentile for respiratory hazard index and is low income"
AIR_TOXICS_CANCER_RISK_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for air toxics cancer risk and is low income?"
RESPIRATORY_HAZARD_LOW_INCOME_FIELD = f"Greater than or equal to the {PERCENTILE}th percentile for respiratory hazard index and is low income?"
HEALTHY_FOOD_LOW_INCOME_FIELD = (
f"At or above the {PERCENTILE}th percentile for low "
f"access to healthy food and is low income"
f"Greater than or equal to the {PERCENTILE}th percentile for low "
f"access to healthy food and is low income?"
)
LOW_READING_LOW_HS_EDUCATION_FIELD = (
f"At or above the {PERCENTILE}th percentile for low 3rd grade reading proficiency"
f"Greater than or equal to the {PERCENTILE}th percentile for low 3rd grade reading proficiency"
" and has low HS education"
)
THRESHOLD_COUNT = "Total threshold criteria exceeded"
FPL_200_SERIES = "Is low income"
FPL_200_SERIES = "Is low income?"
# End of names for individual factors being exceeded
####