Yuma 4×4

Media and Communications

Census Scientific Advisory Committee Meeting Day 2

Census Scientific Advisory Committee Meeting Day 2


VARE VAI >>GOOD MORNING.
TOMMY WRIGHT. WELCOME TO THE SECOND DAY OF THE
FALL MEETING OF THE CENSUS BUREAU SCIENTIFIC COMMITTEE
MEETING. WE HAD A LIVELY DISCUSSION
YESTERDAY AND A KNOW MANY OF YOU HAVE ALREADY TOLD ME YOU’RE
LOOKING FORWARD TO TODAY’S MORNING.
WHO SAID THAT? SO THIS IS SUMMARIZING WHAT
HAPPENED YESTERDAY. WHO SAID THAT?
LET’S JUST GET RIGHT TO IT. SO JUST TO REMIND YOU THERE’S NO
RULES, NO LOSERS AND NO WINNERS. YOU JUST YELL OUT THE ANSWER AND
YOU HAVE IN ERSATZ FACTION HOW YOU SCORED.
ONE HINT, THIS YEAR OF COURSE I THINK IT’S BEEN EVERY YEAR AND
THESE ARE NOT EXACT QUOTES, BY THE WAY.
THE ORDER ACCORDING TO THE AGENDA SO THAT’S A HINT.
I BUT I MAY THROW IN SOME DATA FOR DISASTER RECOVERY.
ALL RIGHT. WE MUST STAY RELEVANT.
WE FOSTER INNOVATION, FOR EXAMPLE, THROUGH THE CENSUS
OPPORTUNITY PROJECT CANVASSING OPERATION BEGAN AUGUST 18.
WE MUST REMAIN A LEADER IN DATA AND STATISTICAL PRODUCTS.
YOU SAID THAT.>>I’M GUILTY.
>>I THINK 3.8 BILLION OF DECENNIAL REQUESTS THEN THE
PERSON WENT ON TO SAY BILLIONS AND BILLIONS AND BILLIONS.
REMINDED ME OF I THINK WE ALL HEARD THAT PHRASE BEFORE.
WHO SAID BILLIONS OVER AND OVER AGAIN?
YES. WHAT ABOUT CONSTRUCTION STATISTICS?
WE CAN’T MOVE LEFT OVER MONEY FROM 2020 CENSUS TO CONSTRUCTION PROGRAMS.
WHO’S IN CENTER, WHAT’S GOING ON?
I FELT SAYING THE FUSION CENTER, WHAT’S GOING ON?
MARVIN GAYEE FOR WHAT’S GOING ON BUT THE FUSION CENTER DEFINED
THAT? TWO PEOPLE COULD BE POSSIBLE ANSWERS.
ENRIQUE AND AL. ENRIQUE MENTIONED IT FIRST BUT
AL FOLLOWED UP. ANY ASSESSMENT SAVINGS FROM
IN-OFFICE ADDRESS CANVASSING? WHAT DO YOU DO WHEN YOU GET
INCOMPLETE RESPONSES? YES. THROW ROCKS AT ME WHEN I START TALKING LIKE THIS.
WE NOW SAY ADULTS, CHILDREN AND BABIES LIVING, STAYING AT THIS
ADDRESS OR WE PLAN TO SAY YES, 1300 PARTNERSHIP
SPECIALISTS ON BOARD AS OF YESTERDAY.
COULD BE AL BUT WILLETTE SAID THAT AND HERE’S A SEQUENCE OF
EVENTS AND WE’RE WRAPPING UP ALMOST, REAL-TIME 2020
ADMINISTRATIVE RECORD SIMULATION AND COMPARE RESULTS BETWEEN A
CENSUS AND THE SURVEY STYLE CENSUS TAKING.
IN REAL TIME AT THE SAME TIME.
JENNIFER SAID AND SOMEBODY SAID WILL THIS BE ADMINISTRATIVE
RECORDS FOR THE ENTIRE COUNTRY.
THEN THERE WAS A SHORT ANSWER, YES.
>>POPULATION EQUALS BIRTHS MINUS DEATH VERSUS POPULATION
VERSUS IMMIGRATION. WE FOUND CHILDREN THAT WERE BORN IN THE
UNITED STATES WERE LIVING IN MEXICO AT THE 2020 CENSUS.
WE LEARN — WE LEARNED THIS FROM THE 2010 CENSUS.
>>THANK YOU VERY MUCH.
>>ANYONE INTENDS TO DO PUBLIC COMMENT THIS MORNING AND HASN’T
DONE SO, PROVIDE YOUR NAME REGISTRATION DESK AND FOR
VIEWERS A CALL-IN NUMBER WILL SHOW UP FOR THE PUBLIC EXCELLENT
SESSION SCHEDULED AT 11:00 A.M. AS WE BEGIN TO REMEMBER MEMBERS
AND THOSE JOINING US IN THE ROOM IT IS NOT ALLOWED TO USE YOUR
SMARTPHONE TO TAKE PICTURES INSIDE CENSUS HEADQUARTERS NOR
CAM — CAMERAS OR RECORDING DEVICES
OR MEMBERS ALLOWED WITHIN AN ESCORT AND WHEN YOU WANT TO
SPEAK TURN YOUR TENT CARD UP AND ALWAYS STATE YOUR NAME FOR THE
RECORD IT’S NEEDED EACH TIME TO SPEAK FOR THE MOST ACCURATE
TRANSCRIPTS. BE ADVISED SIDE CONVERSATIONS
CAN ARE HEARD ON THE MICROPHONES.
EACH MEMBER SHOULD HAVE RECEIVED A TRAVEL REIMBURSEMENT PACKAGE TURN IT IN
BEFORE YOU LEAVE. AND THE BUSSES WILL LEAVE THE
CENSUS AT 2:15 AND MAKE A STOP AT 2:45 AT RONALD REAGAN AIRPORT
AND 3:15 TAT THE HOTEL. — AT THE HOTEL.
DUE TO FEDERAL GUIDELINES REFRESHMENTS ARE PROVIDED FOR
COMMITTEE MEMBERS ONLY. BEFORE I TURN IT OVER TO ALISON,
I WANT TO MAKE NOTE OF SOMETHING.
AND WHAT SCHOOLS ARE PARTICIPATING PUBLIC VERSUS PRIVATE.
>>THIS IS INFORMATION HOW WE FOUND THE PRIVATE POOL.
THE STATISTICS AND POOLS PROGRAM USED PROGRAMS FOR LISTS FOR
STATES SIDE SCHOOLS INCLUDING ALL 50 STATES AND THE DISTRICT
OF COLUMBIA FOR THE NATIONAL EDUCATION STATISTICS AND
MAINTAINS A LIST OF PUBLIC AND PRIVATE SCHOOLS.
THE SCHOOLS ALSO WORKS WITH OTHER GROUPS SUCH AS THE TRIBAL
AREAS AND WORKED WITH PUERTO RICO, DEPARTMENT OF EDUCATION,
SECRETARY OF EDUCATION PROVIDED THE LIST OF ALL PUBLIC AND
PRIVATE SCHOOLS IN PUERTO RICO USED FOR THE MAIL OUT AND FOR
THE ISLAND AREAS WE COORDINATE WITH THE DECENNIAL ADVISER FOR
THE IS LAND AND THEY — ISLAND AND THEY WORK WITH OFFICIALS
DIRECTLY.>>THANK YOU VERY MUCH.
ALLISON.>>GOOD MORNING, EVERYBODY. I WANTED TO LET YOU KNOW, MARIA
IN CHARGE OF THE COMMUNICATIONS PROGRAM WHO COULDN’T BE HERE
YESTERDAY IS HERE THIS MORNING. SHE’S IN THE BACK AND IS HERE
UNTIL 9:45. SO IF FOLKS HAD QUESTIONS THEY
WANTED TO ASK HER, WE HAVE A MOMENT FOR THAT AND WE MIGHT
ACTUALLY IS ONE FOR YOU, MARIA, IF I HAVE TIME THIS MORNING BEFORE WE START.
ALSO IT’S EXCITING THIS MORNING AND WE’LL BE LOOKING AT THE
STATUS UPDATE ON THE CENSUS DATA PRODUCTS AMONG OTHER THINGS BUT
THAT’S EXCITING BECAUSE WE GOT A LOT OF DETAIL AHEAD OF TIME SO
WE THINK IT WILL BE A REALLY ROBUST DISCUSSION JUST LIKE THE
DEMOGRAPHIC ANALYSIS WAS BECAUSE WE GOT A LOT OF DETAIL AHEAD OF
TIME. I HEARD FEEDBACK FROM COMMITTEE
MEMBERS IT WAS FRUSTRATING TO HAVE SHORT AMOUNTS OF TIME ON
THE AGENDA MOSTLY TAKEN UP BY PRESENTATION AND THEN WE RAN
LATE. OF COURSE, EVERYTHING WORKED OUT
BECAUSE WE GOT EVERYTHING DONE WE NEEDED TO BY 5:00 BUT WE
DEFINITELY PREFER HAVING A DISCUSSION AND GETTING A LOT OF
DETAIL AHEAD OF TIME WE CAN DIG INTO BECAUSE WE LIKE THE
DETAILS. SO I’M SURE WE’LL HAVE MORE TO
SAY ABOUT THAT IN OUR RECOMMENDATIONS TODAY.
LOOKING FORWARD TO ALL THE IMPORTANT CONVERSATIONS WE’LL
HAVE TODAY AND ALSO JUST WANT TO SUPER COMMEND THE COMMITTEE
MEMBERS WHO DID AN AWESOME JOB WRITING RECOMMENDATIONS
YESTERDAY. OH, MY GOODNESS.
I HARDLY HAD TO DO ANYTHING LAST NIGHT WHICH GAVE ME CONFIDENCE
ABOUT GOING AHEAD AND TAKING THE FERRIS WHEEL RIDE WHICH OF
COURSE WE PAID FOR WITH OUR OWN DOLLARS AND NOT TAXPAYERS
DOLLARS NOR — FOR THE RIDE. LOOKING FORWARD TO THE
CONVERSATIONS AND THERE WAS A QUESTION, MARIA, IF IT’S OKAY I
COMMUNICATE, FROM THE RECOMMENDATIONS AND I’LL SEE IF
I CAN GET THIS RIGHT. I KNOW THERE’S A KNOWN ANSWER TO
THIS SO RATHER THAN WRITING IN TO THE RECOMMENDATION IT WAS A
QUESTION HOW THE VERY LARGE COMMUNICATIONS PLAN REACHES OUT
TO AND TARGETS MESSAGES TO SO MANY DIVERSE GROUPS AND FOR
COMMITTEE MEMBERS I THOUGHT YOU CAN TALK ABOUT THE STRUCTURE IN
WHICH YOU GENERATE MESSAGING BEYOND JUST THE TRANSLATION OF
THE LOGO AND WHATNOT.>>SO FROM DAY ONE I THINK ONE
OF THE MOST IMPORTANT THINGS IS WE’RE WORKING ON
COMMUNICATIONS. WHICH IS NOT ONLY IN THE
COMMUNICATIONS DIRECTOR BUT WE HAVE PEOPLE WORKING ON MYTHOLOGY
AND THE DIRECTORY AND DECENNIAL ETCETERA.
WE HAVE A DIVERSE TASK WORKING ON THESE WITH A DIVERSE
CONTRACTOR. WE HAVE Y & R BUT THEY LOOK AT
POPULATIONS AFRICAN, HISPANIC, ASIAN, PUERTO RICAN AND OTHERS AND WHAT
I THINK MAKES THE CAMPAIGN UNIQUE WE BROUGHT EVERYONE TO
THE TABLE. THIS WAS NOT A CENSUS YNR
DECISION BUT THE TEAM DECISION AND ALL THE CONTRACTORS MAKING
DECISIONS ON THE RESEARCH. WE MADE FOCUS GROUPS AND SURVEYS
IN WHICH WE COVERED ALL THE GROUPS.
WE DID FOCUS GROUPS WHEN WE TEST THE PLATFORM AND THEN WE MADE
ONLINE RESEARCH AND FOCUS GROUPS AND REPRESENTATIVES INTERVIEWS
DOING OUR CREATIVE TESTING ABOUT FOUR OR FIVE MONTHS AGO.
WE DID ABOUT 142 FOCUS GROUPS. WE TESTED FROM EVERYWHERE FROM
SAN JUAN, PUERTO RICO TO HONOLULU, HAWAI’I TO EVERYTHING
IN BETWEEN. WE TESTED ALL KINDS OF THINGS.
WE TESTED FOR RURAL AUDIENCES AND NOT ONLY TRADITIONAL RURAL
AUDIENCE BUT LOOKED FOR HISPANICS, AMERICAN INDIANS AND
BLACKS LIVING IN RURAL AREAS AND DID VERY URBAN AREAS AND TRY TO
REFLECT THE DIVERSITY AND THE MESSAGES YOU’LL SEE IN THE
CAMPAIGN WILL BE A DIVERSE CAMPAIGN WITH MESSAGES IN
GENERAL THAT SAY WE HAVE TOP FIVE MESSAGES.
WE KNOW FROM THOSE OVERALL WHAT SPECIFICALLY HAVE WORKED BETTER
FOR EACH OF THE PARTICULAR AUDIENCES.
SO WE ARE DEFINITELY PAYING MORE ATTENTION TO SAY IF I HAVE TOP
FIVE MESSAGES THE TOP THREE THINGS WORK BETTER AND THAT’S
WHAT WE’RE USING IN HAWAI’I AND MAKING IT ALSO AVAILABLE FOR THE
ISLAND AREAS TO USE. THEY HAVE A DIFFERENT
COMMUNICATIONS CAMPAIGN BUT IF SOMETHING GETS BETTER IN PUERTO
RICO WE’RE USING THAT. I WAS TALKING TO SOMEONE EARLIER
THIS MORNING AND WE HAVE DIFFERENT FAMILIES ON THE
CREATIVE. THAT’S EXPECT ALSO YOU’LL SEE IN
THE NEXT COUPLE OF MONTHS. AND THOSE FAMILIES OF CREATIVE, I THINK SOME ARE MORE ABOUT
CHILDREN AND SOME ARE MORE ABOUT THE FLAVOR OF YOUR COMMUNITY AND
DEPENDING ON HOW IT WORKS WITH THE AUDIENCE WE HAVE GIVEN MORE
ATTENTION AND SPACE TO THOSE FAMILIES WITHIN THE AUDIENCE AND
PERHAPS A FAMILY NOT BEING VERY USED IN HISPANIC MAY BE USED IN
ASIAN AND THAT’S HOW REFLECTING COMMUNITY.
FOR PAID ADVERTISING THE LARGER AREA THE CAMPAIGN YOU CAN SEE
THE VARIETY. WE TRIED TO DO THAT WITH
PARTNERSHIPS. WE TRIED TO GIVE A LITTLE
DIFFERENCES TO SOCIAL MEDIA. DEFINITELY IT HAS TO BE MORE
STANDARD. SIE SAY — IS A STANDARD
PROGRAM BUT YOU’LL SEE PERSONALIZATION WITHIN THE
GROUP.>>TO REITERATE HOW I UNDERSTAND
IT AND YOU CONFIRM WHAT WE LEARNED AT THE SPECIAL MEETING
YNR HAS FIVE OR SIX SUBCONTRACTORS AND EACH
SPECIALIZES IN AFRICAN AMERICAN, HISPANIC, NATIVE-AMERICAN,
HAWAIIAN. THEY EACH GENERATED CAMPAIGNS
AND CAME UP WITH IDEAS AND MESSAGES AND AMONG THOSE
MESSAGES WERE CHOSEN.>>EXACTLY.
>>IT WASN’T THERE WAS A CAMPAIGN DEVELOPED.
>>EVERYTHING CAME UP IN FACT I WOULD LOVE TO SAY SOMETHING.
A LOT OF THE MESSAGES YOU’LL SEE ON THE DIVERSE AUDIENCE CAMPAIGN
ACTUALLY CAME FROM THE SMALLER CONTRACTORS TO HAVE GREAT IDEAS
AND PEOPLE START COPYING AND THE MAIN CONTRACTOR LIKE THIS BIG
AGENCY IN NEW YORK WAS REALLY ADOPTING THEM TO THE DIVERSE
CAMPAIGN BECAUSE THEY WERE REALLY GOOD.
NOTHING HAS BEEN TOP DOWN IT’S BEEN FROM GROUND UP AND BASED ON
THE RESEARCH. I WOULD SAY FOR CREATIVE, WE
HAVE A LOT OF COMMUNITY INTERVIEWS BECAUSE WE WANTED TO
DOUBLE CHECK A LOT OF MESSAGES WIN SEVEN PARTICULAR — WITH
SEVEN PARTICULAR AUDIENCES SO WE NEEDED TO DIG MORE AND LIKE
HAITIAN. YOU NEED TO DIG MORE ON THOSE.
>>THANK YOU. I APPRECIATE THAT.
>>NOW TO JASON DEVINE AND CYNTHIA DAVIS HOLLINGSWORTH WITH
AN UP DATE ON THE 2020 DATA PROGRAM AND THERE’S A
DISCUSSION, KATHRYN PETTIT.
>>GOOD MORNING. JASON DEVINE.
I WORK IN THE POPULATION DIVISION AND HERE WITH MY
COLLEAGUE, CYNTHIA DAVIS HOLLINGSWORTH TO UPDATE YOU ON
THE PROPOSED 2020 CENSUS DATA PRODUCTS PLAN.
DURING THE SPRING CSAC MEETING WE PROPOSED DATA PRODUCTS FOR
THE 2020 CENSUS AND CHALLENGES FACED WITH A NEW METHOD OF
AVOIDANCE AND THIS WILL BE A CONTINUATION OF THAT DISCUSSION
WITH THE FOCUS ON THE DESIGN OF THE DEMOGRAPHIC AND HOUSING
CHARACTERISTICS FILE OR THE DHC AS WE REFER TO IT. FIRST A LITTLE BACKGROUND.
THE CENSUS HAS A LONG HISTORY OF PROTECTING INFORMATION PROVIDED
BY RESPONDENTS. THROUGH THE DECADES MORE AND
MORE GRANULAR DATA HAS BEEN PUBLISHED AND ADVANCES IN DATA
SCIENCE AND MORE POWERFUL COMPUTERS AND BIG DATA WHICH
CONTAIN PERSONAL INFORMATION HAS INCREASE THE RISK OF IDENTIFYING
INDIVIDUALS FROM PUBLISHED STATISTICS.
TO MITIGATE THIS RISK THE CENSUS BUREAU HAS A NEW DISCLOSURE
METHOD. THE GOAL FOR THE 2020 CENSUS
PRODUCT IS TO MEET DATA USER NEEDS AND IMPLEMENTING THE NEW
DISCLOSURE AVOIDANCE METHOD HOWEVER THERE’S SOME CHALLENGES.
TODAY WE’LL DISCUSS THE 2020 CENSUS DATA PRODUCTS DESCRIBED
WE THE DISCLOSURE AVOIDANCE SYSTEM AND WE’RE DEVELOPING
TABULATION PROVIDE COUNTS OF THE POPULATION BY DETAILED RISK AND
HISPANIC ORIGIN GROUP AND THE POPULATION DETAILED HISPANIC
RACE AND ORIGIN GROUPS, TRIBE AND BY HOUSEHOLD FAMILY TYPES.
FOR THE PRODUCTS NOT SUPPORTED BY THE CURRENT DISCLOSURE
AVOIDANCE SYSTEM WE’RE CONDUCTING OUTREACH TO
UNDERSTAND THE TABLES IN TERMS OF DETAIL AND GEOGRAPHY.
INITIAL FEEDBACK WAS COLLECTED THROUGH A JULY 2018 FEDERAL
REGISTRY NOTICE. ITS EXTENSION AND ASSOCIATED
OUTREACH. ADDITIONAL OUTREACH HAS BEEN
CONDUCTED TO GAIN INPUT FROM DATA USERS AND A 2010
DEMONSTRATION OF DATA PRODUCT FILED BASED ON THE DHC PLANNED
IS BEING DEVELOPED TO ALLOW EXTERNAL DATA USERS TO LOOK AT
IMPACT OF PRIVACY AND ADDITIONAL OUTREACH IS PLANNED INCLUDING A
NATIONAL ACADEMY OF SCIENCES WORKSHOP AND CYNTHIA WILL
PROVIDE MORE INFORMATION ON THE ADDITIONAL OUTREACH AND OUTREACH
THAT’S TAKEN PLACE AND WHAT IS PLANNED IN HER PORTION OF THE PRESENTATION.
STAKEHOLDER FEEDBACK WITH A REVIEW OF TABLE ACCESS RATES AND
SUBJECT MATTER EXPERT KNOWLEDGE IS INFORMING THE ACKNOWLEDGEMENT
OF THE DATA PRODUCTS AND IT WILL INCLUDE AN INITIAL SET OF
PRODUCTS MADE OF LEGALLY MANDATED PRODUCTS AND SUPPORTED
BY THE DIFFERENTIALLY PRIVATE ANNOYEDANCE SYSTEM.
— AVOIDANCE SYSTEM. A HEARD SET OF PRODUCTS WILL BE
PRODUCED WITH TABLES TO BE PROTECTED USING OTHER SYSTEMS.
THE DESIGN OF THE LATER SET OF PRODUCTS WILL BE DESIGNED BY
FEEDBACK ALREADY COLLECTED AND ADDITIONAL FEEDBACK WE HOPE TO RECEIVE.
FOR THE 2020 PRODUCT THEY HAVE A STRUCTURE SIMILAR TO THE 2010
DATA PRODUCTS THEY’RE GROUPED IN THREE CATEGORIES.
AVOIDANCE SYSTEM WHICH WILL HAVE A TOP-DOWN ALGORITHM EITHER
SUPPORTS OR IS BEING EXPANDED TO SUPPORT.
THIS INCLUDES THE TWO LEGALLY MANDATED PRODUCTS, APPORTIONMENT
AND REDISTRICTING FILE AND DEMOGRAPHIC PROFILES AND HOUSING
CHARACTERISTICS FILE AND DISTRICTINGDISTRICTING FILE AND
THE DHC WILL INCLUDE MANY OF THE TABLES AVAILABLE FROM SUMMARIES
IN 2010. THE CONTENTS OF THE PRODUCT IS
THE FOCUS OF THE PRESENTATION ALONG WITH AN OVERVIEW OF HOW TO
USE THE SPECIFICATION DOCUMENT WE PROVIDED PRIOR TO THE
MEETING. THE NEXT CATEGORY INCLUDES
TABLES AND PRODUCT WHERE’S DIFFERENTIALLY PRIVATE SYSTEMS
WILL BE DEVELOPED TO ALLOW FOR RELEASE WITH DETAILED RACE AND
HISPANIC ORIGIN TABLES AND FAMILY ORIGIN FILES IN SUMMARY
FILE OF 2010 AND AVERAGE AGE AND HOUSEHOLD SIZE FOR DIFFERENT
TYPES OF HOUSEHOLDS. ALL TABLES IN SUMMARY 2 IN 2010 AND THE ALASKAN NATIVE SUMMARY
FILE AND THE PUM IS UNIQUE WHERE THE MYTH ODDOLOGY CAN RELEASE
THE MICRO FILE BUT IT WOULDN’T INCLUDE THE HOUSEHOLD MEMBERS OR
THE VARIABLES NOT INCLUDED IN THE REDISTRICTING FILE OR DHC.
THE FINAL CATEGORY INCLUDES TWO PRODUCTS NOT PLANNED TO PRODUCE
FOR 2020 SINCE MOST THE DATA INCLUDED IN THEM IS AVAILABLE
FROM OTHER FILES. THIS WOULD BE THE SUMMARY OF
HOUSING CHARACTERISTICS AND HOUSING UNIT COUNTS REPORT SERIES.
THE PRESENTATION WILL FOCUS WITH THE ALGORITHM
BEING EXPANDED TO SUPPORT. THIS INCLUDES THIS APPORTIONMENT
PRODUCT AND THE DATA WILL BE TABULATE BEFORE DISCLOSURE AND
THE DEMOGRAPHIC HOUSING FILE AND CONGRESSIONALCONGRESSIONAL
DIRECT AND SPECIFICALLY THE DHC AND HOW TO GET INFORMATION ABOUT
WHAT’S INCLUDED IN THE FILES. THE DEMOGRAPHIC PROFILES ARE
SOURCE FROM THE DHC AND CONGRESSIONAL DISTRICT
DEMOGRAPHIC AND REDISTRICTING FILE IS A RETABULATION OF THE
DHC FOR CONGRESSIONAL BOUNDARIES.
A LITTLE BIT ABOUT THE REDISTRICTING FILE IT WILL BE
THE SECOND PRODUCT RELEASED FROM THE 2020 CENSUS.
AFTER THE RELEASE OF THE PORTIONMENT PRODUCT AND IT WILL
INCLUDE DEMOGRAPHIC AND HOUSING CHARACTERISTICS FOR DETAILED
GEOGRAPHIC AREAS. THE DATA PROVIDED THE STATES
UNDER PUBLIC LAW INCLUDES THE DATA ON THE CENSUS BLOCK LEVEL
ON THE VOTING AGE AND 57 POSSIBLE MULTIRACE COMBINATIONS
BY HISPANIC ORIGIN. THE OCCUPANCY STATUS AND GROUP
QUARTERS POPULATION FOR THE SEVEN MAJOR GROUP QUARTERS TYPE.
IT WILL BE AVAILABLE AT THE CENSUS BLOCK LEVEL AND MUST BE
RELEASED WITHIN ONE YEAR OF CENSUS DAY.
ONE UPDATE FOR THE 2020 REDISTRICTING FILE WAS THE
ADDITION OF DATA OR THE 7th MAJOR QUARTER TYPES RECORD —
REPORTED IN A SEPARATE FILE AND THE DATA WILL ALLOW US TO BETTER
MEET STATE REDISTRICTING AND THE TEST WAS MADE AVAILABLE IN MARCH
AND WAS THE FIRST 2020 CENSUS PRODUCT USED IN DISCLOSURE
AVOIDANCE METHODOLOGY. THE PROTOTYPE SPECIFICATION IS
AVAILABLE ON THE WEB PAGE ON CENSUS.gov.
HERE’S A SCREEN SHOT OF WHAT THE COVER OF THE DOCUMENT LOOKS
LIKE. I WON’T GO INTO DETAIL ABOUT
THIS AND WE’LL HAVE MORE TIME TO FOCUS ON THE DHC. IT WILL INCLUDE THE DEMOGRAPHIC
AND HOUSING TABLE PREVIOUSLY INCLUDED IN SUMMARY FILE 1.
YOU MAY US LATER HEAR US REFER TO P WHERE IT STANDS FOR
POPULATION AND AGE HOUSING BUT IN THIS PRESENTATION I’LL REFER
TO IT AT DHC AND THEY’RE ON THE SLIDE AND I’LL SKIP THEM HERE
BECAUSE DATA ON THE DHC WILL BE AVAILABLE FOR ALL GEOGRAPHIES AT
THE LOWEST LEVEL RELEASED FOR WHICH ARE AT THE CENSUS BLOCK
LEVEL. NOTABLE EXCLUSIONS INCLUDE
TABLES THAT PROVIDE COUNTS LIKE DETAILED RACE AND HISPANIC
ORIGIN, TRIBES AND POPULATION BY HOUSEHOLD FAMILY TYPES DUE TO
IMPLEMENT OF THE PRIVACY AND METHODOLOGY AND PRODUCTS
INCLUDING USE TABLES. IT PROVIDES DEMOGRAPHIC AND
HOUSING CHARACTERISTICS FROM SELECTED DHC TABLES AND EXTREME
LINE EASY TO USE FORMAT. AS SOON AS THE RELEASING PRODUCT
AS POSSIBLE AND SUBJECTS WILL INCLUDE FIVE-YEAR AGE GROUPS,
SEX, RACE, HISPANIC OR LATINO ORIGIN, HOUSEHOLD TYPE,
RELATIONSHIP TO HOUSEHOLD OR GROUP QUARTERS POPULATION AND
HOUSING TENURE. IT ALLOWS US TO PROVIDE SELECTED
DHC TABLES FULLER THAN THE FULL DHC AND AVAILABLE TO PLACE IN
MINOR CIVIL DIVISION AND ABOVE.
THERE’S DATA AVAILABLE AT THE CENSUS BLOCK LEVEL.
AND IT’S PROPOSED TO BE AVAILABLE IN THIS CASES THESE
TABLES WILL BE REPEATED BY MAJOR RACE AND HISPANIC ORIGIN GROUPS
AND IT ALLOWS YOU TO SEE WHICH TABLES THESE ARE.
AT THE CENSUS BLOCK THIS WILL INCLUDE TOTAL POPULATION, URBAN
RURAL DESIGNATION, SEX, RACE, HISPANIC, LATINO ORIGIN AND SEX
BY FIVE YEAR AND SINGLE YEAR OF AGE UNDER 20, RELATIONSHIP TO
HOUSEHOLDS BY HOUSEHOLDER CHARACTERISTICS, HOUSEHOLD TYPE
BY RELATIONSHIP AND PRESENCE OF SPECIFIC AGES, HOUSING TENURE,
HOUSING SIZE AND AGE AND SEX AND MEDIAN AGE FOR VARIOUS GROUPS
AND POPULATION BY SINGLE YEAR OF AGE UP TO 100 BY SEX AND 10 YEAR
BY RACE OR HISPANIC ORIGIN AND 10 YEAR BY PRESENCE OF CHILDREN.
AT THE COUNTY LEVEL IT WILL INCLUDE HOUSEHOLD TYPE AND
PRESENCE OF CHILDREN AND HOUSEHOLD TYPE BY AGE OF OWNER
AND PEOPLE OF SPECIFIC AGES AND ONE ADDITIONAL FILE AVAILABLE AT
THE STATE LEVEL TO PROVIDE THE GROUP QUARTERS SECTION AGE AND
POPULATION BY QUARTERS TYPE.
TO ALLOW YOU TO EXPLORE WHAT IS PROPOSED IN THE DNC WE PROVIDE A
PRODUCTS SPECIFICATION AND IN THE NEXT FEW SLIDES I’LL SHOW
YOU HOW TO USE THE SPECIFICATION.
THE WAY THE FILE’S SET UP YOU CAN USE THE INFORMATION AND SEE
THE TABLES INCLUDED IN THE STATISTICS FILE.
IT PROVIDES A CROSSWALK FROM THE 2010 SUMMARY FILE AND NUMBERS
FROM THE 2010DHC TABLE NUMBERS AND ALLOWS TO YOU IDENTIFY
TABLES THAT ARE IN THE FILE AND PROVIDES THE
PROPOSED LOWEST LEVEL OF GEOGRAPHY FOR EACH TABLE.
AND BEFORE I GET INTO SCREEN SHOTS OF THE SPECIFICATION,
THERE’S A TABLE NUMBERING CONVECTION THAT WILL HELP YOU
UNDERSTAND. IF YOU CAN UNDERSTAND THIS YOU
CAN LOOK AT THE TABLE AND SEE AND KNOW WHAT LEVEL OF GEOGRAPHY
THE TABLE’S AVAILABLE FOR AND WHAT TYPE OF TABLE IT IS.
TO P TABLES ARE POPULATION LABELS AND IF IT’S A P IT’S
AVAILABLE AT THE BLOCK LEVEL. IF IT’S A PCT IT’S AVAILABLE AT
THE BLOCK LEVEL AND PCO AT THE COUNTY LEVEL AND PST AT THE
STATE LEVEL AND H IS AVAILABLE AT THE BLOCK LEVEL AND HCT TABLE
IS AVAILABLE AT THE CENSUS TRACT LEVEL.
REMEMBER, IF IT’S AVAILABLE AT ANY OF THESE LEVEL IT’S ALSO
AVAILABLE ABOVE THAT LEVEL.
THIS IS A SCREEN SHOT IN THE SPECIFICATION WE PROVIDED.
THIS IS THE CROSSWALK TAB AND THE FILE INCLUDES FIVE TABS, IF
YOU LOOK ACROSS THE BOTTOM YOU’LL SEE THE TABS.
THE LIST OF TABLES THE DEMOGRAPHIC PROFILE, THE DHC
READ-ME TAB AND THE CROSSWALK AND TABLES TAB.
AGAIN, THOSE ARE AT THE BOTTOM OF THE SCREEN AND IT LETS YOU
SEE TABLES PROVIDED IN SUMMARY FILE 1 IN 2010 AND THE
CORRESPONDING FILES IN THE DHC. THE SCREEN SHOT IS THE FIRST FEW
TABLES IF YOU COMPARE IT WITH 2010 AND 2020 THE COLUMNS ON THE
LEFT SIDE YOU CAN SEE IF IT IS OR NOT INCLUDED.
IF THE LOWEST LEVEL OF GEOGRAPHY HAS CHANGED.
IF YOU LOOK THROUGH THE COLUMN YOU SEE P7.
I’LL FOCUS ON P7. I’LL GIVE EVERYBODY A SECOND TO
IDENTIFY P7. IT’S A THIRD OF THE WAY DOWN ON
THE LEFT-HAND COLUMN. IT’S HAS PANIC ORIGIN BY RACE
AND AVAILABLE AT THE COUNTY LEVEL ON THE DHC.
THE 2010 FS1 IT WAS AVAILABLE AT THE BLOCK LEVEL.
YOU CAN TELL BY THE LABEL. P IS A BLOCK TABLE, PCO IS A COUNTY TABLE.
THEY’RE STILL AVAILABLE AT THE BLOCK LEVEL AND IN THE NEXT
TABLES THEY’VE BEEN REMOVED ALTOGETHER BUT THE DATA AND YOU STILL HAVE DATA
AT THE SAME LEVEL YOU’D HAVE AVAILABLE IN 2010.
AND THE FOURTH COLUMN SHOWS THE LOWEST LEVEL OF GEOGRAPHY BEING
PROPOSED FOR 2020.
THE TAB PROVIDES MORE DETAIL ABOUT THE TABLE AND HERE YOU CAN
SEE P1 THROUGH P4. AND EXACTLY WHAT WILL BE IN THE TABLES.
ANOTHER TAB PROVIDES THE SAME LEVEL OF DETAIL NOR — FOR THE DEMOGRAPHIC
PROFILE. THIS IS BASICALLY A TABLE
LAYOUT. YOU CAN SEE HOW IT WILL BE SET
UP WHAT RACE CATEGORIES WILL BE INCLUDED IN THE TABLES AND
FORMAT. BY START THE DHC CROSSWALK YOU
CAN SEE THE TABLES INCLUDED AND THEN USE THE CROSSWALK FROM THE
2010 LABEL TO THE 2020 LABEL TO GO TO THIS TAB AND SEE MORE
DETAIL ABOUT WHAT’S GOING TO BE INCLUDED.
IT WILL BE INCLUDED IN THE FILE OR TABLE.
WHEN IT’S REPEATED FOR MAJOR RACE OR HISPANIC ORIGIN GROUPS
THEY’RE IDENTIFIED WITH THE LETTERS A THROUGH I AND GROUPS
REPEATED IN 2010. THEY’RE LISTED ON THE RIGHT SIDE OF THE SLIDE.
IT INCLUDES AFRICAN AMERICAN ALONE AND RACE GROUPS AND SOME
OTHER RACES ALONE, TWO OR MORE RACES AND HISPANIC OR LATINO AND
WHITE ALONE, NOT HISPANIC OR LATINO.
AND THE SCREEN SHOT SHOWS THE TABLE REPEATED BY THE GROUPS.
P7, SEX BY AGE BY AGE GROUPS. YOU SEE THE LETTER DESIGNATION
AFTER THE P7. THIS IS A TABLE FOR PEOPLE WOULD
ARE ASIAN ALONE.
THIS SHOWS WHAT TABLES FROM THE DHC WILL BE INCLUDED IN THE
DEMOGRAPHIC PROFILES AND IN THE TAB SHOWS THE DETAILS INCLUDED
IN THE TABLES AS I SHOWED YOU BEFORE.
SO THAT’S HOW YOU CAN SEE WHAT IS PLANNED FOR THE DHC AND
DEMOGRAPHIC PROFILES AND REDISTRICTING PRODUCTS AVAILABLE
ONLINE SO WITH WHAT’S AVAILABLE ONLINE AND SPECIFICATION YOU CAN
SEE WHAT SPROE OPPOSED FOR INCLUSION — FOR WHAT IS
PROPOSED IN THE DHC FROM THE SUMMARY PROVIDED IN 2010.
OF COURSE, WE STILL PLAN TO HAVE ADDITIONAL PRODUCTS INCLUDING
CENSUS BRIEFS AND POPULATION HOUSING TABLES AND SPECIAL
REPORTS TO PROVIDE ADDITIONAL ANALYSIS AND DIGITALIZATION FOR
2010 DATA AND DATA FOR LARGER PLACES.
AND WE INCLUDE THIS SLIDE TO PROVIDE A SUMMARY OF WHAT WE
COVERED IN THE PREVIOUS SLIDE AND THE FIRST BLOCK INCLUDES A
FOCUS ON STAY SUPPORT THE CURRENT AVOIDANCE SYSTEM AND THE
NEXT BLOCK INCLUDES TABLE REQUIRE COMPLEX HOUSEHOLD JOINS
AND DETAILED HISPANIC GROUPED AND ALAN — ALASKAN NATIVE
TRIBES AND GROUPS AND MAINTAIN CONFIDENTIALITY AND THE NEXT
BLOCK ARE MAINLY SOURCED FROM OTHER TABLED — TABLES AND WE
DISCUSS THE FILE BECAUSE CH IS UNIQUE BECAUSE WE CAN SEE THE
PROTECTED FILE BUT WOULDN’T INCLUDE THE OTHER INFORMATION
NOT PROVIDED IN THE REDISTRICTING PRODUCT OR DHC.
THE LAST TWO ARE TABLES WE’RE PROPOSING NOT TO PRODUCE.
NOW TO CYNTHIA TO COMPLETE THE PRESENTATION.
>>THANK YOU, JASON. GOOD MORNING, EVERYONE.
NOW I’LL TAKE A LITTLE BIT OF TIME TO UPDATE YOU ON AN
OUTREACH ACTIVITIES WE’VE BEEN CONDUCTING SINCE OUR LAST
BRIEFING WITH YOU ON THIS POINT. WE’VE PARTICIPATING IN SEVERAL
MEETINGS AND CONFERENCES PROVIDING THE SAME UPDATE AS
WE’VE PROCEEDED WITH THE METHODOLOGY FOR OUR DATA
PRODUCTS. THESE MEETING INCLUDE PAA, A
PANEL ON PRIVACY PROTECTION AND MEETING WITH NCSL TO NAME A FEW.
OTHERS ARE LISTED ABOVE. IN ADDITION, WE’VE RELEASED
INFORMATION REGARDING THE IMPLEMENT STATUS OF THE 2020
DISCLOSURE AVOIDANCE SYSTEM. THIS INCLUDES THE RELEASE OF THE
CODE BASE USED FOR THE 2018 END TO END TEST.
JOHN WILL PROVIDE MORE INFORMATION ON THIS AT HIS PRESENTATION FOLLOWING OURS.
I’LL TAKE A LITTLE BIT TO HIGHLIGHT THE EXPANSION OF THE
2020 CENSUS MEMO SERIES TO NOW INCLUDE FORMAL DECISIONS ON THE
IMPLEMENTATION OF THE 2020 DATA FILES.
IT’S ACTUALLY THE LAST BULLET ON THIS SLIDE.
THE 2020 MEMO SERIES DOCUMENTS SIGNIFICANT DECISIONS, ACTIONS
AND ACCOMPLISHMENTS REGARDING THE 2020 CENSUS PROGRAM.
TO DATE, WE HAVE TWO DECISION MEMOS AVAILABLE PUBLICLY AND THE
TWO DECISION MEMOS ARE ONE SPECIFICALLY ON THE PARAMETERS
FOR THE 2018 END TO END TEST. AS JASON MENTIONED THE IMPLEMENT
OF THE DISCLOSURE AVOIDANCE SYSTEM ON THE DATA AFTER WE
RELEASE THE PORTION SO AT THE START OF THE REDISTRICTING DATA.
— THE RELEASE OF THE APPORTIONMENT.
WE’LL CONTINUE TO GATHER FEEDBACK THROUGH THE CALENDAR
YEAR. LIST ABOVE SHOWS CONFIRMED
UPCOMING MEETINGS AND THEIR RESPECTIVE DATES.
FOR EXAMPLE THIS WAS RECOMMENDED BY THIS GROUP IF THE SPRING
WHERE BEGINNING OUR FORMAL CONSULTATIONS WITH THE AMERICAN
INDIAN AND ALASKAN NATIVE TRIBAL LEADERS.
WE’LL CONDUCT THAT NEXT MONTH.
AS JASON TEASED YOU WITH WE’RE SPONSORING A TWO-DAY PANEL
WORKSHOP CURRENTLY SCHEDULED FOR DECEMBER OF THIS YEAR.
WE’RE STILL IN THE BEGINNING PLANNING STAGES OF THAT WORKSHOP
BUT I’LL JUST TAKE A FEW MOMENTS TO GIVE YOU THE FOCUS AND HOW WE
PLAN TO PROCEED. THE WORKSHOP WILL CONTINUE TO
PROVIDE AN OPPORTUNITY FOR DISCUSSIONS AND COLLECTING DATA
USER FEEDBACK ON USE CASES AND THEIR NEEDS FOR THE 2020 DATA
PRODUCTS AND TO HELP US IDENTIFY AND ASSESS THE TECHNICAL
SOLUTIONS FOR THE DEVELOPMENT OF THE REMAINING DATA PRODUCTS.
AS WELL AS FINE TUNING THE EXISTING METHODOLOGY.
STAFF WILL CONVENE A MULTIDISCIPLINARY GROUP OF
EXPERTS TO SERVE AS THE PLANNING IN THE FOR THIS WORKSHOP.
OF COURSE IN CONSULTATION WITH THE CENSUS BUREAU.
IN ADDITION, THE WORKSHOP WILL ALLOW PRESENTATIONS ON THE
RELEASE AND ASSESSMENT OF THE 2010 DEMONSTRATION PRODUCTS.
HEARD US TEASE A LITTLE BIT ABOUT THAT SO I’LL TALK MORE
ABOUT THAT. THAT WILL BE THE FOCUS OF THE WORKSHOP IN DECEMBER.
>>SOW — SO WE PLAN TO RELEASE THE CAPABLES OFF THE CURRENT
DASH SYSTEM. IT WILL BE RUN ON THE INTERNAL
2010 DATA KNOWN AS THE SEC AND CREATE TWO SPECIFIC DATA
PRODUCTS TO BE PACKAGED IN THIS DEMONSTRATION PROJECT.
ONE IS THE PL94 AS SPECIFIED WITH THE RELEASE OF THE 2018 TESTS.
AS WELL AS THE DHC FILE JASON JUST WENT THROUGH.
IT’S ACTUALLY SELECTED TABLES FROM THAT PRODUCT AND IT IS
BASED ON THE CURRENT METHODOLOGY FOR DAS.
THE TABLES WE CAN SUPPORT WE PLAN TO RELEASE RUNNING THE DAS
ON THE 2010 DATA. THIS WILL ALLOW THE DATA USER
COMMUNITY TO ASSESS THE IMPACTS OF THE CURRENT METHODOLOGY.
THE TARGET RELEASE DATE IS MID OCTOBER SO NEXT MONTH.
AND AS JASON POINTED YOU TO THAT SPREAD SHEET THAT IDENTIFIES THE
TABLES WITHIN THE DHC, WE’LL BE CREATING A VERY SIMILAR SPREAD
SHEET SO WE’LL IDENTIFY WHICH PARTICULAR TABLES ARE INCLUDED
OR WILL BE INCLUDED IN THE 2010 DEMONSTRATION PRODUCTS AS WELL
AS THE PL IN THE SPECIFICATIONS. WE’RE CONTINUING TO WORK ON THE
RELEASE STRATEGY OF THIS SO STAY TUNED FOR MORE INFORMATION TO BE
PROVIDED VERY SOON AS WE WORK OUT THE DETAILS.
AGAIN, IT WILL BE RELEASED PUBLICLY TO ALLOW ALL DATA USERS
TO DO ASSESSMENTS WHILE THE FOCUS OF THE WORKSHOP WILL BE
SELECTED PRESENTATIONS, WE’LL STILL BE GATHERING FEEDBACK FROM
THE GENERAL PUBLIC. WE’RE CONTINUING TO WORK OUT THE
DETAILS OF WHAT THAT PROCESS LOOKS LIKE.
WE’LL KEEP BRIEFING YOU.
>>THANK YOU.>>CAN YOU BACK UP TO THE
QUESTIONS FOR PEOPLE. EACH ONE OF US REPRESENTATIVES
SOME CONSTITUENCY WITH A DIFFERENT ANGLE AND OTHER
CONCERNS I’M SURE WILL COME UP IN THE DISCUSSION.
I LOVE THE SPREAD SHEET. IT WAS VERY HELPFUL.
LOOKING FROM THE LARGE SET OF INPUTS YOU HAVE TO SYNTHESIZE,
I’M SURE THERE’S SOMEONE IN THE COUNTRY WHO THINK EVERY TABLE IS
A MUSS -HAVE SO THE CRITERIA IS HARD TO
SORT OUT. THESE ARE QUESTIONS WE CAN TALK
ABOUT DURING YOUR RESPONSE BUT I’D LOVE TO HEAR IF YOU HAVE ANY
FEEDBACK FROM THE PUBLISHING OF THE PROTOTYPES FROM THE REDISTRICTING FILE THEMSELVES.
THE DISSEMINATION PLAN FOR THE 2010 DATA PROJECT YOU METHED THE
WORKSHOP. — MENTIONED THE WORKSHOP.
I DIDN’T CONNECT THE TWO PIECES. I THINK THAT’S HELPFUL BUT
OBVIOUSLY ONLY SOME PEOPLE WILL BE AT THE WORKSHOP.
SO THINKING ABOUT SOME YOU’LL SEE THROUGHOUT THE PRESENTATION
A LOT OF MY FOCUS IS ON GETTING THE WORD OUT TO DIFFERENT TYPES
OF USERS. SO THINKING ABOUT OTHER
ACADEMICS OR OTHER PEOPLE AROUND THE COUNTRY THAT CAN’T ATTEND
THE WORKSHOP IN PERSON. NOT EVERYONE IS IN D.C.
I THINK IT’S HARD TO COMMENT ON THE TIMETABLE WHEN IT’S LIKE
EVERYTHING WILL BE EXPLORED ON THE FS1 AND OTHER PIECES HARDER.
I THINK PEOPLE WILL BE INTERESTED IN SEEING THAT.
I SAW LIKE LAST TIME THE SF1 WAS RELEASED A MONTH AFTER THE
DEMOGRAPHIC PROFILES IN 2010. THE IDEA IT WILL BE
SIGNIFICANTLY LONGER IF IT’S SIX MONTHS LATER OR SOMETHING LIKE
THAT, GETTING PEOPLE PREPPED FOR DATA LATER THAN THEY EXPECT.
PEOPLE PLAN WAY IN PREPARING THE DATA.
IF YOU EXPECT THE RELEASES TO BE MUCH LATER THAT WOULD BE GOOD TO
KNOW EARLIER THAN LATER. BUT I THINK THE PROTOTYPE FILES
ARE FANTASTIC. AND IN THE END, PEOPLE WANT IT
KNOW HOW IS THIS AFFECTING WHAT I NEED TO DO.
SIMILAR TO THE SAME MESSAGES WE HEARD IN THE OUTREACH PIECE,
THEY DON’T CARE ABOUT THE 2020 CENSUS BUT WANT TO KNOW HOW IT
AFFECTS MY COMMUNITY OR MY NEEDS AND WORK.
SO I THINK THE LIST YOU HAD ON THE PIECE IS GREAT.
I WAS DISAPPOINTED TO SEE IT’S STILL REALLY ACADEMIC FOCUSSED.
AND MAYBE YOU’VE DONE OUTREACH AND MAYBE YOU COULDN’T LIST
EVERYONE. THAT’S TOTALLY POSSIBLE.
I THINK WE SAW THERE WAS A COMMENT LETTER FROM THE NATIONAL
CONGRESS OF AMERICAN INDIANS. I WAS PLEASED TO SEE THE TRIBAL
CONSULTATIONS GOING FORWARD AND I THINK IT WILL BE A DIFFICULT
TASK. I’VE HAD EXPERIENCE IN THE SHORT
COMINGS OF THE TRIBAL DATA THAT EXIST ALREADY.
I THINK IT WILL NEED TO HAVE A LOT OF CONVERSATION AND OTHER
CONSTITUENCIES CAN HELP CHIME IN ON THE IMPACTS OF THE FEWER TABS
ON THE RACE AND ETHNICITY PIECES BUT SOME OF
THIS TOO WE RECOMMENDED ANALYSIS OF THE FEDERAL REGISTER ABOUT
WHO YOU GOT ANSWERS FROM.
SO WHETHER THESE FOLKS WERE ADEQUATELY REPRESENTED IN A
FEDERAL REGISTER PROCESS WHICH IS NOT REALLY INCLUSIVE.
SOME WILL FOLLOW THAT PIECE.
WE HEARD A LOT YESTERDAY ABOUT YOUR REALLY EXTENSIVE CUSTOMER
ASSISTANCE MANAGEMENT PROGRAM AND THE PARTNERSHIP WITH 20,000 GROUPS.
IN THINKING ABOUT THE DISSEMINATION STRATEGY AND HOW
TO LEVERAGE THE GREAT WORK ON THE OUTREACH TO GET FEEDBACK ON
THE PRODUCTS. IT SEEMS IT’S BEING HANDLED
SEPARATELY IN SILOS. BUT ALL THOSE PEOPLE — A SUBSET
OF THOSE PEOPLE HONESTLY INTERESTED IN OUTREACH WILL ALSO
BE INTERESTED IN USING THE DATA. YOU HAVE THE LIST ALREADY AND
THEY’RE LISTENING TO YOU ALREADY IN A FRIENDLY TRUSTED WAY.
IT’S SEEMS LIKE A READY MADE VEHICLE TO GET THAT OUT.
I KNOW THE PARTNERSHIP SPECIALISTS HAVE A LOT OF ON
THEIR PLATE BUT USING THE DISSEMINATION SPECIALISTS
PLATFORMS OR OTHER PIECES TO DO THAT OUTREACH BECAUSE WE KNOW
PARTNERSHIP PIECES — THE PARTNERSHIP REPRESENTATION IS
MUCH MORE DIVERSE THAN THE FOLKS THAT WOULD WATCH THE LIVE STREAM, FOR EXAMPLE.
I ALSO THINK — WELL, I’LL SKIP THAT.
TO GO ON TO SPECIFICS AND IN THINKING ABOUT THE SPECIFICS ONLY A SMALL TABLE
WERE NOT INCLUDED. THE NOS.
I WOULD RECOMMEND CHANGES ON THE SPREAD SHEET I’LL SEND YOU
SEPARATELY. ONE FOR EXAMPLE ABOUT WHETHER
THE TABLE IS AVAILABLE ON ACS. SO I THINK ONE OF THE QUESTIONS
IS WHERE ELSE CAN I GET THIS DATA FOR PEOPLE LOOKING AT IT.
IT WOULD BE HELPFUL IF THERE’S A COMPARABLE TABLE ON ACS AND I
DIDN’T DO THE CROSSWALK BECAUSE THAT WOULD BE ANOTHER PLACE FOR PEOPLE TO GO.
SOME OF THESE I ALSO RECOGNIZE MAY BE SOLVED, REPRESENTED IN
FUTURE DATA PRODUCTS WE DON’T KNOW ABOUT YET.
SEEING WHICH ONES WERE DIFFICULT TO PRODUCE UNDER THE NEW SYSTEM
GAVE FORESHADOWING OF THE ISSUES YOU’LL BE FACING WHEN WE
SEE THE PROPOSED TABLES AND IN SF1 THERE’S OFTEN
TABLES ON OWN CHILDREN, RELATED CHILDREN AND ALL CHILDREN.
SO OBVIOUSLY FOLKSED THERE WE HEARD A LOT YESTERDAY ABOUT HOW
THE INCREASE COMPLEXITY OF FAMILY TYPES SO HAVING LESS
INFORMATION ABOUT THAT MAY HAVE IMPLICATIONS FOR THOSE
SERVING CHILDREN OR ANALYZING THE DEMOGRAPHERS AND PEOPLE CAN
SPEAK UP ABOUT WHAT THEY KNOW ABOUT CHANGING FAMILY STRUCTURE
AND THE HOUSEHOLD TYPE BY RELATIONSHIP FOR PEOPLE 65 YEARS
AND OVER. I WAS THINKING ABOUT THE MASSIVE
CONCERN ABOUT ELDER CARE AND HOME HEALTH NEEDS AND HOW THAT
MIGHT EFFECT PEOPLE THINKING ABOUT THE SENIOR SERVICES FROM A
CITY PERSPECTIVE, FOR EXAMPLE. SO FOR EVERY ONE CAUGHT — CUT OUT ANOTHER ONE HAD AGES
THREE TO FOUR PULLED OUT SO PEOPLE INTERESTED IN SCHOOL READINESS.
THE COMBINATIONS VALENTINE — THE COMBINATIONS WILL BE
AVAILABLE AND WE KNOW LESS ABOUT RACIAL DISPARITIES AND
CONDITIONS FOR RACIAL GROUPS AND I UNDERSTOOD ON THE LAST ONE WE
JUST NEEDED TO UPDATE THE 2010 METHODOLOGY AND IT NEEDED TO BE
UPDATED ON THE DOCUMENTATION AND SPREAD SHEET.
I KNOW YOUR HEAD IS IN THE STATS AND DOCUMENTATION BUT THE
COMMUNICATIONS PLANNING SHOULD BE VALUED.
THE DATA USER GROUP SHOULD BE VALUED AND YOU CAN LEVERAGE
WHAT’S BEEN DONE AROUND PRIVACY AND WA WE KNOW FROM CBAM THAT
PEOPLE VALUE. YOU ARE BUILDING THIS ON A
REASON THAT PEOPLE HAVE ALREADY SAID THEY VALUE AROUND THE
PRIVACY OF INFORMATION. IT’S JUST REALLY HARD.
I CAN IMAGINE IT WILL BE HARD TO FOLLOW.
THIS IS MORE LIKE WHAT I WOULD WANT FROM MY CONSTITUENCY.
SOME EASY PAGE LIKE THE INFORMATION ABOUT DAIT — DATA
PRODUCT THERE AND THE MEMOS LIKE 57 PIECES DOWN THAT ARE REALLY
COMPLICATED WAS HARD TO FIND. EVEN FINDING THE P.O. TECHNICAL
APPLICATION FILE TOOK TIME FOR ME WHEN I HAD THE TITLE.
GENERALLY BEING MORE TRANSPARENT AND PUTTING THINGS IN AN EASIER
ACCESS FORMAT WOULD BE HELPFUL AND I’M MAINLY THINKING AHEAD
FOR YOU. I KNOW THERE’S BEEN SOME
COMMUNICATIONS AROUND DIFFERENTIAL PRIVACY BUT
THINKING ABOUT JUST A BLOG ABOUT THESE ARE REALLY HARD.
THERE’S GOING TO BE TRADE-OFFS AND THESE ARE HARD.
WITH RELEASES OF THE DHC OR OTHERS, I THINK BRINGING THAT
INTO THE CONVERSATION FOR USERS IS PRETTY UNDERSTANDABLE.
THE TRADE-OFF. I WOULD BE UP FRONT WITH THE
USERS ABOUT WHAT YOU’RE FACING I THINK.
I DID TRY TO FIND STUFF ON THE 2020 CENSUS RESEARCH OPERATIONAL
PLANS AND OVERSIGHT WHICH IS WHERE I THINK THE PRODUCTS PAGE SHOULD BE.
RIGHT NOW EVERYTHING IS TOO VARIED TO SEE WHAT’S HAPPENING.
THIS IS THE ACTUAL DATA FILE. THE VAST MAJORITY OF THE USERS
AREN’T EVEN THINKING ABOUT THE 2020 DATA YET, RIGHT?
SO TRYING TO GET IT ON THEIR MINDS NOW.
SO THEY’RE NOT SURPRISED WHEN THE DATA COMES OUT AND THEY’RE
FAVORITE TABLE’S MISSING. I THINK THE LAYOUT OF YOUR SLIDE
EARLIER ABOUT THE INPUTS YOU’RE USING WERE HELPFUL.
I WANTED TO MORE. NOT JUST THE INPUTS BUT SOME
SYNTHESIS FROM THIS. LIKE WHAT CAUSED YOU TO NOT DO
THE TABLES YOU HAD? WAS IT JUST METHOD LOGICAL IT DIDN’T WORK OR NO ONE
ACCESSED THE TABLE ON FACT FINDER OR YOU DIDN’T GET A USE
CASE IN THE FEDERAL REGISTER SO YOU HAVE THE EVIDENCE TO SAY
THIS TABLE CAN BE CUT WITHOUT REPERCUSSIONS?
I WANTED MORE WITH THE REASONING AND I SAW THE KIDS
COUNT PUBLIC COMMENT THAT CAME THROUGH.
PEOPLE ARE FRAYED ALL THE DATA’S — ARE AFRAID ALL THE DATA’S
GOING TO BE TAKEN OUT SO AS MUCH RATIONALE WE CAN PROVIDE TO SHOW
IT’S EVIDENCE-DRIVEN AND DESCRIBING THE INPUT IN SOME LAY
FORMAT WOULD BE HELPFUL AND HOW YOU’RE MAKING DECISIONS AND AS
YOU DO MAKE DECISIONS, SOME OF WHAT YOU’RE THOUGHT PROCESS WERE
IF YOU COULD. AND I DID THIS PLUG BEFORE BUT
WE MADE A CHANGING FROM THE LONG FORM TO THE AMERICAN COMMUNITY
VARE — SURVEY WAS A WRENCHING PROCESS FOR PEOPLE ON THE GROUND
AND MANY HERE LIVED THROUGH THE PAIN AND SUFFERING.
WE KNOW HOW TO DO THIS. THE ACS USER GROUP AS A DISSEMINATION
PIECE IN THINKING ABOUT POST RELEASE HOW YOUR GOING TO SHARE
THIS WITH USERS AND GUIDANCE WOULD BE HELPFUL BECAUSE I FOUND
THEM HELPFUL DURING THAT FORMER PROCESS WHEN WE WERE DOING THE ADJUSTMENTS.
I’M COMING OUT OBVIOUSLY THE CENSUS
IS IMPORTANT FOR REPRESENTATION AND FUNDING BUT FIND OUGHT MORE
ABOUT OUR — FINDING OUT MORE ABOUT OUR
COMMUNITIES IS THE FUN PART.
I ENJOYED GOING THE WORK SO THANK YOU AND I DON’T KNOW IF
YOU WANT TO COMMENT BUT I’M SURE OTHER PEOPLE HAVE QUESTIONS TOO.
>>THANK YOU, CARTH — KATHRYN FOR GREAT COMMENTS AND
I’LL TRY TO RESPOND TO ONE SET AND CYNTHIA CAN RESPOND TO THE
OTHER SET. YOUR FIRST POINT YOU MENTIONED
THE LARGE SET OF INPUT AND LARGE AMOUNT OF INPUTS AND THE HARD
DECISIONS WE HAD TO MAKE.
TO THE INPUT WE RECEIVED COMBINED WITH OUR OWN SUBJECT
MATTER KNOWLEDGE SUPPORTS THE IDEA ONCE YOU MOVE BEYOND A
LEGALLY MANDATED PRODUCT THERE’S IMPORTANT USES FOR THE OTHER
TABLES WE PROVIDE. WE WENT THROUGH THE HIT RATES
FROM INTERNET PAGES AND WE RECEIVED DATA FROM UNIVERSITY OF
MINNESOTA’S NIGIS AND THEY GAVE US DATA ON
TABLES THEY MAINTAIN. WE WENT THROUGH THAT AND TRIED
TO FIND ACTIONABLE DIFFERENCES IN HIT RATES.
IT’S HARD BECAUSE EVERY FILE PROGRESSES THROUGH THE
TOTAL POPULATION FILES GET A LOT MORE HITS AND GO INTO DETAILS.
DON’T WANT TO DISCOUNT THE SMALL NUMBER OF HITS FOR THE DETAILED
POPULATIONS BECAUSE THEY’RE IMPORTANT FOR THE GROUPS. AS WE WORK THROUGH AND KNOW WE
HAVE THE PL THAT’S MANDATED AND LOOK HOW LONG TO FORMULATE THE
PHC, A NATIONAL BREAK EMERGED BASED MORE ON THE RISK EACH
VARIABLE PRESENTED FROM THE DISCLOSURE AVOIDANCE PERSPECTIVE.
THAT CREATED A DISTINCTION FROM WHAT WE CAN SUPPORT AND THE DHC
WHAT YOU SEE IN THE SPECIFICATION AND WHAT HAD TO BE
MOVED TO OTHER PRODUCTS. IN THE FIRST SET OF TABLES THERE WERE SMALL CHANGES
MADE AND YOU CAN SEE THE USES WERE LOWER AND WE DIDN’T KNOW
THE GOOD USE DATA AND FOR THE MOST PART YES, THE METHODOLOGY
WAS PUSHING TO MORE COMPLEX TABLES INTO THE LATER PRODUCTS
WE COULD STILL MAINTAIN ACCURACY FOR THE FIRST SET OF PRODUCTS
AND WORK ON OTHER PRODUCTS AS WELL.
AND I CAN ADDRESS ONE OF YOUR OTHER COMMENTS WITH THE
INCREASING COMPLEXITY OF FAMILY TYPES.
AS WE WORKED THROUGH THIS, IT’S A LITTLE DIFFERENT THAN IN
PREVIOUS DECADE WHERE WE PRODUCED DATA AND PROTECTED AND
HAVE YOU ALL THE VARIABLES TO WORK WITH AND FORMULATE THE
TABLES THE WAY YOU WANTED TO LATE IN THE PROCESS.
WITH DIFFERENTIAL PRIVACY YOU HAVE TO BUILD.
IF YOU WANT A TABLE WITH AGE AND HOUSEHOLDS OF A
PARTICULAR AGE YOU HAVE TO BUILD THAT IN THE AVOIDANCE SYSTEM UP
FRONT BECAUSE YOU LOSE THE CONNECTION WEN MEMBERS OF THE
HOUSEHOLD AND THEIR FAMILY MEMBERS.
THAT’S SOMETHING WE HAD TO CONTEND WITH AND THE OTHER
TABLES WILL BE PART OF ANOTHER DIFFERENTLY PRIVATE SOLUTION.
TRYING TO GO THROUGH. I WROTE A LOT OF NOTES FROM YOUR
COMMENTS AND YOU COVERED A LOT OF GROUND.
I THINK THE TIMING OF THE DHC WILL BE WHAT YOU FALL IN FOR THE
SF1 BECAUSE IT’S THE SAME DISCLOSURE AVOIDANCE SYSTEM WILL
PROTECT THE PRODUCT AND DLC TABLES.
WITH THAT I’LL LET CYNTHIA TALK ABOUT THE OUTREACH ASPECTS OF YOUR COMMENTS.
>>THANKS, JASON. THANK YOU, KATHRYN FOR YOUR
COMMENT. ONE THING I SHOULD HAVE POINT
THE OUT ABOUT THE PLANNED WORKSHOP IS WE DO PLAN TO
WEBCAST IT LIVE. FOR THOSE WHO CANNOT ATTEND IN
PERSON THEY CAN STILL EXPERIENCE THE WORKSHOP.
ALSO, WITH THE FACT WE’RE GRABBING THIS PLANNING COMMITTEE
IS VARIOUS BACKGROUNDS SO THEY KNOW WHO THEIR PEOPLE ARE AND
WHO THEY SHOULD GO TO AND THEIR CONSTITUENCIES.
THAT WAS THE POINT OF GRABBING THE GROUP AND WE SHOULD BE
REPRESENTATIVE. AND JASON POINTED OUT WITH THE
RELEASE SCHEDULE WE’RE TRYING AS BEST AS WE CAN TO KEEP THE SAME
RELEASE SCHEDULE AS WE DID IN 2010 AT LEAST FOR THE PL BECAUSE
WE MANDATE IT AND OF COURSE METADATA AND FOR THE DHC.
AS WE CONTINUE TO FINALIZE THE METHODOLOGY FOR THE REMAINING
PRODUCTS, OUR GOAL IS TO TRY TO STICK TO THE SCHEDULE WE’VE HAD
IN THE PAST AS BEST AS POSSIBLE BUT WE WILL BE TRANSPARENT ABOUT
THAT AND BE VERY OPEN AND COMMUNICATE THAT.
THE OTHER THING I JUST WANTED TO TOUCH ON IS YES, WE DO OUTREACH
AND TALK TO VARIOUS GROUPS. WHEN YOU MENTION THE ACS USER’S GROUP WE HAVE A
LEAD-IN WITH THE POPULATION BUREAU AND WE WORK CLOSELY WITH
THEM AND PROVIDE INFORMATION AND GAIN FEEDBACK AND WE’RE TRYING
TO AS WE COMPLETE THE DETAILS OF HOW TO GET INPUT BASED ON THIS
RELEASE, WE WILL INCLUDE THAT GROUP TOO SO THEY CAN GET THAT
INFORMATION OUT TO THE USERS GROUP.>>GREAT.
WE’LL HAVE MORE QUESTIONS. I JUST WANT TO ECHO WHAT CATHY
SAID. THIS IS THE KIND OF THING JUST
FREAKING PEOPLE OUT. THE MODEL OF HOW YOU ALL DID IT
WAS BRILLIANT. THOSE USERS GUIDES ARE
FANTASTIC. I LOVE THEY’RE SPECIFIC BY
AUDIENCE. INCREDIBLY HELPFUL.
I WANT TO ECHO THAT RECOMMENDATION FROM HER. PUSH.
— MICHELLE.>>THIS IS RICHELLE WINKLER.
THANKS FOR THE DETAILED PRESENTATION.
I LOVE ALL THE TABLES THAT YOU PROVIDE AND YOU’RE DOING WORK ON
A DIFFICULT TOPIC SO THANK YOU.D AND YOU’RE DOING
WORK ON A DIFFICULT TOPIC SO THANK YOU.
I HAVE A COUPLE QUESTIONS ONE IS EASY TO ANSWER.
REGARDING THE RESEARCH DATA CENTERS AND RDCs AND WONDERING
IF THE FULL DATA OR HOW IS DAS GOING TO BE IMPLEMENTED IN
SETTINGS LIKE THAT WILL PEOPLE STILL BE ABLE TO ACCESS DATA AT THE CENTERS.
MY OTHER QUESTION, YOU TALKED ABOUT A LITTLE BIT IN THE
PRESENTATION MORE THAN I SAW IN THE MATERIALS PROVIDED AHEAD OF
TIME ABOUT PUMS. I WAS HOPING YOU COULD TALK A
LITTLE BIT MORE ABOUT MAYBE ANY TIMETABLES FOR WHEN THOSE FILES
MIGHT BE RELEASED. IT SOUNDED LIKE INDIVIDUALS WILL
NO LONGER BE ABLE TO BE NESTED WITHIN HOUSEHOLDS.
I WANTED TO UNDERSTAND THAT A LITTLE BIT BETTER AND YOU
MENTIONED SOMETHING IN THE PRESENTATION ABOUT SOME OF THE
VARIABLES MAY NOT BE AVAILABLE IN THE PUM FILE.
AND SO I JUST LIKE TO HEAR A LITTLE BIT MORE DETAIL ON THAT.
THANK YOU.>>LET ME START WITH THE FRDC
QUESTION. AT THE MOMENT, RESEARCH IN THE
SFRDCs CANNOT ACCESS THE CENSUS EVIDENCE FILE FROM 2010.
IT ACCESSES THE VERSE — VERSION PREPARED KNOWN AS THE
100% DETAIL FILE WHICH HAS HAD DISCLOSURE AVOIDANCE APPLIED TO
IT. WE HAVE A CONTINUING COMMITMENT
TO MAKE THE RESOURCES OF THE CENSUS BUREAU AVAILABLE IN
SFRDCs BUT DON’T AT THE MOMENT HAVE THE EQUIVALENT FOR THE 2020
SEF. WE’LL HAVE TO EXPERIMENT WITH
THE METHOD BY WHICH THE 2020 CENSUS WILL BE EDITED.
THAT’S NOT A NO BUT SITTING THERE WITH A PUBLIC USE MICRO
SAMPLE WHICH IS WHERE I’LL GO NEXT.
TECHNOLOGY BEING USED TO DO ALL THE PRODUCTS, JASON AND CYNTHIA
JUST DISCUSSED IS THE TOP-DOWN ALGORITHM WHICH WAS THE FIRST
PRODUCTION ALGORITHM FOR PRIVACY IN THE 2020 CENSUS ENHANCED TO
COVER SEPARATELY TABLE HOUSING UNITS AND PERSONS.
THE SCIENCE DIDN’T EXIST AT THE TIME WE DEVELOPED THE PRODUCTION
ALGORITHMS TO DO PERSONS WITHIN HOUSEHOLDS OR DO DETAILED RACE
AND ETHNICITY. AT THE TIME THESE PRODUCTS THEY
JUST DISCUSSED WERE ISOLATED FROM THE BULK OF THE OTHER
PRODUCTS, WE WERE WORKING WITH PRODUCTION SYSTEM OR BASICALLY A
BETA PRODUCTION SYSTEM, LET’S BE HONEST, WITH A KNOWN SCIENTIFIC
SOLUTION. SINCE THEN, PEER METHOD SCIENTIFIC
SOLUTIONS FOR DETAILED RACE AND ETHNICITY AND FOR JOIN TABLES
HAVE BEEN DEVELOPED AND WE HAVE READ A CONTRACT TO USE THE
SYSTEMS TO DEVELOP THE SYSTEMS JASON AND CYNTHIA TALKING ABOUT
OUTSIDE OF DHC. WE CAN’T PROVIDE MORE DETAILS
RIGHT NOW BECAUSE WE HAVEN’T EXAMINED SAMPLES BUT THE
INTERNAL TEAM IS FAMILIAR WITH THE SCIENCE THAT UNDERLIES THOSE
AG RIMS AND — ALGORITHMS AND NOT FAMILIAR WITH
IT BUT WE’RE TALKING ABOUT A SYSTEM KNOWN AS ECFLO AND PRIV
SEQUAL SOLVES THE JOIN PROBLEM BUT IT DOES SO BY REQUIRING THE
ANALYST SPECIFY A SLIGHTLY MORE DETAILED VERSION OF THE
DIFFERENTIAL PRIVACY POLICY APPLY TO THE TABLES.
SO OUR OWN ANALYST NEEDS SOME EXPERIENCE IN DOING THAT BEFORE
THEY CAN GET COMFORTABLE WITH THE OUTPUT.
IN ADDITION, NEITHER OF THAT TECHNOLOGY DOES NOT DELIVER
MICRODATA TO A TABULATION SYSTEM SO WE HAD TO AGREE RELAX THE
ASSUMPTION AND REQUIREMENT THE SYSTEM HAD TO PRODUCE MICRO
DATA. IT WILL PRODUCE THE TABLES THAT
ARE REQUESTED SO NOW WE NEED A PUBLICATION ENGINE THAT CAN
DISSEMINATE THE TABLES WITHOUT PASSING THEM THROUGH OUR OWN
TABULATION ENGINE. THAT’S ALSO A COMPONENT OF THE
CONTRACT WE JUST LED. WE HAVE MADE CONSIDERABLE
PROGRESS BUT APPROPRIATELY ARE NOT ABLE TO GIVE THIS COMMITTEE
OR ANYONE ELSE SPECIFICS ABOUT THAT UNTIL THE CONTRACTS HAVE TO
START FIRST AND THEY HAVEN’T STARTED YET AND THE WORK PRODUCT
HAS TO BE DELIVERED AND REVIEWED BY THE PEOPLE INSIDE THE CENSUS
BUREAU RESPONSIBLE FOR THAT AND THEN CAN BEGIN DISCUSSING IT
PROBABLY. THAT’S A LONGER ANSWER THAN I
THINK YOU WERE EXPECTING BUT I THINK THIS COMMITTEE IN
PARTICULAR NEEDS TO UNDERSTAND WHAT THE PLAN IS IN SOME DETAIL
AND I HOPE THAT WAS HELPFUL.
>>THIS IS RICHELLE. YES, THANK YOU FOR THE DETAIL.
THE BIG PICTURE I TAKE AWAY FROM IT IS
YOU HAVE BIG PLANS AND THEY’RE IN THE PROCESS.
>>JUAN PABLO.
>>THANK YOU, CYNTHIA AND JASON FOR YOUR PRESENTATION.
I HAD A SUGGESTION TO TRY TO EXPAND THE SET OF USERS IN THE
FUTURE AND TAKE ADVANTAGE OF DECENNIAL AND VISIBILITY THE
CENSUS WILL HAVE THIS COMING YEAR TO DO THAT.
I THINK ONE WAY THAT CAN BE DONE IS TO THINK OF ALL THE GROWTH
AND DATA SCIENCE AND ANALYTICS AND INFORMATION, VISUALIZATION
ACADEMIC PROGRAMS AND COURSES OUT THERE NOW.
THESE ARE TYPICALLY THE TEXTBOOKS USED AND THE PEOPLE
WHO TEACH THOSE COURSES ARE TYPICALLY NOT FAMILIAR WITH
CENSUS DATA BUT I THINK CENSUS DATA IS IDEAL FOR THOSE COURSES
AND VERY INTERESTING. IT’S FREE DATA.
THERE’S A LOT OF DATA. AND I THINK IT MAY BE AN
OPPORTUNITY TO EXPAND THE SET OF PEOPLE WHO ARE FAMILIAR WITH CONSISTENCY DATA
PRODUCTS. AND MAKE THEM APPRECIATE THEM
AND HOPEFULLY USE THEM LATER IN WAYS THEY HAVEN’T BEEN USED.
IT’S NOT A MUST BUT AN INTERESTING OPPORTUNITY.
>>I’M ASSOCIATE DIRECTOR FOR COMMUNICATIONS.
WE FIND IT TO BE AN EXCITING AND INTERESTING OPPORTUNITY.
WE WENT OVER THE INTEGRATION COMMUNICATIONS CONTRACT
YESTERDAY AND I DON’T KNOW IF THERE’S A SEPARATE SLIDE BUT WE
HAVE A DATA DISSEMINATION PLAN WE’LL BE STARTING NOW AND
REVIEWING GOING FORWARD. WE LOOK FORWARD TO SHARING THOSE
DETAILS WITH THE COMMITTEE IN THE FUTURE AND WE SEE THE
OPPORTUNITY TOO AND PLAN A ROBUST PLAN TO REACH OUT AND ARE
LOOKING AT CENSUS DATA.>>KEN SIMONSON FOLLOWING ON
CATHY’S RECOMMENDATION I RECOMMEND CYNTHIA YOU INCLUDE
THE HOUSING STATISTICS USERS GROUP THE NATIONAL ASSOCIATION
OF HOME BUILDERS HOSTS ANNUALLY. THEY GET TOGETHER TRADE
ASSOCIATIONS AND RESEARCH ORGANIZATIONS SUCH AS HARVARD
JOINT CENTER FOR HOUSING STATISTICS.
>>CHRIS.>>CHRIS MOORE.
I’M REAL INTERESTED IN DATA ABOUT CHILDREN AND YOUTH.
IT LOOKS LIKE IT’S QUITE PROMISING.
AGE BREAK AND I’M NOT IMMEDIATELY FAMILIAR WITH TABLE
A1. I WANTED TO FOLLOW ON THE CROSS
TO BE — TABULATIONS. IT LOOKS LIKE THEY’LL BE
PROVIDED AND WONDERING AT WHAT LEVEL OF GEOGRAPHY IT WILL GO DOWN TO.
IT WAS SLIDE 22.
>>SO THIS IS A LATER TABLE BUT YOU ASKED ABOUT SINGLE YEAR OF
AGE UP TO A BLOCK LEVEL TABLE UP TO AGE 20 AND SINGLE YEAR OF AGE
AT THE TRACK LEVEL.>>AND THEY LOOK LIKE COUNT DATA.
THEY’RE NOT BY ADDITIONAL VARIABLE.
>>THE TABLES ARE THE ONES IT RATED BY THE RACE GROUPS.
>>SO THERE WILL BE CROSS TABULATIONS WITH THESE
CATEGORIES. LIKE FOR EXAMPLE, FAMILY STRUCTURE.
>>NOT IN THE CURRENT VERSION OF THE DHC RIGHT NOW.
NO, THEY DO NOT EXIST.
AND WE’RE WORKING ON THE DETAILS OF RACE AND FAMILY HOUSEHOLD
JOINED.>>THE 5522 — 522, CAN YOU SAY MORE ABOUT
WHAT IT IS. IT LOOKS GOOD.
>>IT’S DIFFICULT TO RECALL THE DETAILS OF OUR TABLES.
BUT THIS TABLE SHOW IT BY AGE ALONE AND THIS IS ONE OF THE
SETS OF ABLES REITERATED BY RACE.
IT’S ONE OF THE IDEAS OF PROVIDING THE SPECIFICATIONS AND
WHAT WE PROVIDED IN THE SPECIFICATION IS WHAT WE’RE
PROPOSING. THIS IS WHAT IT WAS IN 2010.
THAT PROVIDES WHAT WE’RE PROPOSING FOR 2020.
IF YOU HAVE TABLES ABOUT HOW TO INTERPRET THE DETAILED TABLES
WHEN THEY’RE LAID OUT, IF YOU FOLLOW UP WITH US WE CAN PROVIDE MORE INFORMATION.
>>THERE’S A BLOCK LEVEL WITH AGE UP TO 20 YEARS OF AGE.
>>JOHN.>>JOHN CZAJKA.
WILL THEY BE RELEASED BASED ON ONE EPSILON VALUE?
IS THAT DETERMINED?
>>THIS IS CYNTHIA. YES, THEY’LL BE RELEASED AT ONE
EPSILON VALUE AND WE’RE IN THE PROCESS OF DETERMINING WHAT THAT VALUE WILL BE.
IT WILL BE DECIDED ON AS A POLICY DECISION.
ONCE THAT IS DONE, WE’LL RELEASE THAT INFORMATION AS PART OF OUR
DECISION MEMO BUT IT WILL ONLY BE ONE EPSILON VALUE.>>THANK YOU.
>>I HAD A QUESTION. ON THE GROUP QUARTERS DATA FOR
THE REDISTRICTING YOU DIDN’T HAVE THE BREAKOUT SO IT MEANT
ALL THE POPULATIONS WOULD BE INCLUDED BUT WITH THE BLOCK
STATES CAN CHOOSE TO INCLUDE OR EXCLUDE THEM AS THEY SEE FIT.
IS THAT WHAT THAT MEANS?>>I MISSED THAT QUESTION.
CAN YOU REPEAT IT, PLEASE.>>I KNOW I TALK FAST.
THE GROUP QUARTERS DATA IN THE REDISTRICTING FILE, YOU NOW HAVE
A BREAKOUT ON THAT. SO PREVIOUSLY I’M ASSUMING THAT
MEANS ALL THOSE POPULATIONS WERE IN THOSE COUNTS?
NOW STATES CAN CHOOSE TO INCLUDE THEM OR EXCLUDE THEM.
>>I’LL TAKE THAT ONE. WHAT HAPPENED IN 2010 IS THAT
PARTICULAR TABLE WAS PRODUCED IN A PUBLICATION AND YOU CAN FIND
IT ON THE 2010 FULL PAGE CALLED THE ADVANCED GROUP QUARTERS
TABULATION. IT WAS PRODUCED AND STATES USED
THAT ADVANCED GROUP QUARTERS TABULATION.
OVER THE COURSE OF THE DECADE THE REDISTRICTING OFFICE IN ITS
NEGOTIATIONS WITH THE NATIONAL CONFERENCE ON STATE LEGISLATURES
ASK THAT BE INCLUDED DIRECTLY IN THE PL94 PACKAGE.
IT’S THE SAME TABLE DONE IN 2010.
IT’S JUST NOW MOVED FROM A SPECIAL PUBLICATION THE ADVANCED
GROUP QUARTERS TABULATION INTO THE PL94 AND IT’S EXACTLY THE
SAME TABLE LAYOUT.>>SO IT’S NOT REALLY A CHANGE
BUT BY HAVING THE BREAKOUT STATES CAN INCLUDE OR EXCLUDE AS
THEY SEE FIT?>>IT’S IMPORTANT TO EMPHASIZE
CENSUS BUREAU UNDER PUBLIC LAW 74171 PROVIDES DATA TO THE
STATES. IT DOESN’T MAKE DECISIONS HOW
THE DATA WILL BE USED FOR THE PURPOSE OF REDISTRICTING.
>>I’M TRYING TO UNDERSTAND THE FUNCTIONALITY.
I’M A LITTLE SLOW. I WAS JUST TRYING TO UNDERSTAND
THE FUNCTIONALITY. IT SEEMS LIKE THE ANSWER’S YES?
>>IN TERMS OF FUNCTIONALITY IT’S IDENTICAL.>>TOMMY, DO YOU HAVE A
QUESTION?>>TOMMY WRIGHT.
I I USED TO WORK WITH TRANSPORTATION TABLES AND SO ARE
USERS WEIGHTED TO THESE TABLES. AND EVERYBODY’S LOOKING AT THE
SAME THING AT THE SAME TIME SO THAT COULD BE AN ADVANTAGE BUT
DOES A USER SAY I WANT ACCESS TO EVERYTHING EVERYTHING AND EVERY
RESPONSE AND I WANT TO FORM MY OWN TABLES.
ARE YOU FEELING THIS IN THE FEDERAL REGISTRY NOTICE.
DO WE THINK OF THIS IN THE FUTURE AND WITH THE DIFFERENTIAL
PRIVACY AND PROTECTION WOULD TO BE ACCOMMODATE THESE THINGS.
WHAT DO USERS REALLY WANT? ARE THEY FOCUSSED ON THESE
TABLES BECAUSE WE’VE BEEN HISTORICALLY PRODUCING HE’S
TABLES. DOES ANYONE SAY I WANT TO GET TO
THE DATA AS MUCH AS I CAN? THEY HAVE TO LEARN HOW THE
TABLES ARE PRESENTED. AS WE SAID, THIS IS VERY
PAINFUL. SO CAN THE USER COME TO A
WEBSITE AND SAY I WANT IN PRINCE GEORGE’S COUNTY THE NUMBER OF
MALES OVER 32 EMPLOYED AS ENGINEERS.
FOR EXAMPLE, IT’S IN THE UNDERLYING DATABASE.
AND IT WOULDN’T BE IN THE DECENNIAL CENSUS BECAUSE OF
ENGINEERS BUT ARE USERS STILL TALKING ABOUT THESE TABLES?
>>THIS IS RICHELLE. I THINK WHAT USERS WOULD LOVE TO
HAVE THE KIND OF ACCESS YOU TALKED ABOUT.
YOUR EVERY DAY USERS, THAT’S WHAT WE WOULD LOVE TO HAVE.
AND I REMEMBER YEARS AGO THERE, I WORKED AT A STATE DATA CENTER
AND WE HAD ACCESS TO A NICE WEBSITE TO CREATE OUR CROSS
TABS. IT’S NO LONGER AVAILABLE AND I
DON’T REMEMBER THE DETAILS ABOUT IT BUT I USED THAT ALL THE TIME
BECAUSE IT WAS STUFF THAT WASN’T IN THE PUBLISHED TABLES.
BUT I COULD CREATE CROSS TAB WOULD BE USEFUL TO THE PARTNERS
WE’D WORK WITH AND I THINK THAT’S WHAT USERS WOULD LOVE TO
HAVE ACCESS TO IF IT WERE AVAILABLE.
>>TOMMY WRIGHT. MAYBE IT’S A VISION OF SOME SORT IN THE FUTURE.
>>ARE WE AHEAD OF SCHEDULE? THANK YOU, JACK.
>>TOMMY, WHAT A GREAT THOUGHT. I WANT TO JUMP ON THAT WITH WHAT
JUAN PABLO WAS SAYING LIKE BACKUP WHAT IS THE DEFINITION OF
A USER RIGHT NOW. MY BELIEF IS THAT THAT’S
CHANGING. IN THE LAST 10 YEARS, OR AT
LEAST TODAY, THERE ARE SO MANY MORE DATA SCIENTISTS, ANALYSTS,
CALL IT WHATEVER YOU WANT IN BUSINESSES EVERYWHERE AND MY GUT
IS THAT AS JUAN PABLO SAID, THEY’RE NOT AWARE OF ALL OF
THESE TABLES RIGHT NOW.
WITH ALL THAT BUSINESS INTELLIGENCE AND PEOPLE WANT
ACCESS TO THE DATA, THEY MAY DWARF THE EXISTING USERS YOU’RE
USED TO. IT MAKE SENSE TO DO WHAT TOMMY
SAID AND STEP BACK AND SAY WHO ARE THE USERS OF THE FUTURE AND
WHAT ARE THOSE USERS GOING TO WANT BECAUSE I CAN SEE IT AND I
CAN FEEL IT IN MY BUSINESS WHERE I WAS.
I CAN SEE IT IN EVERY PLACE. THOSE ARE THE CALLS I GUESS.
SO I THINK YOU NEED TO THINK AHEAD OF WHO IS THE NEXT USER
AND HOW DO YOU GIVE THEM ACCESS TO THE DATA AND WHAT WILL THEY
WANT? I’M PRETTY SURE THEY’LL DWARF
THE USERS OF 10 YEARS AGO.
>>I’LL RESPOND TO THE SECOND PART OF TOMMY’S QUESTION WHICH
IS HOW COMPATIBLE IS THIS WITH PRIVACY PROTECTION.
UNFORTUNATELY THE ANSWER IS IF YOU WANT USERS TO ASK ARBITRARY
QUESTIONS THERE’S A MATHEMATICAL PORTION TO THE DATABASE THAT
STILL SAYS WE CANNOT ALLOW REASONABLE ACCURACY OF THE DATA
SET. I’M KUNAL TALWAR.
AND IT’S POSSIBLE THAT IF THE USER IF WE RESTRICT THE KINDS OF
QUESTIONS THE USER CAN, SAY THEY CAN ONLY ASK TWO OR THREE-WAY
CROSS TABS WE MAY BE ABLE TO DO SOMETHING BUT AS MUCH AS WE’D
LIKE TO BE ABLE TO ANSWER ARBITRARY THINGS IT’S
INCOMPATIBLE WITH PRIVACY.>>JOHN.>>SO A COUPLE OF COMMENTS. FIRST, RICHELLE THE PRODUCT YOU
WERE USING HAD 3 MILLION TABULATED TABLES PRELOADED
THROUGH A DISCLOSURE AVOIDANCE SYSTEM SO YOU WASHINGTON —
WEREN’T MAKING A QUERY ON DEMAND YOU’RE DOING WHAT KUNAL SAID AND
IF YOU ASK FOR SOMETHING NOT IN INVENTORY YOU CAN’T HAVE THAT.
WHAT I WAS GOING TO SUGGEST IS IF THE COMMITTEE WANTS IT, WE
SHOULD BRING IN PEOPLE FROM OUR NEW CENTER FOR ENTERPRISE
DISSEMINATION INCLUDING THE SET SIDE DISSEMINATION UNIT IN THE
FEDERAL BUDGET FOR DEVELOPING NEW METHODS DEFAULT PRODUCTION
AND IT’S THE DISSEMINATION METHOD BEGINNING THIS MONTH WITH
ACS ONE-YEAR RELEASES AND ECONOMIC CENSUS RELEASES.
YOU SHOULD GET MORE FAMILIAR WITH IT.
IT HAS A CONCEPT OF OPERATIONS THAT IS VERY SIMILAR TO WHAT
TOMMY LAID OUT AND TO WHAT JACK REFERRED TO AND I THINK TO WHAT
YOU WERE DOING. THE IDEA IS TO GO TO
DATA.CENSUS.gov AND BE ABLE TO ASK A LESS STRUCTURED QUERY THAN
GIVE ME TABLE A00135 BECAUSE I KNOW WHAT’S IN IT AND WE’LL
STILL HAVE THE API FOR DOING THAT BECAUSE IT’S ANOTHER WAY
USERS EXPECT TO BE ABLE TO INTERACT BUT DATA.CENSUS.gov
WHICH IS UP SO YOU CAN GO THERE NOW, WAS DESIGNED TO BE MORE
USER FRIENDLY AND TO INTERACT ITSELF WITH THE VARIOUS DATA
SOURCE INCLUDING THE APIs TO GENERATE CANDIDATE RESPONSES AND
ALLOW YOU TO GO THROUGH INCLUDING GEOSPATIAL ANSWERS AS
WELL AS TABULAR ANSWERS. I THINK IT MAY BE GOOD TO HAVE
THIS GROUP SEE SOMETHING IN THE NEXT MOOETD MEETING IF YOU WANT
— MEETING IF YOU WANT. I’LL OFFER THAT.
I THINK IT WILL ANSWER MANY QUESTIONS RAISED.
>>THANK YOU VERY MUCH, JASON, CYNTHIA AND EVERYONE.
KATHRYN. WE’LL MOVE ON TO THE NEXT SESSION.
>>I SHOULD SAY DURING THE SESSION WE MIGHT HAVE MIGHT
OVERLAP WITH THE PUBLIC COMMENT AT 11:00.
WE’LL NOW HEAR FROM JOHN ABOWD AND VICTORIA VELKOFF ON METHODOLOGY AND
DEMOGRAPHIC PROGRAMS AND PUBLIC DISCUSSION.
>>WE’RE GOING TO LEAD WITH A VIDEO.
I THINK WE’RE READY.
>>EVERY YEAR WE COUNT EVERY PERSON LIVING IN THE UNITED
STATES OF AMERICA AND COLLECT INFORMATION ABOUT THEM LIKE AGE,
SEX, RACE AND ETHNICITY. THE WHOLE PURPOSE OF DOING
SURVEYS LIKE THE CENSUS AND OTHER DEMOGRAPHIC SURVEYS IS TO
GET AN OVERALL PICTURE OF HOW MANY
PEOPLE LIVE IN MINNESOTA OR MISSISSIPPI AND HOW DO THEY
DIFFER IN DIFFERENT PLACES BY SEX OR RACE?
THE RESULTS OF THE U.S. CENSUS ARE OF POLITICAL RELEVANCE SINCE
THEY’RE USED TO DETERMINE THE SEATS STATES GET IN THE HOUSE OF
REPRESENTATIVES AND LEGISLATIVE DISTRICTS FROM CONGRESS TO CITY
COUNCILS AND THEY’RE USEFUL FOR UNDERSTANDING OTHER ISSUES.
THE PROBLEM IS THE CENSUS LIKE OTHER STUDIES IS SUPPOSED TO BE
PRIVATE. LIKE NO ONE OUTSIDE THE CENSUS
BUREAU IS SUPPOSED TO BE ABLE TO LOOK AT JUST THE PUBLISHED
STATISTIC ABOUT THE DEMOGRAPHIC AND FIGURE OUT THERE’S A WHITE
MALE 31-YEAR-OLD WITH KIDS LIVING IN MY NEIGHBORHOOD.
THAT’S ME. THE CENSUS BUREAU IS SUPPOSED TO
KEEP MY INFORMATION CONFIDENTIAL AND THE INFORMATION OF EVERY
PERSON CONFIDENTIAL WHICH IS A TAL ORDER.
THE SHORT ANSWER, YOU CAN’T. THERE’S AN INHERENT TRADE-OFF
BETWEEN PUBLISHING SOMETHING YOU LEARN FROM A SURVEY AND
MAINTAINING THE PRIVACY OF PARTICIPANTS.
IT MAY SEEM LIKE YOU CAN JUST REMOVE PEOPLE’S NAMES FROM THE
SPREAD SHEETS AND PUBLIC AVERAGES AND TOTALS BUT IT’S
EASY TO RECONNECT USING POWERFUL S AND THERE’S A THEOREM THAT
INFORMATION RELEASED WILL INHERENTLY VIOLATE THE PRIVACY
OF PARTICIPANTS TO SOME DEGREE AND THE MORE INFORMATION THE
MORE YOU VIOLATE THE INDIVIDUAL PRIVACY AND HOW DO YOU QUANTIFY
LOSS OF PRIVACY AND PROTECT IT? TO UNDERSTAND HOW TO MEASURE
PRIVACY IT’S HELPFUL TO LOOK AT HOW SOMEONE WOULD PIECE TOGETHER
THE PRIVATE INFORMATION OF THE PEOPLE SURVEYED.
THEY COULD JUST TRY TO STEAL OR GAIN ACCESS TO THE PRIVATE
INFORMATION ITSELF WHICH CAN’T BE PROTECTED AGAINST
MATHEMATICALLY. THAT REQUIRES PHYSICAL DEFENSES
SO WE WON’T CONSIDER IT HERE. THE KIND WE CAN DEFEND AGAINST
MATHEMATICALLY LOOKS AT PUBLICLY PUBLISHED STATISTICS AND APPLIES
BRUTE FORCE COMPUTATIONAL POWER AND SEE WHICH ONES ARE THE MOST
PLAUSIBLE. WHICH FIT THE PUBLISHED
STATISTICS THE BEST. IMAGINE TRYING ALL LETTERS AND NUMBERS FOR
PASSWORDS UNTIL IT WORKS SO IMAGINE WHICH COMBINATIONS COME
CLOSEST TO AVERAGE AGE, RACIAL BREAKDOWN AND SO ON.
THE MORE CLOSELY IT MATCHES THE PUBLISHED FIGURES THE MORE IT
CAN MATCH. THE POORER IT MATCHES THE LOWER
LEVEL OF CERTAINTY. IF THERE’S SEVEN PEOPLE IN A
PARTICULAR AREA AND TELL ME FOUR ARE FEMALE, FOUR LIKE ICE CREAM
AND FOUR ARE MARRIED ADULTS AND THREE OF THE ICE CREAM LOVERS OF
FEMALE I CAN RECONSTRUCT THE AGE, SEX AND ICE CREAM
PREFERENCES OF EVERYONE INVOLVED.
I’D START WITH THE THREE ICE-CREAM LOVING FEMALE.
THOUGH THERE’S THOUSAND OF COMBINATIONS FOR THREE PEOPLE
ONLY A SMALL FRACTION, 36 IN FACT, ARE PLAUSIBLE.
THEY’RE IN THE RIGHT COMBINATION TO GIVE THE MEDIAN AGE AND THE
SAME THING WORKS FOR THE FOUR FEMALES OVERALL.
THERE’S ALMOST 100 MILLION POSSIBLE COMBINATIONS OF AGES
THEY COULD HAVE AND ONLY 24 AGE COMBINATIONS CONSISTENT WITH THE
MEDIAN MEDIAN AND ONE OF THE PLAUSIBLE
AGE COMBINATIONS WITH THE ICE CREAM LOVERS IT LEADS TO A
SINGLE PLAUSIBLE AND RECONSTRUCTION OF THE AGES,
SEXES AND ICE CREAM PREFERENCES OF THE PEOPLE INVOLVED.
A 100% VIOLATION OF PRIVACY. IF YOU DIDN’T LIST HOW MANY ICE
CREAM LOVERS WERE FEMALE, THERE’D BE TWO PLAUSIBLE
POSSIBILITIES. I’D BE LESS CERTAIN OF THE
COMBINATION OF AGES AND GENDERS AND ICE CREAM PREFERENCES AND IF
YOU PUBLISHED LESS DATA THERE’D BE A HIGHER NUMBER OF PLAUSIBLE OUTCOMES AND THERE’S
RESULTS OF THE STUDY. IF ALL POSSIBLE COMBINATIONS OF
AGES AND SEXES ARE SIMILARLY PLAUSIBLE, THEN AN ATTACKER
CAN’T DISTINGUISH WELL AND THE PRIVACY IS PROTECTED BUT IF
THEY’RE MOW PLAUSIBLE THAN THE REST THEY STAND OUT AND BECAUSE
THEY STAND OUT ON PLAUSIBILITY THEY’RE LIKELY TO BE CLOSER TO
THE TRUTH. TO PROTECT PRIVACY ALL
POSSIBILITIES NEED TO SEEM SIMILARLY PLAUSIBLE OR NOT TOO
CONSPICUOUS. IT’S QUANTIFIED BY MEASURES THE
MAXIMUM SLOPE OF THE GRAPH. IF THE SLOPE NEVER GETS TOO
STEEP YOU CAN’T HAVE PEAKS OF PLAUSIBLE POSSIBILITIES.
HOW DO WE MAXIMIZE THE SLOPE AND MINIMIZE THE PEAKS?
IN PRACTICE THE BEST WAY TO LIMIT AN ATTACKER’S ABILITY IS
TO RANDOMLY CHANGE THE PUBLISHED VALUE.
LIKE ROLLING A DIE AND ADD THAT TO THE PUBLISHED NUMBER AND
JITTERING PUTS A LIMIT ON THE SLOPE OF THE PLAUSIBILITY GRAPH
AND MAKES IT HARDER FOR A POSSIBILITY TO STAND OUT.
JITTERING RESULTS MAY SEEM LIKE LYING BUT AS LONG AS THE SIZE OF
THE ADJUSTMENT ISN’T BIG ENOUGH TO MAKE BIG ENOUGH CHANGES AND IF I JUST
TOLD YOU MY AGE OBVIOUSLY THERE’S ONE PLAUSIBLE
POSSIBILITY, 31. SUPPOSE I SECRETLY PULL A NUMBER
AND ADDED IT TO MY AGE. ALL YOU KNOW IS MY TRUE AGE IS
WITHIN THE FIVE YEARS OF THE NUMBER I TOLD YOU BUT YOU DON’T
KNOW MY AGE EXACTLY. MY PRIVACY HAS BEEN PRESERVED.
TO PROTECT MY AGE MORE I’D HAVE TO PULL A NUMBER BETWEEN MINUS
10 AND 10 OUT OF A HAT AND THAT INCREASE THE PLAUSIBLE
POSSIBLES. IT ALSO INCREASES YOUR
UNCERTAINTY ABOUT MY ACTUAL AGE. THE TRADE-OFF FOR PRIVACY IS
INACCURACY. IF I WANTED YOU TO KNOW MY AGE
WITHIN A YEAR I COULD ONLY PULL A NUMBER BETWEEN MINUS ONE AND
ONE. IN GENERAL, MORE PRIVACY MEANS
YOU GET LESS ACCURACY. LESS PRIVACY MEANS YOU CAN HAVE
MORE ACCURACY IF YOU DO THINGS RIGHT.
WHEN YOU PUBLISH RESULTS HOPEFULLY THERE’S A SWEET SPOT
TO SHARE SOMETHING USEFUL WHILE STILL SUFFICIENTLY MAINTAINING
SOMEONE’S PRIVACY AND MAINTAINING DECENT PRIVACY AND
ACCURACY GETS EASIER AND EASIER WITH LARGE DATA SETS LIKE AS I
ADD MORE NOISE TO THE IMAGE YOU GET THE GENERAL PICTURE EVEN
ONCE YOU LOST THE TRUE VALUE OF A PARTICULAR PIXEL SO WE CAN AND
SHOULD RANDOMLY JITTER PUBLIC STATISTICS WHICH THE U.S. CENSUS
HAS BEEN DOING SINCE THE ’70s. YOU CAN’T JUST ADD ‘OTHER RANDOM
NOISE. IF I SIMILE ADD RANDOM NOISE TO
THIS PICTURE, YOU CAN GET BACK THE ORIGINAL CLEAN IMAGE AND YOU
DON’T WANT THIS HAPPENING TO YOUR DATA.
THERE’S A FIELD OF COMPUTER SCIENCE DEDICATED TO FIGURING
OUT THE LEAST POSSIBLE OF NOISE TO GET THE MOST PRIVACY AND
ACCURACY AND TO FUTURE PROOF THE PUBLICATION OF DATA SO WHEN YOU
GET MULTIPLE JITTERED STATISTICS THEY CAN’T BE COMBINED IN A
CLEVER WAY TO CONSTRUCT PEOPLE’S DATA AND THE CENSUS BUREAU
COULDN’T PROMISE. THEY WERE JITTERING THE FIGURES
IN CENSUS TABLES AND CHARTS BUT NOT IN A MATHEMATIC WAY AND
COULDN’T PROMISE ANYTHING ABOUT PROTECTING PRIVACY UNTIL HOW.
2020 CENSUS WILL FOR THE FIRST TIME BE USING MATHEMATICALLY
VIGOROUS PRIVACY PROTECTION. IT RELIABLY COMPOUNDS OVER
MULTIPLE PIECES OF INFORMATION. IF WE PUBLISH AN AVERAGE AND
MEDIAN AGE WITH A PRIVACY LOSS FACTOR OF 3 AND HAVING BOTH
PIECES OF INFORMATION IS AT MOST 6.
THIS MEANS YOU CAN DECIDE ON THE TOTAL AMOUNT OF PRIVACY LOSS
YOU’RE WILLING TO SUFFER AND THEN DECIDE IF YOU WANT TO
RELEASE 10 PIECES OF INFORMATION AND LESS ACCURACY OR RELEASE ONE
PIECE OF INFORMATION WITH THE FULL PRIVACY LOSS AND HIGHER
LEVEL OF ACCURACY. HOW MUCH PRIVACY WE NEED IS A HARD QUESTION TO ANSWER.
FIRST IT INVOLVES WEIGH HOW MUCH AS A SOCIETY WE VALUE KNOWING
STUFF VERSUS THE DRAWBACKS OF RELEASING PRIVATE INFORMATION.
SECOND, SINCE THEY CAN BE MEASURED WE HAVE TO TRANSLATE
THE IDEAS OF PRIVACY AND ACCURACY LOSS AS SOMETHING
RELATABLE TO PEOPLE IN SOCIETY. FIRST AND FOREMOST, IT’S HARD TO PUBLIC DATA WITHOUT
VIOLATING THE INDIVIDUAL QUESTION AND IF YOU WANT A LIMIT
ON THE PRIVACY VIOLATION YOU HAVE TO RANDOMLY JITTER THE
STATISTICS TO PROTECT THE DATA. THE ACCURACY OF THE INFORMATION
AFTER BEING JITTERED IS GENERALLY DESCRIBED BY SAYING
SOMETHING LIKE IF WE RANDOMLY JITTER THE VIEW VALUE FOR THE
POPULATION OF THE TOWN 98% OF THE TIME IT WOULD BE WITHIN 10
PEOPLE OF THE VIEW VALUE. THERE’S HOW CLOSE YOU WANT IT TO
THE ANSWER AND THE LOSS OF PRIVACY IS DESCRIBED IN TERMS OF
HOW CONFIDENTLY AN ATTACKER WOULD BE ABLE TO SINGLE OUT A
POSSIBILITY FOR THE TRUE DATA. GIVEN THE PUBLISHED INFORMATION
IS THERE A FEW POSSIBILITIES OR MANY PLAUSIBLE FOR THE WHAT THE
TRUE DATA WOULD BE. IT’S MEASURED BY THE PROMINENCE
OF PEAKS AND THE PROTECTION OF PRIVACY REQUIRES POLICING THE
POSE BUILT FOR SUCH PEAKS. IF INDIVIDUALS WILL WILLINGLY
PARTICIPATE IN STUDIES OR SURVEYS OR USE SERVICES WHERE WE
REVEAL SENSITIVE INFORMATION, WE SHOULD DEMAND THE RESEARCHERS OR
ORGANIZATIONS UTILIZE A ROBUST WAY OF USING PRIVACY.
IF THEY CAN’T GUARANTEE THERE WON’T BE A PEAK IN PLAUSIBILITY
WE SHOULDN’T AGREE TO GIVE THEM A PEEK AT OUR DATA.
THANKS TO THE CENSUS BUREAU FOR SUPPORT THIS VIDEO AND THE
FOUNDERS KNOW A COMPLETE COUNT IS NECESSARY FOR A FAIR REPRESENTATION AND THESE NEW
SAFEGUARDS WILL PROTECT CONFIDENTIALITY WHILE ALLOWING
THE CENSUS BUREAU TO DELIVER THE COMPLETE AND ACCURATE COUNT OF
THE NATION’S POPULATION AND GIVE THOSE WHO RELY ON DATA INCREASED
CLARITY ON ANALYSES AND DECISION MAKING.
THE CENSUS BUREAU VIEWS THE ADOPTION OF A MATHEMATICAL
GUARANTEE OF PRIVACY AN AS WIN-WIN.
PREVENTING CENSUS BUREAU HAVE HACKING IS A DIFFERENT QUESTION.
THERE’S A REAL CHOICE EVERY CURATOR HAS TO MAKE.
IF THEY WANT THE RESPONDENTS TO TRUST THEM TO PROTECT
CONFIDENTIALITY THEY HAVE TO PROVIDE PROVABLE GUARANTEES.
THERE’S A CONSTRAINT ON THE AMOUNT OF INFORMATION PUBLISHED
FROM DATA. IT’S MATHEMATICALLY IMPOSSIBLE
TO PROVIDE ACCURATE ANSWERS FROM STATISTICS WHILE PROTECTING THE
PRIVACY OF RESPONDENTS. THEY NEED TO DO TWO THINGS,
UNDER THE NEEDS AND DESIRES OF THE PEOPLE WHO PROVIDE DATA AND
WANT TO USE DATA TO DETERMINE WHAT BALANCE OF ACCURACY TO
PRIVACY TO CHOOSE AND NOT WASTE THE LIMITED PRIVACY BUDGET BY
PROVIDING INACCURATE ANSWERS TO UNIMPORTANT QUESTIONS.
>>IF YOU’RE NOT FAMILIAR WITH METAPHYSICS I INVITE YOU TO
LISTEN TO HENRY REICH’S EXPLANATION OF SPACE TIME
CONTINUUM AND MANY OTHER CON CEMENTS IN FIZZ.
— CONCEPTS IN PHYSICS AND WAS AN ACADEMIC PHYSICIST UNTIL
DOING THINGS LIKE THIS AND THE CENSUS BECAME HIS FULL-TIME
OCCUPATION. WE WANTED TO SHOW YOU THAT AND
IT’S BRAND NEW AND JUST RELEASED YESTERDAY.
IT’S ONE OF OUR ATTEMPTS TO MAKE CLEAR THE DECISION MAKING
FRAMEWORK UNDERLYING THE CONFIDENTIALITY AND THE REASONS
WHY WE HAVE MOVED TO A MODERNIZED SYSTEM WHICH WE
DISCUSSED BEFORE. I’LL GIVE THE/ — DISCLOSURE
AVOIDANCE UPDATE AND THIS IS WHERE WE’RE CURATING THE CURRENT
CODE BASE THAT WAS USED FOR THE 18 END TO END TEST.
THE PLACE WHERE THE MEMOS UNDERLYING THE CODE BASE HAVE BEEN MEMORIALIZED.
I ALSO WANTED TO BRIEFLY SAY BECAUSE WE SAID IT IN SEVERAL
FORUMS THE SURVEY WILL NOT BE IMPLEMENTED BEFORE 2025.
RON HAS WRITTEN A NICE BLOG EXPLAINING THAT WE HAVE OUR
HANDS FULL WITH THE 2020 CENSUS AND IN THE PROCESS OF MOVING
TOWARDS FORMAL PRIVACY FOR THE AMERICAN COMMUNITY SURVEY, YOU
CAN’T EXPECT THE SAME ENGAGEMENT OF THE USER COMMUNITY THAT THE
WONDERFUL DOCUMENTATION ALLISON WAS PRAISING EARLIER PRODUCED.
THERE WAS NEVER ANY INTENTION TO DO THIS WITH A HAMMER BLUDGEON AND TO TELL THE USER
COMMUNITY WE WILL BE DEVELOPING THE COLLABORATIVE WAY IN WHICH
WE’RE OPENED ABOUT OUR MODERNIZATIONS OF THE AMERICAN
COMMUNITY SURVEY OVER THE COURSE OF THE NEXT SEVERAL YEARS AND
NOTHING WILL BE IMPLEMENTED BEFORE 2025 THAT INVOLVES FORMAL
PRIVACY. IT DOESN’T MEAN WE’LL BE USING
THE TRADITIONAL METHODS. THAT’S WHAT WE’LL CONTINUE AND
THEY’RE NOT PARTICULARLY TRANSPARENT.
THAT’S THE TRADE-OFF.
SO THE END TO END TEST PRODUCT USED A COLLECTION OF QUERIES
THAT HAD WHAT COMPUTER SCIENTISTS CALLED A HISTOGRAM
AND THERE’S A CONTINGENCY TABLE OF 2012 CELLS AT THE NATIONAL
LEVEL. EVERY INTERACTION OF THE
VARIABLES USED FOR THE REDISTRICTING DATA.
ALL POSSIBLE COMBINATIONS ARE IN THE 2012 NATIONAL HISTOGRAM. IN ORDER TO EXPAND THAT AT THE
DEMONSTRATION LEVEL, THAT FULLY SATURATED HISTOGRAM AND I SEE AN ERROR IN THE SLIDE.
IT’S THE HISTOGRAM THAT WAS DISCUSSED FOR THE DEMONSTRATION
PRODUCT. THAT AGGREGATION WAS NOT IN FACT
IMPLEMENTED. THAT’S THE WAY OUT.
HAVE A HOUSEHOLD AND ONE OF TYPES AND RAISE THE OMB
CATEGORIES IN THE PL94 DATA AND HISPANIC ORIGIN IS HISPANIC OR
LATINO ORIGIN, YES OR NO AND SEX IS MALE OR FEMALE AND
CITIZENSHIP IS IN THE PROTOTYPE THAT WAS USED FOR THOSE PRODUCTS
BUT NO CITIZENSHIP TABLES WILL BE RELEASED AS PART OF THE
DEMONSTRATION. WE WANTED TO MAKE SURE THE
SYSTEM COULD HANDLE THE CVAP PRODUCT.
FOR THE HOUSING AND HOUSEHOLDS, THE FULLY SATURATED HISTOGRAM IT CONTAINS
HOUSEHOLDER ATTRIBUTES, RACE, ETHNICITY.
IT’S ALWAYS OMB RATES. RACE, HISPANIC ORIGIN, SEX AND
AGE AND THE AGE IN THE HOUSEHOLD LEVEL TABLES HAS BEEN GROUPED
ACCORDING TO THE SCHEME JASON PRESENTED.
MULTIGENERATIONAL STATUS AND HOUSEHOLD SIZE TOP CODED AT 7.
AND THE PRESENCE OF PEOPLE 60, 65 OR 75 OR OVER IN THE
HOUSEHOLD AND CERTAIN HOUSEHOLD TYPE ATTRIBUTES IN THE SPECKS JASON TALKED ABOUT.S JASON
TALKED ABOUT. THE CURRENT VERSION OF THE DISCLOSURE AVOIDANCE SYSTEM CAN
HANDLE THE TWO SETUPS AND THEY’RE SUFFICIENT TO PRODUCE
AVERAGES WHICH ARE NOT PARTICULARLY USEFUL BUT ABOUT
70% OF THE FULL DHC PRODUCT. SPECS ARE WHAT WILL BE RELEASED
FROM THE RUN ON THE 2010 DATA WILL BE DEVELOPED OVER THE
COURSE OF THE NEXT MONTH BECAUSE AS SOON AS THAT SYSTEM STOPS
RUNNING, IT WILL BE HANDED OVER TO DECENNIAL AND DEMO TO DO THE QUALITY CONTROL AND
BASED ON THE EDITED FILE AND COVER ABOUT 20% AND BASED ON A
WORK LOAD OF 400,000 OR SO HISTOGRAM AND FULLY SATURATED
CONTINGENCY TABLE AND USE A NEW TOP-DOWN TO BE RELEASED IN THE
DATA MODE IN THE SAME REPOSITORY ONCE WE FINISHED THE PRODUCTS.
THE DETAILS OF THE OPTIMIZATION MANAGEMENT BUDGET WAS YESTERDAY
AND IT HAS NOT BEEN RELEASED TO THE PUBLIC AND WE’RE DEVELOPING
A COMMUNICATION CAMPAIGN AS THE QUALITY REVIEW GOES. I WON’T TELL YOU THE PORTION BUT
WE DID TWO HOURS WORTH OF HOW DO YOU DO IT AND WE
WRESTLED WITH IT. WE HAVE A SOFT RELEASE DATE OF
OCTOBER AND MORE DETAILS IN DECEMBER.
I DON’T WANT TO GIVE THE IMPRESSION THAT WORKSHOP IS
EXCLUSIVELY FOR ANY PARTICULAR USER COMMUNITY.
IT’S ALSO FOR THE PRIVACY COMMUNITY WE WANT ALL INPUT AND
THE CODE BASE RELEASE WILL FOLLOW IN OCTOBER.
YOU ALSO ASKED FOR AN UPDATE ON ADMINISTRATIVE DATA.
THE BULLETS STATE APPORTIONMENT FOR HOW IT WILL BE DONE AND ONE
OF THE DATA PRODUCTS FROM THE 2020 CENSUS.
THE APPORTIONMENT COUNTS ARE DUE TO THE SECRETARY WHO DELIVERS
THEM TO THE PRESIDENT BY SEPTEMBER 31 BY 2020.
IT’S A STATUTORY DEADLINE. THE COUNTS ARE USED TO
REAPPORTION THE HOUSE OF REPRESENTATIVES INCLUDING THE
STATE’S TOTAL POPULATION, CITIZENS AND NON-CITIZENS AND
ACCOUNTS OF EMPLOYEES AND THOSE LIVING WITH THEM ALLOCATED TO
THEIR HOME STATE AND THE PROCESS OF DETERMINE WHO WILL COUNTS AND SAY — IS A RESIDENT AND THE APPORTIONMENT
COUNTS ARE CALCULATED FROM AN INTERNAL FILE KNOWN AS THE
CENSUS UNEDITED FILE. WHICH IS BASICALLY THE RAW
CENSUS RESPONSES AND THEN UNDUPLICATED.
AT THE PERSON LEVEL THE RECORD COUNT IS FIXED AND THE GEO
IDENTIFIER THE STATE IN WHICH YOU LIVE ARE FIXED.
THEY CAN’T BE FURTHER EDIT HEAD TO COUNT OF PERSONS WITHIN EACH
STATE IS FIXED THAT THE POINT. THERE’S NO CITIZENSHIP DATA.
REDISTRICTING. PL94171, THERE’S BEEN MULTIPLE
FEDERAL REGISTER NOTICES STATING THE CONTENT OF THAT FILE.
IT’S PRODUCED BY A STATUTORY DEADLINE OF APRIL 1, 2021.
THE FLAP IS TO MAKE THESE DATA AVAILABLE TO THE PUBLIC BUT
SPECIFICALLY TO STATES TO ASSIST IN REDISTRICTING.
THESE DATA ARE COMPUTED FROM THE CENSUS FILE PRODUCED FROM THE
CEF BUT NOT TO BE USED UNTIL LATE JANUARY 2021.
AT THAT POINT, THE CEF IS SENT TO THE DISCLOSURE AVOIDANCE
SYSTEM WE’VE BEEN TALKING ABOUT AND THAT DISCLOSURE AVOIDANCE
SYSTEM RELEASES A MICRODATA DETAIL FILE TO THE TABULATION
SYSTEM. THE TABULATION SYSTEM THEN BEGINS COMPUTING.
THE PRODUCTION PL9171 DATA AND TE DATA ARE
RELEASED ON THE FLOW BASIS RUNNING THROUGH THE MARCH 31,
2021 STATUTORY DEADLINE.4171 DATA AND TE DATA
ARE RELEASED ON THE FLOW BASIS RUNNING THROUGH THE MARCH 31,
2021 STATUTORY DEADLINE. THE FORMAT OF THE DATA WAS DONE
EARLY AND THE TABLES ARE EXACTLY THE SAME FORMAT AS THE END TO
END TEST PRODUCT. THERE’S FOUR POPULATION TABLES
AND ONE GROUP QUARTERS AND ONE HOUSING TABLE.
THEY HAVE WITHIN REVIEWED BY STATE REDISTRICTING OFFICES AND
THE CONTINUOUS REVIEW PROCESS THAT’S PUBLIC AND HAS FEDERAL
REGISTRY NOTICED ASSOCIATED WITH IT AND THAT’S THE FORMAT IN
WHICH THE STATE REDISTRICTING OFFICES ARE EXPECTED TO HAVE THE
DATA DELIVERED OF THE PL94171. THE VOTING AGE POPULATION DATA.
THE CLEARANCE PACKAGE FOR THE 2020 CENSUS AND PRESIDENT’S
EXECUTIVE ORDER, 13880 COMMITS TO RELEASING DATA BY MARCH 21,
2021. THE SAME AS THE DEADLINE FOR THE
REDISTRICTING DATA. IT WILL BE PRODUCED BY COMBINING
ADMINISTRATIVE DATA FROM A NUMBER OF FEDERAL AND POSSIBLY
STATE AGENCIES INTO A SEPARATE MICRODATA FILE TO CONTAIN A BEST
CITIZENSHIP VARIABLE FOR EVERY PERSON IN THE 2020 CENSUS.
THE FILE AND CEF WILL BE SIMULTANEOUSLY SENT TO THE
DISCLOSURE AVOIDANCE TOMORROW TO PLACE A CITIZENSHIP VARIABLE ON
THE SAME KREM. IT MEANS ALL THE CODES HAVE BEEN
HARMONIZED. THE DATA WILL BE PRODUCED AT THE
BLOCK LEVEL AND REDUCED TO THE PUBLIC BY MARCH 1, 2021.
WHAT’S THE FORMAT FOR THE CVAP DATA.
NO DECISIONS HAVE BEEN MADE WITH THE FORMALITY OF THE DATA.
TO AVOID MISINTERPRETATION, IT WILL USE THE SAME CODES AS ALL
THE OTHER PUBLICATIONS. SO THE BLOCK GEOCODES THE SAME
AS THE PL94 AND DHC.
NO DECISIONS HAVE BEEN MADE REGARDING THE CVAP DATA SINCE 2011.
THE CENSUS BUREAU’S INTERNAL WORKING GROUP SET MARCH 31, 2020
AS THE FINAL DATE FOR DETERMINING THE VIABLE OF
SOURCES ON CITIZENSHIP. THAT MEANS IF WE CAN’T DETERMINE
IT’S USEFUL BY MARCH 31 OF NEXT YEAR, WE WON’T USE THAT
ADMINISTRATIVE RECORD SOURCE. A SIMILAR ZIG WAS MADE FOR THE
— DECISION WAS MADE FOR THE 2020 CENSUS LAST YEAR.
MARCH 31, 2020 IS ALSO BEING USED BY THE INTERNAL TEAM AS THE
DATE FOR SPECIFYING WITH A CVAP PRODUCT WOULD LOOK LIKE THOUGH
IT WON’T BE RELEASED. AT SOME POINT YOU HAVE TO SAY
WHAT YOU’RE GOING TO TABULATE AND HOW YOU’LL DO IT SO THAT’S
THE INTERNAL DEADLINE FOR THAT.
WE’RE ALSO CONSIDERING PRODUCTS LIKE THE ONES WE TALKED ABOUT
FOR THE OTHER DATA PRODUCTS.
SO USING ADMINISTRATIVE DATA. THE SECRETARY’S MARCH 26, 2018
INSTRUCTIONS WE BEGAN DOING MODELING EFFORTS THAT FOCUSSED
ON HAVING AVAILABLE TWO SOURCE OF DATA FOR CITIZENSHIP.
A SURVEY RESPONSE TO THE QUESTION THAT WAS PROPOSED FOR
THE 20 CENSUS AND ADMINISTRATIVE RECORDS.
ONCE THE SUPREME COURT HELD THE IN JUNCTION AND THE PRESIDENT
ISSUED THE EXECUTIVE ORDER 13880 WE HAD TO SHIFT TO USE JUST ADMINISTRATIVE RECORDS.
IT ISN’T A QUESTION OF WHETHER IT’S FEASIBLE BUT HOW TO ADAPT
THE MODELING EFFORTS. THAT’S WHAT’S NEVER — UNDERWAY NOW.
THERE’S FEDERAL AGENCIES THAT HAVE ACCESS TO PERSONAL LEVEL
DATA RELEVANT TO ESTIMATING CITIZENSHIP.
THERE’S TWO PRIMARY USES. ONE IS KEEPING THE NAMES,
ADDRESSES AND OTHER INFORMATION CURRENT AND DETERMINING
CITIZENSHIP FILE FROM ELIGIBILITY CONDITIONS THAT
CAUSED YOU TO BE ON THE FILE.
HERE’S THE CURRENT SOURCES OF CITIZENSHIP. FOR 94% OF THE
UNIVERSE BUT IT’S NOT THE ENTIRE POPULATION OF THE UNITED STATES.
INDIVIDUAL TAXPAYER NUMBERS. THE BUREAU HAS NOT RECEIVED OR
REQUESTED FROM THE IRS THE APPLICATION FORM AND OTHER DATA
ASSOCIATED WITH AN I-10 BUT HAS A NUMERICAL FORMAT WHICH MAKES
IT POSSIBLE TO DETERMINE ON ANOTHER ADMINISTRATIVE RECORD
THE RECORD WAS FILED WITH AN I-10 RATHER THAN A SOCIAL
SECURITY NUMBER AND THEY’RE ONLY ISSUED TO NON- CITIZENS.
THAT’S AN INFERENCE ON A NON-CITIZEN THAT COMES FROM
KNOWING THE IDENTIFIER IS AN I-10.
IRS FORMS 1040 AND 1099 WHICH ARE PRIMARILY USED TO
KEEP THE INFORMATION CURRENT AND MEDICARE AND MEDICAID DATA ARE
PRIMARILY USED TO KEEP THE SYSTEM CURRENT.
HOUSING AND URBAN DEVELOPMENT DATA FROM THE AGENCIES I’VE LISTED THERE.
THAT’S THE LAST ONE ON THE LIST.
THOSE ARE VARIOUS DATA FILES FROM HOUSING AND URBAN
DEVELOPMENT. WE’VE HAD MOUs IN PLACE TO
ACQUIRE THESE ADMINISTRATIVE RECORDS FOR QUITE SOME TIME.
THEY’RE USED IN OTHER PRODUCTS. THEY HAVE A FIXED LENGTH AND
HAVE TO BE CONTINUOUSLY SERVICED.
THE ADDITIONAL SOURCES COME FROM THE DEPARTMENT OF HOMELAND
SECURITY FROM THREE AGENCIES FROM WITHIN THE USCIS AND CPB
AND I.C.E. WHERE WE GET LAWFUL PERMANENT CITIZENS AND VISAS AND
ARRIVAL INFORMATION AND FROM THE DEPARTMENT OF STATE WE GET
PASSPORT DATA FROM THE SOCIAL SECURITY ADMINISTRATION WHERE WE
REQUESTED THE MASTER BENEFIT RECORD AND WE REQUESTED PATIENT
REGISTRATION AND FROM THE DEPARTMENT OF JUSTICE AND U.S.
MARSHALS AND CITIZENSHIP AND IMMIGRATION DATA.
THOSE ARE THE SOURCES WE’RE SEEKING TO AUGMENT OUR SYSTEM
WITH. WHAT’S THE RESEARCH PROGRAM LOOK LIKE?
WE HAVE DEVELOPED STATISTICAL MODELS TO COMBINE CITIZENSHIP
DATA TO ESTIMATE BEST CITIZENSHIP FOR EACH PERSON
KNOWN AS THE PERSONAL VALIDATION SYSTEM.
THE PRODUCTION RECORD LYNCHAGE SYSTEM.
THE SAME THE 2020 CENSUS USES TO INJEST ADMINISTRATIVE RECORDS
FOR THE OTHER PURPOSE IT’S USING THEM FOR.
AND USED TO BE COMBINED WITH THE 2020 INFORMATION EVIDENCE FILE
TO PROVIDE THE DISCLOSURE AVOIDANCE SYSTEM TO BE USED
AFTER IT’S BEEN CONFIDENTIALITY PROTECTED.
THIS RESEARCH BEGAN IN APRIL OF 2018.
FINAL SPECIFICATIONS AND MODELS ARE PLANNED TO BE RELEASE BEFORE
MARCH 1, 2020. IT’S IN TERNAL FOR MAPPING SERVICES.
YOU HAVE TO HAVE SOME POINT IN WHICH YOU FIX THE SPECS.
THE REST OF THE RESEARCH PROGRAM IN ENHANCES THE CAPABILITY.
THE PBS CAN LINK ADMINISTRATIVE DATA FOR ABOUT 90% OF THE U.S. RESIDENT POPULATION.
ANOTHER 10% HAVE ADMINISTRATIVE RECORDS AND WOULD BE LINKABLE IF
THE SYSTEM ITSELF HAD THE PII FOR THE RECORD LINKAGE.
MANY REQUESTS FILES ARE PRIMARILY TO EXPAND THE
CAPABILITY TO IDENTIFY THE CITIZEN STATUS OF THE REMAINING
10% OF THE U.S. POPULATION.
ALL DATA INCLUDING WHAT WAS JUST ARE PROTECTED AND CAN ONLY BE
USED AND COVERED BY SECTION 9 WHICH PROHIBIT MAKING THE DATA
FOR AN ESTABLISHMENT OR INDIVIDUAL CAN BE IDENTIFIED.
WE’RE NOT ALLOWED TO PUBLISH THE DATA AND WE’RE BEEN DISCUSSING
THE LAST 45 MINUTES OUR EFFORTS TO ENSURE THE 2020 PUBLICATIONS
INCLUDING ANN CVAP PUBLICATION SATISFY TITLE 13.
THE TABLES WILL BE PRODUCED USING THE AVOIDANCE SYSTEM USING
TOP-DOWN. THE CVAP TABLES WILL SHARE THE
BUDGET DETERMINED BY THE EXECUTIVE POLICY COMMITTEE FOR
THE 2020 CENSUS PUBLICATIONS. WE’LL ACCOUNT FOR THE PRIVACY
LOSS ASSOCIATE WITH THE TABLES AND OVERALL ACCOUNTING OF THE
2020 CENSUS PUBLICATIONS. OKAY.
THANKS.>>OKAY. ANY QUESTIONS.
WE CAN START WITH WASN’T THAT A FANTASTIC VIDEO?
YEAH. I THINK KATHLEEN SENT AN E-MAIL
TO SAY IT WAS FANTASTIC AND WANTED TO CHIME IN ON THAT AS WELL.
KUDOS TO KUNAL WHO FOUND IT LAST NIGHT SO WE’RE EXCITE ABOUT THAT.
>>ANDREW STANWICK. THANK YOU FOR THE PRESENTATION.
I’LL ASK A NAIVE PRESENTATION ABOUT WHAT WAS ON THE LAST
SLIDE. I THINK I WOULD BENEFIT FROM A
BETTER INFORMATION OF WHAT IS MEANT BY TITLE 13 SECTION 9 IN
THINKING ABOUT IT FOR THE DATA I’LL SUBMIT TO THE CENSUS.
I WOULD WAGER MY FAMILY BECAUSE OF ITS CONFIGURATION OF GENDERS
AND AGES AND THE BLOCK GROUP WHERE IT LIVES IS PROBABLY
IDENTIFIABLE IF THE RECORD WERE EVER SEEN AND WOULD KNOW THINGS
ABOUT US YOU DIDN’T NEED TO KNOW IDENTIFY US, MAYBE HISPANIC
ORIGIN. I’M WONDERING, DOES THAT MEAN
THAT YOU CAN’T RELEASE THE INFORMATION OR IT CAN’T BE
DISCOVERABLE EVEN IF IT’S NOT SPECIFICALLY ATTACHED TO ME?
YOU SIMPLY COULDN’TFIDE AN INDIVIDUAL LINE OR HOUSEHOLDS RECORD?
YOU’RE NOT RELEASING MY NAME OR ADDRESS BUT SOMEBODY DELIVER
COULD FIGURE THAT OUT AND LEARN THINGS ABOUT US.
I NEED MORE BACKGROUND ON WHAT THE IMPORT OF THAT TITLE AND SECTION IS.
>>SO SECTION 9 OF TITLE 13 HAS BEEN CONSISTENTLY INTERPRETED BY
THE CENSUS BUREAU AS PROHIBITING THE RELEASE OF DATA EVEN IN
TABULAR FORMAT WHERE YOU CAN ATTACH THE NAME TO THE RECORD.
AND SEEING IT WAS ONLY A SINGLE ENTITY IN A PARTICULAR CELL OF A
TABULAR SUMMARY AND SAYING WELL, IF YOU DO THIS IT DOESN’T TAKE
MUCH BACKGROUND INFORMATION TO FIGURE OUT WHAT THAT ENTITY IS.
SO AT THE BLOCK LEVEL, THE AVERAGE BLOCK HAD IN 2010 HAD 30 PERSONS.
THAT MEANS BLOCK LEVEL POPULATIONS STATISTICS ARE
ALMOST ENTIRELY DEALING WITH POPULATION UNIQUES.
SO WE HAVEN’T RELEASED STATISTICS LIKE THAT BUT HAVE
SAID BLOCK, AGE, SINGLE YEARS AND ETHNICITY IS A POPULATION
FOR MORE THAN HALF THE POPULATION AND ANY ADDITIONAL
ATTRIBUTE WILL INCREASE THAT AND VIRTUALLY EVERY HOUSEHOLD IN THE
UNITED STATES IS A POPULATION UNIQUE AT THE BLOCK LEVEL.
PUBLISHING BLOCK LEVEL DATA IS EXTREMELY TRICKY.
THE TECHNIQUES THAT HAVE BEEN USED HISTORICALLY TO PROTECT
CONFIDENTIALITY GOT THEIR UNCERTAINTY PRIMARILY FROM ME
NOT BEING ALLOWED TO TELL YOU EXACTLY WHAT WE DID AND STATING
IN THE TECHNICAL DOCUMENTATION WE’VE DONE GEOGRAPHIC SWAPPING
TO PROTECT THE GEOSPATIAL LOCATION AND MAY HAVE DONE OTHER
THINGS. WE ALSO GIVE SOME CHARACTERISTIC
OF THE GEOGRAPHIC SWAPPING, FOR EXAMPLE, IN 2010 IT DID NOT
CHANGE THE POPULATION TOTALS IN A BLOCK.
WE HAVE DISCOVERED THROUGH OUR RESEARCH, FAILURE TO PROVIDE
GEOSPATIAL LOCATION TO THE BLOCK POPULATIONS IS A SERIOUS
COMPROMISE OF THE CONFIDENTIALITY OF THE DATA.
WE HAVE INCLUDED THE BLOCK LEVEL AGREO SPACIAL LOCATION BEING
PROTECTED BY THE FORMAL PRIVACY SYSTEM.
THAT MEANS TO USE METAPHYSICS LANGUAGE, EVERY ONE OF YOUR
CHARACTERISTICS AND IN YOUR FAMILY’S CHARACTERISTICS HAVE
BEEN JITTERED INCLUDING WHERE YOU WERE GEOSPATIALLY LOCATED IN
OUR DATABASE. WE DON’T DO IT RECORD BY RECORD
BUT THROUGH AN ALGORITHM BUT GETS DOWN TO EVERY SINGLE
CHARACTERISTIC GETS JITTERED AND ENOUGH TO AVOID REIDENTIFICATION
ATTACKS. THE PROBLEM IS OUR OLD METHOD OF
DOING IT WAS TO TRY TO SPECIFY EXACTLY WHAT THE ATTACKER WAS
GOING USE, HOW MUCH COMPUTING RESOURCES THE ATTACKER HAD AND
WHAT SORT OF OUTSIDE KNOWLEDGE HE OR SHE BROUGHT TO THE
TOMORROW. YOU SPECIFIED A SPECIAL DATABASE AND THEY HAD
SOPHISTICATED LINKAGE TECHNOLOGY THAT COULD MAP DATA WITH BLOCK
I.D. THAT CONTAINS A NAME AND ADDRESS.
ON THAT TECHNOLOGY WE DISCOVERED IN 2010 IF YOU HAD DONE THAT YOU
WOULD HAVE BEEN ABLE TO LINK ABOUT 45% OF THE POPULATION AND
YOU WOULD HAVE BEEN RIGHT WHO THEY WERE 38% OF THE TIME.
THAT WAS JUDGED TO BE AN UNACCEPTABLE CONDITION FOR 2020.
WE HAD TO ABIDE BY BETTER CONFIDENTIALITY PROTECTION.
USING FORMAL PRIVACY YOU’RE TRYING TO CONTROL
NOT THE ABSOLUTE REIDENTIFICATION RATE BECAUSE IT
DEPENDS ON THINGS OUT OF OUR CONTROL BUT THE CONTRIBUTION THE
DATA WE RELEASE MAKE TO THAT REIDENTIFICATION.
SO THE DATA SCIENCE EXERCISE. SO WHEN YOU TURN THE DATA
SCIENTIST LOOSE ON THE 2020 TABLES THEIR INFERENCES ABOUT
SPECIFIC VARIABLES WILL BE CONTROLLED BY THAT EPSILON WE
SPECIFIED FOR THE FINAL PRODUCTS.
THAT’S A WORSE CASE. SO BASICALLY IF YOU’RE
CONTROLLING THE WORSE CASE, YOU’RE CONTROLLING ALL THE OTHER
CASES. SO WORSE CASE PROTECTION IS WHAT THE
SYSTEMS ARE BASED ON AND THEY’RE HARD TO BUILD AND HARD TO REASON
OUTSIDE THE MATHEMATICAL COMMUNITY BUT THEY
WORK. AND MAINTAIN THE CONFIDENCE OF
PEOPLE THAT THIS DOES WORK AND IT’S IMPORTANT IT WORK AND IT’S
IMPORTANT PEOPLE BELIEVE US WHEN WE SAY IT WORKS.
>>I WANT TO COMPLIMENT YOU ON THE RELEASE OF THE CODE BASE
IT’S EXCELLENT PRACTICE TO MAKE PUBLIC WHAT YOU’RE DOING AND
RELEASING THE CODE BASE IS CLEARLY THE RIGHT WAY TO DO IT.
I HAVE A COUPLE QUESTIONS. FIRST, YOU MENTIONED YOU’RE
GOING IT RELEASE THE 2010 DATA AND THE SYSTEM OF THE
ORIGINAL DATA OTHER THAN THE NEW ONE.
AND TRYING TO DIE COMPARISON BETWEEN THE OLD SYSTEM COMPARED TO THE NEW ONE.
AND THE NEW ONE ALLOWS TO YOU DO REASONING ON TOP OF IT.
I GUESS THE CITIZENSHIP DATA PEOPLE CONSIDER MORE SENSITIVE
THAN OTHER ATTRIBUTES. AND ALSO PERHAPS MORE CORRELATED
ACROSS BLOCKS THAN SOME OF THE OTHER ATTRIBUTES.
IF YOU LOOK AT THE DISTRIBUTION OF SECTION OF
NON-CITIZENS IN THE BLOCK IS PROBABLY MORE BIMODAL AND
MAKE IT EASIER TO MOUNT TO RECONSTRUCT THE ATTRIBUTE THAN
FOR A COMPLETELY INDEPENDENT ATTRIBUTE.
ALL OF WHICH TO SAY IT MAKES SENSE TO HAVE A HIGHER LEVEL OF
PRIVACY GUARANTEE FOR THIS ATTRIBUTE COMPARED TO OTHER.
— OTHERS. I WONDER IF YOU HAVE THOUGHT
ABOUT THAT. DOING THAT MAY NOT BE THAT HARD
WITH A RANDOMIZED RESPONSE BEFORE FEEDING IT TO THE
DISCLOSURE AVOIDANCE SYSTEM.>>WE INVESTIGATED THE
FEASIBILITY OF PUTTING EXTRA CONFIDENTIALITY PROTECTIONS ON
THE CITIZENSHIP VARIABLE. THAT’S WHAT WE’LL LOOK AT.
>>GREAT MINDS THINK ALIVE.>>THIS GOES BEYOND WHAT’S HERE
BUT THERE’S A VERY INTERESTING IMPLICATIONS ON YOUR SLIDE 54.
A BIG LIL — LIMITATION OF THE CENSUS BUREAU SYSTEM IS IT ONLY
WORKS FOR PEOPLE WHO HAVE FSN AND YOU COULDN’T GET A PICK IF
YOU WORKED IN THAT DATABASE.BUREAU SYSTEM IS
IT ONLY WORKS FOR PEOPLE WHO HAVE FSN AND YOU COULDN’T GET A
PICK IF YOU WORKED IN THAT DATABASE.
THEY SURVEYS TYPICALLY LEAVE 5% TO 10% UNMATCH AND AND HERE
YOU’RE CREATING LINKAGE BEYOND THE SYSTEM.
IS THE BUREAU CONTEMPLATING CREATING PICKS FOR THESE FOLKS
TO EXPAND THE LINKAGE IN ALL FUTURE SURVEYS?>>WE ACTUALLY ALREADY DO THAT
AND CAN GET A PICK WITH A SOCIAL SECURITY NUMBER.
EVERY PERSON AND RECORD WE INGEST GETS PUT THROUGH AN
ANONYMIZATION ROUTINE THAT REMOVES THE NAME AND ADDRESS AND
REPLACES THEM WITH CODES.
ONE OF THE BIPARTISAN BYPRODUCTS OF THE RESEARCH IS TO
REMOVE THE PRODUCTION ENVIRONMENT.
THAT’S HOW WE DO IT. THE CITIZENSHIP VARIABLE WILL
BENEFIT FROM THIS RESEARCH THE OTHER PRODUCTION
USES OF THE 2020 CENSUS WILL RUN ON THE PRODUCTION CODE NOT
LOCKED BUT IN A PRODUCTION PROCESS.
IT’S TO EXPAND BEYOND PEOPLE ISSUED SOCIAL SECURITY NUMBERS.
>>THAT’S GREAT.>>IT SEEMS LIKE A GOOD TIME TO
REMIND EVERYBODY IT’S A ONE-WAY STREET.
YOU GUYS INGEST THE DATA BUT BY LAW THERE’S NO SHARING IT ANY OTHER WAY.
>>THE CENSUS BUREAU IS THE DATA ROACH HOTEL.
THEY COME IN IDENTIFIABLE AND COME OUT UNIDENTIFIABLE.
>>KRISHNA.>>MAYBE THIS MIGHT HAVE BEEN A
BETTER THE LAST TWO POSITIONS HAVE FAIRLY LINKED.
IT WAS INTERESTING. IN THINKING ABOUT ALL OF THIS
CONTRAST SEEMS SUPER IMPORTANT MORE SO IN LIGHT OF THESE
CONVERSATIONS ABOUT CITIZENSHIP. PART OF THIS IS MAYBE MADE ME
THINK ABOUT HOW PEOPLE THINK ABOUT CONFIDENTIALITY AND HOW IT
MAPS AND THIS IS RELATED TO ANDREW’S QUESTION AND HOW IT
MAPS TO HOW YOU GUYS ARE THINKING ABOUT IT INTERNALLY.
I DON’T — WE HAVE A NICE SET OF RESULTS WITH TAB LARRY —
TABULARRY INSTRUCTIONS AND I THINK THEY WANT TO LINK
THEMSELVES TO THEIR RECORDS. TAB
AND 538 THAT HAS A BOT THAT TWEETS OUT AND I’D LOVE TO
UNDERSTAND THE DEGREE TO WHICH WE HAVE THIS ABILITY TO
GUARANTEE PRIVACY USING THE DEFINITION AND HOW CLOSE THE
DEFINITION MATCHES WHAT WE THINK PEOPLE ARE CONCERNED ABOUT ARE
WE LOOKING FOR OUR KEYS THOUGH THEY MAY BE SOMEWHERE ELSE JUST
BECAUSE THERE’S LOTS OF LIGHT BUT MAYBE THERE’S MORE OF A
DELTA BETWEEN THE THEORY OF WHAT WE CAN DO VERSUS WHAT MATTERS
MOST TO INDIVIDUALS AND THEIR EXPECTATIONS ABOUT PRIVACY AND
WHAT IT DOES AND DOESN’T MEAN.
>>WE HAVE A RESEARCH QUESTION ADDRESSING WHAT YOU JUST RAISED
IN THE CENTER FOR BEHAVIORAL SCIENCES METHODS.
THEY HAVE ALREADY FIELD TEST THE QUESTIONNAIRE AND THEY’LL BE
DOING THE PRODUCTION SURVEYING OVER THE COURSE OF THE LAST HALF
OF THIS CALENDAR YEAR. THE BASIC WAY IT WORK IS THAT
THE SAMPLED INDIVIDUALS ARE FIRST GIVEN THE CENSUS
QUESTIONNAIRE AND THEN ASKED A SERIES OF QUESTIONS ABOUT THE
SENSITIVITY OF THE QUESTIONS IN THAT QUESTIONNAIRE. A SERIES OF QUESTIONS ABOUT HOW
THEY PERCEIVE CONFIDENTIALITY PROTECTION TO APPLY TO THOSE
DATA. HOW WORRIED THEY ARE ABOUT
PARTICULAR CHARACTERISTICS AND IN THE LAST DESIGN THE
QUESTIONNAIRE THEY SAW CONTAINED A CITIZENSHIP QUESTION.
THEY WERE ASKED ABOUT THAT AS WELL.
IN ADDITION, THE EARLY RESEARCH FROM THAT GROUP CONFIRMS MOST OF
WHAT YOU SAID. IT’S EXTREMELY DIFFICULT FOR A
RANDOMLY SAMPLED HOUSEHOLD IN THE UNITED STATES TO RELATE TO
THE WAY TECHNICAL PROFESSIONALS IN THIS AREA DESCRIBED
CONFIDENTIALITY PROTECTION AND DESCRIBE PRIVACY RISK AND EVEN
MORE DIFFICULT FOR THEM TO RELATE TO THE QUANTIFICATIONS OF
IT THAT WE ARE USING. SO BEHAVIORAL SCIENTISTS ARE
WORKING HARD AT TRYING TO UNDERSTAND WHAT PEOPLE ACTUALLY
UNDERSTAND ABOUT CONFIDENTIALITY PROTECTION AND THEN GET MORE INFORMATION
ABOUT THE SENSITIVITY OF VARIOUS ATTRIBUTES.
THAT GROUP IS MAKING A REPORT TO THE NATIONAL ADVISORY COMMITTEE
IN ONE OF ITS UPCOMING MEETINGS AND WE CAN SCHEDULE A SIMILAR
REPORT FOR CSAC. EVERYTHING YOU SHED IS ON THE MARK.
AND IS KNOWN TO OUR GOOD CONFIDENTIALITY RESEARCH GROUP.
WE SHOULD HAVE THEM REPORT MORE OF THE RESEARCH.
>>FROM DEBORAH ON THE PHONE WRITES I APPRECIATE JOHN’S
PRESENTATION AND TERRIFIC VIDEO CAN HE COMMENT ON PRIVACY WITHIN
THE RDC NETWORK. IT SEEMS THEY MAY BECOME MORE
IMPORTANT AS ADDITIONAL PROTECTION IS PUT IN PLACE ON
PUBLICLY RELEASED DATA GIVEN SECURITY AND REQUIREMENTS TO USE
RDC IS THERE A COMMITMENT TO MINIMALLY PER–THE DATA TO MAX
— PERTURB THE DATA TO MAXIMALLY USE IT.
>>AS YOU MIGHT IMAGINE I GET THAT QUESTION RELATIVELY OFTEN.PERTURB THE DATA TO
MAXIMALLY USE IT.>>AS YOU MIGHT IMAGINE I GET
THAT QUESTION RELATIVELY OFTEN.THE DATA TO MAXIMALLY USE
IT.>>AS YOU MIGHT IMAGINE I GET
THAT QUESTION RELATIVELY OFTEN.DATA TO MAXIMALLY USE IT.
>>AS YOU MIGHT IMAGINE I GET THAT QUESTION RELATIVELY
OFTEN.TO MAXIMALLY USE IT.>>AS YOU MIGHT IMAGINE I GET
THAT QUESTION RELATIVELY OFTEN.MAXIMALLY USE IT.>>AS YOU MIGHT IMAGINE I GET
THAT QUESTION RELATIVELY OFTEN. I’LL GIVE FOLLOW-UP IF NEEDED.
CURRENTLY IN THE FSRDCs WHEN THE PROBLEM INVOLVED CENSUS BUREAU
DATA THE RESEARCHERS GET MINIMALLY PERTURBED VERSIONS OF
THOSE FILES. SOMETIMES THEY ACTUALLY GET THE
FULL UNDERLYING FILE. SOMETIMES IN THE CASE OF THE
CENSUS THEY GET THE 100% DETAILED FILE TO WORK WITH AND
IT’S PICKED AMONG OTHER THINGS. THEN THERE ARE STUDIES THAT
PRODUCE STATISTICAL SUMMARIES INTENDED FOR THE OPEN SCIENTIFIC
COMMUNITY AND GO THROUGH A CUSTOM DISCLOSURE AVOIDANCE
REVIEW THAT IS ACTUALLY EXEMPLAR IN THE WORLD IN THE SENSE THAT
THEIR FULL CODE BASE FROM THE ASSETS THEY WERE ALLOWED TO ACCESS ALL THE WAY THROUGH TO
THE TABLES THEY WISH TO RELEASE ARE BY A TRAINED AVOIDANCE
OFFICER AND THEY APPLY TRADITIONAL DISCLOSURE
LIMITATION. WE ARE NOT CHANGING THAT SYSTEM
BUT WE ARE CHANGING WHAT THE TRADITIONAL AVOIDANCE DOES
MIGRATING IT TO FORMALLY PRIVATE DISCLOSURE AVOIDANCE BUT LIKE
FOR THE AMERICAN COMMUNITY SURVEY WE HAVE NOT PUT THOSE
PRACTICES IN PLACE FOR CERTAIN RDC PROJECTS THAT WERE PROPOSED
TO MAKE MAJOR LARGE-SCALE DATA RELEASES WHICH IS UNUSUAL.
WE NORMALLY DON’T ALLOW THAT BUT THERE’S SOME ASSOCIATED WITH
INTERNAL AS WELL AS EXTERNAL RDC PROJECT.
IN THOSE CASES WE HAVE ALREADY USED FORMALLY PRIVATE METHODS AS
WE DO FOR NEW PRODUCTS BEING DEVELOPED INSIDE THE CENSUS
BUREAU. THERE ISN’T A CURE FOR THE
GENERAL SCIENTIFIC HYPOTHESIS. IT’S POSSIBLE TO MAKE PUBLIC USE
PRODUCTS TO SUPPORT THE HYPOTHESES A MICRO SAMPLE WOULD
HAVE PREPARED INCLUDING THE ONES FROM THE 2020 CENSUS.
THEY SUPPORT MORE SCIENTIFIC HYPOTHESES THAN THE RELEASE
PACKAGE DID BECAUSE THE FULL INTERACTIONS ARE IN THE ESTIMATION SET.
THERE’S NO ONE SIDE FITS ALL ANSWER.
THERE’S NO CHANGE IN THE PROCESS FOR APPLYING.
YOU COULDN’T USE THE 2010CEF SO THE WAY TO GET ACCESS TO THE
2020 CEF STILL HAD TO BE WORKED OUT BUT THAT’S A FAIR ANSWER
FROM MY POINT AND HAVE CHANGE THE WAY YOU ACCESS THE COMMUNITY
SURVEY BUT CAREFUL ABOUT THE DISCLOSURE AVOIDANCE REVIEW FOR
THE OUTPUTS AND RESEARCHERS KNOW THAT AT THE PROPOSAL STAGE.
THEY HAVE TO LAY OUT WHAT IT IS THE ARTICLE IS SUPPOSED TO TAKE
OUT. WHERE WE GET THE MOST PUSHBACK
IN THE RDCs IN THE REVISE AND RESUBMIT PROCESS FOR SCIENTIFIC
JOURNALS. THE EDITORS AND REVIEWERS, SINCE
THEY CAN’T TOUCH THE CONFIDENTIAL DATA RIGHTLY ASK
FOR THE ANALYST TO DO THINGS THAT CHALLENGE THE ROBUSTNESS OF
THEIR RESULTS. WE DON’T WANT TO RELEASE ALL THE
SITE ANALYSES. WE WANT TO PROVIDE A MECHANISM
BY WHICH THE JOURNAL EDITOR CAN COME IN AND CERTIFY WHAT THE
AUTHORS CLAIMED IS TRUE AND ARE WORKING HARD ON THAT IN A
REPRODUCIBLE SCIENCE PROJECT THAT HAS THE PARTICIPATION OF
THE MAJOR JOURNAL EDITORS. I SAID I’D GIVE A NUANCED ANSWER
BUT I DON’T WANT TO LEAVE THE IMPRESSION EVERY USE OF THE DATA
NOW HAS TO BE IN AN FSRDC. MORE DO AND WE’VE SAID THAT
SINCE 2016 BUT THAT’S NO SURPRISE TO ANYONE.
IN MOST COUNTRIES YOU CAN’T ACCESS THE MICRO DATA MANY RESEARCHERS CAN ACCESS.
THIS IS NOT A MIXED ENVIRONMENT. IT’S A MOVING ENVIRONMENT AND WE
HAVE TO ADJUST AS THE RISKS AND BENEFITS CHANGE.
THERE’S ENORMOUS BENEFITS DOING THE RESEARCH.
I’VE DONE IT MYSELF MORE THAN 20 YEARS.
THOSE ARE NOT GOING TO BE CUT OFF BY ADOPTING FORMAL PRIVACY
AS THE MAIN STANDARD FOR DISCLOSURE.
IT’S GOING TO TAKE MORE WORK TO GET IT IMPLEMENTED IN ALL PARTS
OF THE CENSUS BUREAU.>>THAT SOUNDS TOTALLY FRAZ TO
ME. — TOTALLY FRAZ — TOTALLY FAIR
TO ME AND IT HAS TO GO THROUGH DISCLOSURE AVOIDANCE.
I LOVED THE VIDEO FOR MANY THINGS AND ONE WAS THE LAST
QUESTION WHICH IS LIKE HOW DO YOU MAKE THAT TRADE-OFF BETWEEN
ANSWERING EVERY POSSIBLE QUESTION AND ANSWERING IMPORTANT QUESTIONS.
POS — POLICY QUESTIONS ARE IMPORTANT AND — TOMMY WRIGHT.
I’M SO SORRY.>>I’M GOING HAVE TO INTERRUPT.
IT’S NOW TIME FOR PUBLIC COMMENT.
I BELIEVE I ACTUALLY HAVE FIVE TIMEPIECES AND THEY ALL HAVE
FIVE DIFFERENT TIMES. I’M GOING TO GO BY MY SMARTPHONE.
WE TAKE PUBLIC COMMENTING IN WRITING AND IN PERSON AND PHONE.
IN PERSON COMMENTS ARE LIMITED TO 20 — TWO MINUTES AND
ANYTHING LONGER NEEDS TO BE SUBMITTED BY WRITTEN COMMENT AND
IF YOU’RE CALLING IN DIAL 1-888-946-7606 AND THE PASS CODE
IS 3615102 AND THE NUMBER SIGN. I’LL REQUEST COMMENTS THE PHONE
LINES LAST TO GIVE YOU TIME TO DIAL IN.
WE HAVE RECEIVED THREE WRITTEN PUBLIC COMMENTS WHICH ARE NOW
POSTED ON THE CSAC MEETING PAGE AT CONSISTENT — CENSUS.gov.
THE THREE WRITTEN SUBMISSIONS RATHER THAN SUMMARIZED THEM I
READ THE OPENING PARAGRAPHS OF EACH AND LOOK LIKE THEY FALL
UNDER TWO MINUTES AND IF YOU’LL INDULGE I’LL READ THE OPENING
PARAGRAPH. PUBLIC COMMENT OF DEBORAH STEIN.
[LOSS OF AUDIO]
>>I UNDERSTAND THERE ARE RECOMMENDATIONS TO NOT HAVE THE
DATA PRODUCTS AVAILABLE FROM THE 2010 AND ASK YOU RECONSIDER
THIS. THESE DATA ARE VITAL TO THOSE OF
US WHO FOCUS ON POLICIES RELATED TO YOUNG CHILDREN AND THEIR FAMILIES.
WITHOUT THE DATA WE CANNOT BE AS STRONG AS ADVOCATE FOR THE VULNERABLE POPULATION AND
CAN YOU PROVIDE DATA FOR SUBCOUNTY LEVEL AND AS MANY
AREAS PERTAINING TO CHILDREN AND FAMILIES AS YOU CAN.
THAT’S THE ENTIRE CORRESPONDENCE.
WE TO HAVE ANYONE WHO WANTS TO MAKE COMMENTS IN THE ROOM AND IF
SO, PLEASE PROCEED TO THE SIDE OF THE TABLE LABELLED TO THE
SIDE LABELLED PUBLIC COMMENT. THIS IS LIMITED TO TWO MINUTES.
BEFORE MAKING YOUR COMMENT, PLEASE STATE YOUR NAME AND AFFILIATION CLEARLY.
>>THANK YOU FOR THE OPPORTUNITY TO SPEAK HERE.
MY NAME’S MIA. I PRESENTED TESTIMONY ALMOST
EVERY BECAUSE I THINK SOCIAL ISSUES ARE INTERRELATED.
BUT I THINK I NEED TO REPEAT ALL THE COMMENTS IF YOU THINK ABOUT
THIS, IT’S SO SERIOUS AND I’M SURE CENSUS BUREAU IS IMPORTANT
BECAUSE YOU ARE COUNTING EVERY PERSON, EVERY PERSON’S VOICE
COUNTS. BUT I THINK THE PROBLEM IS EVERY
AGENCY IS JOINING ENFORCEMENT WITH PRIVATE ATTORNEYS ALMOST EVERYWHERE AND THEY WANT
TO ROB THEIR RIGHTS, ROB THEIR PROPERTY, ROB THEIR FAMILIES AND
THEY WOULD SHACKLE THEM AND SEND THEM TO JAIL OR SEND THEM TO
HOSPITAL OR BEHAVIOR INSTITUTE ALMOST EVERYWHERE.
YOU SAY THEY HAVE NO VOICE BECAUSE THEY’RE IN JAIL AND NOT
DO ANYTHING. THEY’RE ROBBED OF THEIR
RESOURCES, MULTI-MILLION DOLLARS AND THEY’RE MODEL
FAMILIES AND GIFTED CHILDREN AND HAVE A HIGH
ACHIEVING RECORD AND RUIN THEIR FAMILIES AND OUR SOCIETIES.
I WANT TO REALLY MENTION ABOUT THIS BECAUSE I MYSELF AM A Ph.D.
IN ECONOMICS. MY HUSBAND IS A Ph.D. IN
ECONOMICS FROM COLUMBIA UNIVERSITY.
MY TWO CHILDREN ARE GRADUATES OF M.I.
M.I.T. BUT ALL LABELLED UNFIT AND EVERYTHING WORSE AND I HAVE
TO TURN THEM AROUND BUT STILL MY HUSBAND WAS MURDERED BY WHAT I
CALL THIS [INDISCERNIBLE] AND WHAT I MEAN IS IF YOU RELATE IT
TOGETHER IT’S THE MURDER, ABUSE OF [INDISCERNIBLE] NETWORK OPERATIONS.
[INDISCERNIBLE]. NOW THE USE OF PPP, PUBLIC PRIVATE PARTNERSHIP
IS PROPAGANDA FROM ALL ORGANIZATION TO GOVERNMENT
AGENCIES. WHETHER IT’S THE DEPARTMENT OF
JUSTICE OR SMALL BUSINESS ADMINISTRATION AND
[INDISCERNIBLE] AND ALL THREE BRANCHS FROM LOCAL TO FEDERAL TO
GLOBAL DOESN’T MEAN ANYTHING. TODAY [INDISCERNIBLE] AND IGNORE
ALL THOSE. YOU HAVE TO FILE WITH THE
SUPREME COURT WITH THE CONSTRUCTIVE DISCHARGE AND STILL
GO NOWHERE BECAUSE THEY TAKE AWAY ALL OF YOUR RIGHTS AND
LITIGATION AND IGNORE EXCUSES AND THEY DON’T ALLOW YOU TO FILE AND —
>>I’M VERY SORRY. THIS IS TOMMY WRIGHT.
THE TWO MINUTES HAVE BEEN REACHED ACTUALLY EXCEEDED.
IF YOU HAVE WRITTEN COMMENTS, WE’LL BE HAPPY TO RECEIVE THOSE
BUT YOUR ALLOWANCE FOR TWO MINUTES HAS BEEN EXCEEDED.
I’M SO VERY SORRY.>>I HAVE A LOT OF COMPLAINTS.
THEY TAKE AWAY THE RESOURCES. I’D LIKE TO SUBMIT A COMMENT.
>>PLEASE DO SUBMIT.>>WITH ATTACHMENTS.
EVERY VOICE SHOULD COUNT. I HOPE THIS CENSUS BUREAU TAKES
SERIOUS ACCOUNT AND THE INFORMATION FROM THERE.
>>WE WILL RECEIVE YOUR COMMENTS.
>>THANK YOU.>>THANK YOU VERY MUCH.
ARE THERE ANY PUBLIC COMMENTERS ON THE PHONE?
MAYBE I SHOULD ASK IF THERE’S OWNER PUBLIC COMMENTS IN THE
ROOM? NONE ON THE PHONE?
OKAY. WELL, THANK YOU VERY MUCH.
WE’LL CONTINUE WITH THE PRESENTATION AND DISCUSSION.
ALISON AND MAYBE JOHN.
>>THE VIDEO ENDED WITH A GOOD ABOUT DETERMINING WHAT ARE
IMPORTANT USES AND QUESTIONS.
WONDERED HOW THE CENSUS BUREAU IF THE CENSUS BUREAU CURRENTLY
HAS GUIDANCE ON DETERMINING THAT.
>>IS THIS JOHN ABOWD AGAIN. I SHOULD SAY IT’S NEW GROUND FOR
EVERYBODY IN THE ROOM AND IN THE CENSUS BUREAU.
SO WE HAVE MADE A VERY STRONG EFFORT TO IDENTIFY FIRST KNOWN
STATUTORY USES OF THE DATA AND WE OURSELVES A LARGE
TEAM IN THE BUREAU ARE ATTEMPTING TO ASSESS THE PRIVACY
LOSS BUDGET NECESSARY TO MEET REASONABLE INTERPRETATIONS OF
THE STATUTORY REQUIREMENTS. ONE OF THE THINGS AND COMMUNICATING THE NOTION AT THE
GEOGRAPHIC LEVEL THOSE ARE OFFICIAL AND BUILD ARBITRARY
GEOGRAPHIC AGGREGATES THEY’RE NOT INTENDED STO — TO STAND
ALONE AND IS THE BLOCK LEVELS FROM SYNTHESIS WERE NEVER
INTENDED TO STAND ALONE FROM THE BLOCK GROUP.
THEY’RE DESIGNED FOR PIXELS OF BLOCK SIZES.
THE BLOCK IS THE SMALLEST PIXEL AND HOW IT CAME INTO THE
TABULATION PRODUCTS. WHAT WAS THE SMALLEST GEOGRAPHIC
UNIT TO MOVE PEOPLE AROUND TO MEET.
IT WASN’T TO DESIGN VOTING DISTRICTS THAT HAD EQUAL
POPULATION AND ONE BLOCK EACH. THAT’S NOT THE USE CASE.
THE USE CASE IS VOTING DISTRICTS FOR LEGISLATIVE BODIES FOR
GOVERNMENTAL LEVELS THAT HAVE TO HAVE EXACTLY OR APPROXIMATELY
EQUAL POPULATIONS SO YOU NEED A PIXEL THAT CAN ACCOMMODATE THAT
AND WITH THE OFFICIAL POPULATION IN THE MIXEL AND THE ABILITY —
PIXEL AND THE ABILITY TO MOVE THAT AROUND AND THE GEOGRAPHIC
DEFINITION IS MAKING THAT MORE ACCURATE AS THOSE GEOGRAPHIC
AREAS GET MORE POP ULOUS AND SOME OTHER USE CASE
MORE DIFFICULT AND WE’RE WORKING ON THEM.
ANOTHER EXAMPLE IS OUR OWN ESTIMATES ON PYRAMIDS.
WE’RE EXAMINING THAT KIND OF SUMMARY TO MAKE SURE THE DATA
ARE USEFUL FOR THAT PURPOSE TOO. THERE’S A LOT OF INTEREST GROUP
AND IT’S EXTRAORDINARILY HARD TO BALANCE THE INTEREST AND NOT
SOMETHING WE HAVE A LOT OF PRACTICE DOING AND WE GET GOOD
AT AT IT FAST. WE SEEK INPUT NOT JUST ON THE
ACCURACY BUT THE DECISION MAKING PROCESS AND WHAT IS INHERENT IN
SETTING THE PRIVACY LOSS BUDGET.
>>THIS IS DEBORAH. SINCE THE ENVIRONMENT IS MOVING
WILL WE SEE IT IN THE RRC BUT I THINK SHE MEANT THE CENSUS RPCs.
THANK YOU. DO YOU UNDERSTAND THE QUESTION?
IT’S KIND OF VAGUE.>>JOHN ABOWD.
I DON’T SEE HOW IT’S DIFFERENT FROM THE ONE I ANSWERED EARLIER.
>>NO, WE’RE GOOD. KATHY WANTED TO SAY SOMETHING.
>>I WAS INTERESTED IN THE QUESTION AND
IN THINKING WITH THE TIMING OF THIS RELEASE OF THE CITIZENSHIP
STRUCTURE IS MARCH 31, 2020 IN THE HEAT OF TRYING TO ENCOURAGE
PEOPLE TO ANSWER THE CENSUS. SO HAVE THERE BEEN THOUGHTS
ABOUT COMMUNICATION ON THAT. NOT THAT I UNDERSTAND THAT IT’S
SILOED BUT ALSO WHAT WOULD BE PICKED UP BY THE NEWS OR SOCIAL
MEDIA OR OTHER THINGS THAT COULD ENCOURAGE MISINFORMATION.
>>SO WE’RE NOT GOING TO GIVE YOU ANYTHING THAT DISRUPTS THE
OPERATIONS OF THE 2020 CENSUS. THAT WOULD BE COUNTERPRODUCTIVE.
IN EXPLAINING WHERE THOSE DEADLINES COME FROM, THEY COME FROM THE NECESSITY TO
FORMALIZE WHAT WE’LL DO AND THEY’RE HARD INTERNAL DEADLINES
BUT THE COMMUNICATION PLAN WILL BE DEVELOPED IN A MANNER THAT
ENSURES THAT THE 2020 CENSUS CAN BE CONDUCTED WITH MINIMAL
DISRUPTION.>>AND JOE.
>>THANK YOU FOR THE PRESENTATION AND THE VIDEO I
THOUGHT WAS SUPERB AND WE HAD A SNEAK PREVIEW NEXT TO OUR
COLLEAGUE. I THOUGHT THAT WAS EXCELLENT.
IN TERMS OF THE COMPLEXITY OF THIS ISSUE IN TERMS OF THE
CENSUS LOOKING AT SOME OF THE CITIZENSHIP RELATED ISSUES IS
SOMETHING WE LOOKED AT CLOSELY IN HOMELAND SECURITY IS WHAT WE
DID. THE CBP AND CITIZENSHIP AND
IMMIGRATION SERVICES AND THE HOMELAND SECURITY
INVESTIGATIONS. THOSE THREE GROUPS CITIZENSHIP
OF HOMELAND WAS DIVIDE INTO THREE PIECES FROM THE HOMELAND
SECURITY ACT IN THE WAKE OF 9/11 AND NOT IN THE BROAD CENSUS
FORMAT WE’RE THINKING ABOUT AND IT WAS IMPORTANT TO BE INVOLVED
IN SOME LARGER EQUATION LOOKING AT THIS FROM A QUANTIFICATION
POINT OF VIEW. IN OTHER WORDS, I THINK THERE IS
SOMETHING AND MAYBE KRISHNA MADE THE POINT ABOUT QUANTIFICATION.
IT’S IMPORTANT IT BE MOVED FROM THE BLOCK LEVEL CONCERNS AND
LOOK AT IT FROM A MORE GEOGRAPHIC SPACE AND FROM
MY POINT OF VIEW AT THE STATE LEVEL WHAT’S THE ISSUE.
WHAT’S CITIZENSHIP LOOK LIKE IN AMERICA IN 2020.
I WANT TO ARTICULATE THE MEI CRO USES OF U.S. HOMELAND OF
ENTRY AND WHO WOULD GET A VISA AT A CONSULAR SITUATION AROUND
THE WORLD. TRUSTED TRAVELER PROGRAMS.
MANY OF YOU USE THOSE PROGRAMS.
AND LOOK AT THE DATA HIGHLY SENSITIVE AND IN HOMELAND WE DID
A GOOD JOB OF KEEPING IT EXTREMELY CLOSE BUT TO GET THE
PRIVILEGES PEOPLE HAD TO SUBMIT INFORMATION TO HOMELAND SECURITY
TO OBTAIN THOSE PRIVILEGES. SOME OF WHICH THE CITIZENSHIP
WAS A RELEVANT FACTOR IN US EVALUATING THE CIRCUMSTANCES BUT
I DO THINK AS I SIT WHERE I SIT CURRENTLY, MY VIEW OF THIS IS IT
SEEMS AS THOUGH THE CENSUS HAS TO BE PROTECTED FROM THESE
CONCERNS ABOUT THE INFORMATION WILL BE USED SOMEHOW TO PUSH
SOMEONE ASIDE WHO IS CURRENTLY IN THE UNITED STATES.
I STRONGLY ENDORSE MAKING SURE WE PROTECT THIS LIKE WE DID AT
HOMELAND THE PERSONALLY IDENTIFIABLE INFORMATION WE HAD
THE SENSITIVE HOMELAND SECURITY INFORMATION AND HAD THE NATIONAL
SECURITY INFORMATION AND HAD ANY NUMBER OF DIFFERENT ISSUES.
I THINK AND HOPE THE PUBLIC HAD CONFIDENCE WE WERE MAINTAINING THOSE IN A DISCREET WAY.
THAT’S SOMETHING I APPRECIATE YOU’VE BEEN DOING AND I
APPRECIATE THE LEADERSHIP THAT’S BEEN GIVEN THE ISSUE.
I ALSO LOOK FORWARD TO HEARING FROM I DON’T KNOW WHAT WE CAN
HEAR FROM THE INTERNAL WORKING GROUP WHEN IT’S MEETING AND
DISCUSSING SOME OF THESE ISSUES BUT I DO VALUE THE PARTICIPATION
IN THIS. AS A STEP BACK FROM IT, I DON’T
KNOW I’M NO LONGER IN GOVERNMENT BUT I WONDER WITH THE UTILITY OF
THIS OR ON A MACRO LEVEL AND I DON’T KNOW WHAT THE REQUESTED
USES HAVE BEEN. IN OTHER WORDS, OF PEOPLE SAYING
WE WANT THIS DATA AND FOR UTILITY, WHAT UTILITY THIS
SERVE. IN OTHER WORDS, WHAT ARE THE
REQUESTED USES, JOHN, THIS INFORMATION WOULD BE FOR.
IN OTHER WORDS, IS THERE UTILITY IN THE STATE OF GEORGIA TO KNOW THE ANECDOTAL NUMBER AND
THERE’S 11 MILLION RESIDUAL PEOPLE WITHOUT DOCUMENTS AND
WHATEVER THE NOMENCLATURE IS UNDOCUMENTED OR ILLEGAL THERE’S
NEW TERMINOLOGY UTILIZED. THE AWARENESS OF THIS MAY BE A WAY FOR
LEGISLATURES TO SET ASIDE ASSIMILATION OF POPULATIONS THAT
ARE THERE BUT I’M JUST MAKING SOME OF THIS UP RIGHT NOW SO
I’LL END MY COMMENTS BECAUSE I APPRECIATE THE TOLERANCE OF
TOMMY AND ALLISON THROUGH THESE LENGTHY COMMENTS BUT I STRUGGLE
WITH THIS BECAUSE I HAVE THE PREVIOUS BACKGROUND.
IT’S PERSON TO KNOW THESE THINGS IN THESE DISCREET WAYS BUT IN TERMS OF THE BROADER UTILITY OF
THE INFORMATION INTERESTED IN WHAT REQUEST OF USE ARE
ARTICULATED. THE LAST THING WE NEED TO HAVE
HAPPEN IS SOMEONE NOT FILL OUT THEIR CENSUS FORM.
THIS IS MY ARTICULATED CONCERN BECAUSE IT MAY BE USED AGAINST
THEM INDIVIDUALLY. I DON’T THINK THAT’S THE KIND OF
WAY THE INFORMATION WILL BE USED BUT I’M NOT CERTAIN BECAUSE I
DON’T KNOW HOW IT WOULD BE USED. I SET OUT A COUPLE EXAMPLES
WHICH ARE HOW WE APPROPRIATE MONEY.
CERTAINLY APPORTIONMENT WON’T BE AFFECTED BECAUSE WHETHER
SOMEONE’S A CITIZEN OR NON-CITIZEN CONGRESSIONAL
DELEGATION ARE BASED ON PEOPLE IN A GEOGRAPHIC AREA.
THAT MAY BE MORE OF A COMMENT THAN A QUESTION.
I WANTED TO SHARE THOSE THOUGHTS WITH YOU.
SEE IF YOU CAN DIG A QUESTION OUT OF THAT, JOHN.
>>THIS IS JOHN ABOWD, I DON’T THINK IT WILL COME AS A SURPRISE
I TAKE QUESTIONS WITH EXTREME SENSITIVITY.
WHAT I’D LIKE TO SAY IS THE SLIDES I PRESENTED ON THE USES
OF THE 2020 CENSUS FOR APPORTIONMENT AND REDISTRICTING
AND CVAP IS WHAT WE UNDERSTAND WILL BE OUR PROCEDURE.
I ALSO WANTED TO CLARIFY THAT APPORTIONMENT IS BASED ON THE
RESIDENT POPULATION AND RESIDENT POPULATION IS A WELL DEFINED
ENTITY THAT INCLUDES BOTH CITIZENS AND NON-CITIZENS.
THE PRODUCTION OF CVAP DATA WE’RE GATHERING INFORMATION ON
THE USE CASE IT’S SUPPOSED TO MEET AND WE’LL BE AS TRANSPARENT
AS WE CAN BE ABOUT THAT PRODUCT WAS INTENDED FOR THOSE USE CASES.
>>IT’S THE ROACH MOTEL, RIGHT, COMES IN, DOESN’T GO OUT.
THAT’S ALWAYS IMPORTANT
>>LET ME REPEAT IT BECAUSE THIS IS JOHN ABOWD IT’S ALWAYS
IMPORTANT TO NOTE WE ARE INGEST ING IDENTIFIABLE DATA BUT MAKING
WHAT I THINK CAN BE REASONABLY CHARACTERIZED AS AN EXTREMELY
STRONG EFFORT TO MAKE SURE THE PUBLICATIONS CAN ONLY BE USED
FOR STATISTICAL PURPOSES AND THEY DON’T ENDANGER ANY OF THE RESPONDENTS
AND WE’RE DOING OUR BEST TO PROTECT THE CONFIDENTIALITY OF
THE DATA AND ANY OTHER DATA IN COMBINATION WITH THE 2020
CENSUS.>>THAT’S A MANTRA.
IT HAS TO BE TREATED. I DON’T KNOW HOW IDEAL WITH THE
ROACH MOTEL METAPHOR BUT IT’S CRITICAL INFORMATION BE TREAT
KNEAD WAY THAT ENCOURAGES PEOPLE TO ENGAGE IN WHATEVER THEIR CITIZENSHIP STATUS MIGHT BE.
THIS IS A UNIQUE AMERICAN PHENOMENON.
I’M PREACHING TO THE CHOIR BECAUSE THOSE WHO LIVED HERE
YOUR ENTIRE CAREERS LIVED THIS BUT I APPRECIATE THE
RESPONSIBILITY HAVING DONE SOME OF THIS IN A PREVIOUS LIFE TIME
WITH SOMETHING CALLED THE CENSUS MONITORING BOARD.
THANK YOU.>>THIS IS JUAN PABLO.
I HAVE A QUICK COUPLE SUGGESTIONS.
FIRST, A LITTLE BIT OF PRAISE FOR JOHN’S WORK AND THE CENSUS
BUREAU’S WORK ON DIFFERENTIAL PRIVACY.
IN THE RESEARCH WORLD WE TIMES TALK ABOUT RESEARCH OFTEN PAPERS
COME OUT ABOUT A TOPIC AND TAKES 10 YEARS FOR ANY IMPLEMENT TO
HAPPEN AND WHAT WE’RE SEEING HERE IS A LOT OF RESEARCH BEING
IMPLEMENTEDLEAD-IN RIGHT WAY. RELATED TO THAT JUST SUGGESTIONS ON A COUPLE COMMUNITIES THAT
KEEP COMING TO MIND EVERY TIME I LISTEN TO JOHN’S PRESENTATIONS.
ONE IS A NORMAL METHODS IN SCIENCE.
THERE’S BEEN A LITTLE BIT OF RESEARCH BUT I
THINK IT WILL A GOOD MATCH.
AND IN WORKING WITH HEALTH RECORDS A LOT OF THAT IS BEING
PRODUCED. I THINK WHAT I HAVE SEEN FOR
FOLKS WHO DO THAT KIND OF RESEARCH IS NOT NEARLY AS
SOPHISTICATED AS WHAT YOU’RE DOING HERE.
AND IN MANY WAYS THAT DATA FOR A LOT OF PEOPLE MAY BE MORE
SENSITIVE. I THINK THAT MIGHT BE ANOTHER
INTERESTING CONNECTION WHERE MAYBE MORE MINDS, MORE THINKING
ABOUT THIS PROBLEM COULD LEAD TO GOOD OUTCOMES.
>>I THINK THAT’S IT. FOLKS WHO HAVE COMMENTS OR
SUGGESTIONS, KRISHNA IS GOING TO BE ACCUMULATING THEM AND SO
PLEASE BE SURE TO PLAN TO GET WITH HIM ON THAT.
WE’LL DO THAT A LITTLE BIT LATER TODAY AND I THINK WE’LL TAKE A
BREAK. ARE WE ON TIME?
WE GET TWO EXTRA MINUTES OF BREAK.
I THOUGHT WE WERE EARLY. SO WE’RE COMING BACK AT 11:45. AS SOON AS THE CHAIR COMES AND TAKES HER CHAIR.
>>TOMMY WRIGHT. AND BILL DAVY TO PRESENT ON THE
PLANSPLANS.>>I’M THE ECONOMIC CENSUS
SURVEY DIRECTOR. NEXT TO ME IS BILL DAVY THE
METHODOLOGY DIRECTOR AND WE’LL BE SLATING UP THE PRESENTATION TODAY.
WE’LL PRESENT AN UPDATE ON THE ECONOMIC CENSUS REGARDING OUR
DATA DISSEMINATION FOR THE ECONOMIC CENSUS.
IN 2018 I PROVIDE AND AN UPDATE IN THE DATA COLLECTION PROCESS
WE WERE IN THE MIDDLE OF. I CLOSED OUT IN MARCH OF 2019
AND WE’VE NOW MOVED INTO DATA REVIEW AND EDITING AND WILL
RELEASE OUR FIRST SET OF DATA FROM THE 2017 ECONOMIC CENSUS
NEXT THURSDAY SEPTEMBER 19. SO BRIEF BACKGROUND INFORMATION
JUST TO GET EVERYBODY ON THE SAME PAGE.
THE ECONOMIC CENSUS IS AUTHORIZED UNDER TITLE 13 OF THE
UNITED STATES CODE SECTION 131. THE DATA’S COLLECTED EVERY FIVE
YEARS IN ENDING TWO AND SEVEN. THE CONTENT IS DEVELOPED IN
COORDINATION WITH BUSINESS AND GOVERNMENT.
THE ECONOMIC CENSUS IS THE MOST COMPREHENSIVE MEASUREMENT OF THE
U.S. ECONOMY. WE PROVIDE STATISTIC AT THE
NATIONAL STATE AND LOCAL LEVELS AND IT PROVIDES INFORMATION ON
INDUSTRY REVENUES AND OWNER REVENUES OF AMERICAN BUSINESS
PERFORMANCE THAT ARE CONSISTENT, COMPARABLE AND COMPREHENSIVE
ACROSS ALL INDUSTRIES AND GEOGRAPHIC AREAS.
WE HAVE A WIDE RANGE OF DATA USERS.
SOME BIG ONES ARE LISTED HERE. DISTRICT OF ANALYSIS USE THE
DATA FOR GROSS DOMESTIC PRODUCT AND PRODUCT ACCOUNTS.
CENTER FOR MEDICARE AND MEDICAID USE IT INTO THEIR GEOGRAPHIC
HEALTH SPENDING BY STATE. THE FEDERAL RESERVE WILL USE IT
AS A BENCHMARK FOR THEIR DEVELOP PRODUCTION AND CAPACITY NUMBERS.
BUREAU OF LABOR USES IT AS INPUT INTO THEIR PRODUCER PRICE INDEX AND THERE’S OTHER USERS
THAT USE THE DATA FOR MAKING PERSONAL AND BUSINESS DECISION OR POLICY.
THAT’S CURRENTLY 31.1 BUSINESS ESTABLISHMENTS IN THE UNITED
STATES. THE KEY WORD IS ESTABLISHMENTS.
IT’S AN ESTABLISHMENT-BASED SURVEY.
OF THOSE 7.8 MILLION HAVE EMPLOYEES.
THAT IS THE UNIVERSE OF THE ECONOMIC CENSUS. SO TO WRAP UP DATA COLLECTION
WHICH I SPOKE ON IN DECEMBER, I HAVE A COUPLE GRAPHS I WANT TO SHARE.
THESE ARE RESPONSE RATES GRAPHS ABOUT 75%.
THE FIRST GRAPH IS THE RESPONSE RATE BY STATE.
ON THE MAIL LOCATIONS. WHERE DID WE MAIL THE LETTERS
NOR RESPOND ENT TO REPORT. YOU CAN IMAGINE LARGE COMPANIES
RECEIVE AT THEIR HEADQUARTERS LOCATION TO REPORT.
YOU CAN SEE THE RESPONSE BY STATE.
THE LOWEST RESPONSE BY STATE WOULD HAVE BEEN SEEN IN CONNECTICUT.
IF IT’S CLOSER TO RED IT’S A LOWER RESPONSE AND YOU CAN SEE
WHERE WE GOT A LOW AND HIGH RESPONSE.
LOWEST WAS CONNECTICUT, U.S. VIRGIN ISLANDS AND PUERTO RICO
AND HIGHEST WERE IOWA, ARKANSAS AND THE COMMONWEALTH OF THE
ISLANDS. THE NEXT IS A RESPONSE RATE BY
FISCAL ESTABLISHMENT WE MAY HAVE GONE TO THE HEADQUARTERS AND IT
SHOWS BY THEIR ACTUAL PHYSICAL LOCATION WHERE THE RESPONSE
RATES WERE BY STATE. SAME COLOR CODING.
ANYTHING CLOSER TO THE DEEP RED IS LOWER RESPONSE BY STATE AND
DEEP BLUE IS A HIGHER RESPONSE. LOWEST ARE D.C., NEW YORK AND
RHODE ISLAND AND HIGHEST WERE IN IOWA AND WISCONSIN.
IT GIVES A GOOD GRAPHICAL REPRESENTATION OF WHERE WE SAW RESPONSES.
LEADING UP TO DATA DISSEMINATION IT GOES THROUGH THE BUSINESS
REGISTER FIRST AND IT UPDATES THINGS SUCH AS ORGANIZATIONAL
STRUCTURE, CHANGES IN OWNERSHIP, NEW ESTABLISHMENT ADS AND NEW
LOCATIONS ADDED TO A COMPANY, INFORMATION RELATED TO GEOGRAPHY
AND SO FORTH. WE MAKE SURE WE HAVE THE COMPANY
STRUCTURE CORRECT ON THE BUSINESS REGISTER.
IT’S IMPORTANT NOR CENSUS AND INDICATORS THAT RELY ON CENSUS
DATA FOR BENCHMARKING. THE SPECIFIC TRADE AREA SUBJECT
MATTER EDITS. WHAT WE CALL THE BACK END WHEN
WE RECEIVE THE DATA IN THE TRADE AREAS WE LOOK AT THE DATA AND
PERFORM CHECKS FOR COMPLETENESS AND CONSISTENCY AND COMPARE TO
INDUSTRY AVERAGES AND LOOK AT HISTORIC INFORMATION AND
VALIDATE LOGICAL SCENARIOS SUCH AS DOES THE REVENUE AND PAYROLL
MAKE SENSE AND THEY’RE NOT REPORT BEING AT A LARGER RATE
THAN REVENUE AND PERFORMING AMPUTATIONS OF THE DATA TO MAKE
SURE WE HAVE A COMPLETE RECORD ARE THERE MISSING ITEMS WE HAVE
TO ACCOUNT FOR. BOTH IN THE BUSINESS REGISTER
AND TRADE AREAS WILL HAVE ANALYSTS REVIEW THE CASE AND
MAKE UPDATES AS NECESSARY BASED ON WHAT THE REFERRAL WAS.
IN GENERAL OUR REFERRAL RATES ARE BETWEEN THE NEIGHBORHOOD OF
17% TO 20% BETWEEN THE BUSINESS REGISTER AND TRADE AREAS. IT’S A FAIR NUMBER OF REFERRALS
THAT GET WORKED IN THE PROCESS. WITH THE REFERRAL, SOME
REFERRALS TAKE NO TIME TO FIX AND SOME TAKE HOURS DEPENDING ON
THE AMOUNT OF RESEARCH NEEDED TO UPDATE THE RECORD.
SO CONTINUING WITH DATA EDITING REVIEW WE’LL LOOK AT THE LARGEST
COMPANIES TO ENSURE MICRODATA OF THE MOST COMPLEX COMPANIES ARE
REVIEWED AND COORDINATED KROOGS — ACROSS THE AREA AND WE’LL
MAKE SURE THE COMPANY HOLISTICALLY MAKES SENSE ACROSS
THE TRADE AREAS. FOR THE MOST PART THEY RECEIVE
THE MOST REVIEW AND WE’LL SPEND A LOT OF TIME LOOKING AT THE
LARGE COMPANIES AND LOOK AT INFLUENCERS AND THEY HAVE TO THE
MOST REVENUE OR MOST PAYROLL. WE’LL BE SURE TO MAKE THOSE
CASES ARE CORRECT AND THEN UTILIZE VARIOUS PROGRAMS IN THE
EDITING INCLUDING A MASS CORRECTION.
IF THERE’S SOMETHING SYSTEMATICALLY GOING WRONG IS IT
A WAY A COMPANY REPORT AND WE’LL CORRECT RECORDS QUICKLY RATHER THAN ONE BY ONE.
SO ONCE THE MICROREVIEW IS COMPLETE WE’LL MOVE TO THE STAGE
OF MACRO REVIEW WHICH IS WHERE WE ARE NOW.
IT’S CONDUCTED WHEN THERE’S A FULL SET OF TABULATED DATA THAT
INCLUDES HAVING ALL CASES THAT REPORTED BACK TO US AND NON
MAILS AND ADMINISTRATIVE DATA ASSOCIATED AND DELINQUENT CASES
WE HAVE TO IMPUTE FOR. AT THAT POINT WE’LL TAKE THE
RECORDS AND AGGREGATE THEM TO A MACRO LEVEL TO START A REVIEW OF
STATE OR INDUSTRY LEVEL DATA. AND TO DO THAT IN ORDER TO
SUMMARY THE DATA IS READY TO BE RELEASED.
WE’LL START TO REVIEW PROBLEM IDENTIFICATION OF SPECIFIC
CELLS. CELLS THAT DON’T MAKE SENSE AT
AN AGGREGATE. IF WE’RE LOOKING AT COMPARISONS
OF 2017 AND 2012 DATA AND SEE AN INDUSTRY OF A TREND UP 150% AND
IT DOESN’T MAKE SENSE AND WE’LL DRILL INTO THE SPECIFICS OF THAT
INDUSTRY TO SEE WHAT’S GOING ON THERE AND IF THERE’S A SET OF
RECORDS THAT STILL NEED TO BE CORRECTED.
WE’LL LOOK FOR DARTH DIFFERENCES FROM 2012 TO 0217 — 2017 AND
LOOK AT STATES WITH HIGH IMPUTATION LEVELS AND TRY TO
FIGURE OUT HOW TO GET THEM DOWN BY REVIEWING THE REPORTED DATA
AND MAKING SURE MAYBE THERE WEREN’T DATA SLIDES OR OTHER
ISSUES THAT CAUSED THE RECORD TO BE IMPUTED.
WE’LL LOOK AT HISTORIC LEVELS AND DOCUMENT ALL THE FINDINGS.
THE LAST BULLET POINT TALKS ABOUT RECONCILIATION.
WE’LL RECONCILE THE DATA WITH OUR ANNUAL SURVEYS.
OUR QUARTERLY AND MONTHLY PROGRAMS TO MAKE SURE THE DATA
ACROSS ALL PROGRAMS MAKE SENSE AND RELY ON OUTSIDE SOURCE DATA
FROM OTHER AGENCIES OR TRADE ASSOCIATIONS AND OUTSIDE EXPERTS
TO MAKE SURE THE DATA MAKES SENSE BEFORE WE RELEASE. WE ARE RELEASING OUR FIRST DATA
PRODUCT FROM THE 2017 ECONOMIC NEXT WEEK SEPTEMBER 19.
IT’S THE FIRST-LOOK REPORT AND PRELIMINARY DATA BY INDUSTRY.
THE DEMAND IS FOR DETAILED DATA. THAT DATA WILL COME OUT IN THE
NEXT YEAR OR TWO WITH OUR GEOGRAPHIC AREA SERIES.
NEEDS OF THE PRIMARY FEDERAL AGENCIES AND THEIR REQUIREMENTS
HAVE THE MOST WEIGHT. WE DO WORK AND HAVE DISCUSSIONS
WITH DDA, FRB AND CMS ON WHAT THEIR NEEDS ARE.
WE’LL BE UTILIZING THE NEW ENTERPRISE TOOL FOR DISSEMINATING DATA.
THEY’RE RELEASES ARE BEING STANDARDIZED TO WHAT WE’RE
CONSIDERING THE ECONOMY-WIDE STATISTICS DATA.
THAT WAY THE DATA BY INDUSTRY ARE CONSISTENT AND COMPARABLE
WHEN YOU LOOK ACROSS INDUSTRIES. IN PAST CENSUS YEARS WE’D
RELEASE RETAIL OR WHOLESALE AND THE DATA WAS NOT ALWAYS RELEASED
IN THE SAME FORMAT OR TIME AND NOW WE’RE DOING RELEASES ECONOMY
WIDE. AS I MENTIONED THE RELEASES
START IN SEPTEMBER, THIS MONTH AND THEY’LL BE COMPLETED BY DECEMBER 2021.
SO THIS IS CHART THAT SHOWS YOU THE DIFFERENT DATA PRODUCT
RELEASES WE HAVE. THE FIRST-LOOK REPORT WHICH IS
GENERAL STATISTICS AT THE TWO THROUGH SIX DIGIT INDUSTRY
LEVEL. IT WILL BE RELEASED NEXT WEEK
AND WE HAVE THE SERIES TO RELEASE GENERAL STATISTIC DATA.
REVENUE, PAYROLL, EMPLOYMENT AND OTHER VARIABLES.
AT THE U.S. STATE AND METRO COUNTY AND PLACE LEVEL AGAIN FOR
THE TWO THROUGH SIX-DIGIT INDUSTRIES.
THOSE WILL START TO BE RELEASED IN JANUARY OF 2020 ON A FLOW
BASIS THROUGH NOVEMBER 2020. SO WE’LL BE RELEASING DIFFERENT STATES AND
WE’LL RELEASE THE GOODS AND SERVICES PRODUCED BY INDUSTRY.
THAT WILL BE RELEASED IN NOVEMBER 2020 AND RELEASES FROM
2020 TO DECEMBER 2021. THE BIG POINT TO TAKE AWAY FROM
THE SLIDE IS THROUGH THE SET OF RELEASE THE FIRST LOOK TO THE
FINAL THERE’S INCREMENTAL IMPROVEMENT AND NEXT WEEK IT’S
PRELIMINARY AND AS WE RELEASE THE SERIES THE DATA BECOMES
FINAL. SO YOU GET BETTER DATA THE
FURTHER WE GO ALONG AND MORE SPECIFIC DATA AS WE GO ALONG.
IL SHOWS HOW ALL THE DATA RELATES THROUGH THE RELEASES.
THE TOP BAR THAT RUNS ACROSS. THE FIRST LOOK REPORT IS U.S.
LEVEL DATA, PRELIMINARY DATA AND SHOWS ESTABLISH MANIES AND
PAYROLL AND REVENUE FOR A PARTICULAR INDUSTRY AND
MANUFACTURING. THIS IS DATA FROM 2012.
THEN WE RELEASE THE GEOGRAPHIC AREA SERIES.
THE BOXES UNDERNEATH. YOU’LL SEE FOR EACH STATE WE NOW
PRODUCE THE DATA AND TOTAL EMPLOYMENT, PAYROLL AND REVENUE.
ACROSS THE 50 STATES AND D.C., IF YOU AGGREGATE THOSE THEY GIVE
THE U.S. TOTAL SHOWN IN THE FIRST-LOOK REPORT.
FROM THERE, WE RELEASE THE PRODUCTS OR GOODS AND SERVICES
DATA THE BOTTOM BAR THAT RUNS ACROSS.
THAT’S TAKING THE TOTAL REVENUE ORIGINALLY REPORTED IN THE FIRST
LOOK AND FINALIZED THROUGH THE DATA AND BREAKS OUT THE REVENUE
TO THE DIFFERENT GOODS AND SERVICES PRODUCED.
YOU SEE THE RED LINE CONNECTING THE DATA FROM THE TOP REVENUE
SHIPMENTS OF $20 BILLION DOWN TO PASSENGER AND
TRUCK TIRES ET SET VA. YOU CAN SEE HOW THE — ET SEAT
RA. YOU CAN SEE — ETCETERA.
YOU CAN SEE HOW THE DATA CONNECTS.
THIS IS A QUICK LINK TO SHOW YOU THE RELEASE SCHEDULE PAGE YOU
CAN GO IT ON YOUR PONE IT’S BASICALLY THE INFORMATIONIVE
JUST PROVIVEDED. PROVIDED.
THEN WE WANT TO TALK TO YOU TODAY A LITTLE BIT ABOUT OUR
DATA DISSEMINATION RISKS.
THE BIG ONE BILL WILL TALK ABOUT IS THE IMPACT ON THE DATA
RELATED TO DISCLOSURE METHODOLOGY AND THE REQUIREMENTS
IN RELEASING THE DATA. THE SECOND IS BUDGETARY IMPACTS.
WE’RE ALWAYS IN A STATE OF TIGHT BUDGETS AND THERE’S RISK TO
FUTURE DATA PRODUCTS BEING RELIEVED INCLUDING OUR
GEOGRAPHIC SERIES AND PRODUCT TABLES AND FINAL DATA RELEASE
TABLES. ALL THOSE RELEASES ARE PLANNED
FOR BUT WE ALSO NEED THE BUDGET FUNDS IN ORDER TO SPEND THE TIME
TO REVIEW AND RELEASE.
I THINK BILL WILL TALK ABOUT THE DATA DISSEMINATION ISSUES.
>>THANKS, BILL. AS MENTIONED EARLIER WE USE
ADMINISTRATIVE RECORDS DATA FROM OTHER AGENCIES IN THE ECONOMIC
CENSUS FOR VARIOUS REASONS WHETHER IT’S FOR EDITING,
AMPUTATION OR DIRECT SUBSTITUTION IN LIEU OF A
RESPONSE? SO THE SOURCE, ONE OF OUR
PRIMARY SOURCES OF THE ADMINISTRATIVE DATA IS IRS.
I WANT TO TALK ABOUT IRS HAS DONE IN TERMS OF THEIR
CONFIDENTIAL REQUIREMENTS. SO BACK IN SEPTEMBER 2016 OF THE
IRS MADE A REVISION TO THE CONFIDENTIALITY REQUIREMENTS FOR
AGENCIES THAT RECEIVE AND USE TAX RETURN INFORMATION IN THEIR STATISTICAL REPORTS.
AND THE PRIMARY CHANGE WAS THEY ADDED ADDITIONAL THRESHOLDS
REGARDING THE NUMBER OF TAX RETURNS THAT MUST BE INCLUDED IN
A CELL FOR THE ESTIMATE TO BE RELEASED.
THE GUIDANCE SAID THE AGENCIES MUST ADHERE TO THE IRS
GUIDELINES OR EQUIVALENT APPROVED BY THE IRS.
UP TO THIS POINT, UNTIL THE IRS RELEASED THE REVISIONS, THE
CENSUS BUREAU AND THE IRS HAD AN AGREEMENT IN PLACE THAT
BASICALLY THE EXISTING CENSUS BUREAU DISCLOSURE METHODS WERE
IN FACT TREATED AS AN EQUIVALENT ALTERNATIVE.
SO BY THAT WE WERE ABLE TO USE I AM — WE WERE ABLE TO ESTABLISH COUNTS
OF ESTABLISHMENTS AND FIRMS. THAT IS SOMETHING CHANGING WITH
THE CHANGE IN THE IRS CONFIDENTIALITY REQUIREMENTS.
I’LL SHOW A BIT OF THAT IN A MINUTE.
WE HAVE WORKED ON AN AGREEMENT WHERE WE CERTIFIED BASICALLY THE
TWO DISCLOSURE METHODS AND A FORMALLY PRIVATE METHOD YOU
HEARD ABOUT FROM JOHN EARLIER. THESE ARE ACCEPTABLE EQUIVALENT
ALTERNATIVE TO THE IRS REQUIREMENTS THAT SHOW UP IN
THEIR PUBLICATION. THIS AGREEMENT WAS DONE IN
AUGUST OF ’19 AND EFFECTIVELY THE SAME AS THE PRIOR AGREEMENT
WE HAD EXCEPT FOR THE CAVEAT ON THE COUNTS.
I’LL GO INTO THOSE IN A MINUTE. WHAT’S THIS MEAN FOR THE
ECONOMIC CENSUS? THE IRS’S STIPULATION WAS COULD
NO LONGER PUBLIC COUNTS OF ESTABLISHMENTS LESS THAN THREE.
SO FOR THE ECONOMIC CENSUS, TYPICALLY WE DID NOT TREAT
COUNTS OF ESTABLISHMENT OR FIRMS AS SENSITIVE COUNTS OF ONE AND
TWO ETCETERA. .
IT’S THE MAGNITUDE DATA WE’VE ALWAYS TREATED AS SENSITIVE.
I MEAN RECEIPTS, PAYROLL EMPLOYMENT.
THOSE VALUES WITH THE RULE OF
SELF-SUPPRESSION. WITH THE ADDITIONAL CAVEAT IT
MEANS THERE’LL BE ADDITIONAL CELLS AND COUNTS OF
ESTABLISHMENTS WE’LL NO LONGER ABLE TO PROVIDE THAT WE’LL HAVE TO SUPPRESS.
AND IT ATTEMPTS TO SHOW AT A HIGH LEVEL THE APPROXIMATE
NUMBER OF ROSE IN THIS CASE IT’S — ROWS IN THIS CASE IT’S ONLY
COUNTS SUPPRESSED. SO WHAT THIS TABLE BASICALLY
SHOWS THE 18 SECTORS COVERED BY THE ECONOMIC CENSUS.
THE NEXT ROW, THE SECOND ON THE TABLE, THAT GIVES THE NUMBER OF
ROWS WHERE THE ESTABLISHMENT COUNT IS LESS THAN THREE.
AND THE FOURTH COLUMN IS THE PERCENTAGE WHICH IS COLUMN TWO
DIVIDED BY COLUMN THREE.
AND WHERE THERE’S A FEW NUMBER OF ROWS LESS THAN THREE WE PUB
THAT TO THE STATE LEVEL. YOU WON’T EXPECT A LOT OF CELLS
THAT SHOW UP WHERE HAVE YOU FEWER THAN THREE ESTABLISHMENTS
CONTRIBUTING. FOR THE OTHER ROWS, AS BILL
MENTIONED THE DATA IS PLACED TO THE ECONOMIC PLACE LEVEL.
YOU CAN SEE FOR THOSE ROWS, BECAUSE OF THE DETAIL
WE’RE PROVIDING YOU’RE EXPRESS NUMBER OF COUNTS — EXPRESSING
DID SUPPRESSING THE NUMBER OF
COUNTS AND PAYROLL WOULD BE SUPPRESSED BASED ON OUR SUPPRESSION RULES.
AND THIS LOOKS AT WHAT HAPPENS TO ZIP CODE DATA.
FOR ZIP CODES THE ONLY THING WE PUBLISH FROM THE ECONOMIC CENSUS
ARE COUNTS OF ESTABLISH MANIES BY RECEIPT CLASHES.
WE DON’T PUBLIC MAGNITUDE DATA AT THE ZIP CODE LEVEL.
IT SHOWS THE PERCENTAGE OF ROWS SUPPRESSED BY INCREASING THE
DETAIL YOU CAN SEE IN THE SECOND COLUMN.
IT GOES FROM TWO AND THREE-DIGIT WITHIN THE FIRST ROW SECTOR 44
RETAIL TRADE AND YOU CAN SEE THESE PERCENTAGES RANGE FROM
ABOUT 37% AND HEALTH CARE AND THE SMALLEST AND UP TO 82% AND
THE LEVEL FOR SECTOR 71 WHICH IS ART AND ENTERTAINMENT. AND ADDITIONAL REQUIREMENT IN
TERMS OF THE PROTECTION OF THE
ADMINISTRATIVE RECORDS FROM THE IRS.
SO WE HAVE A COUPLE QUESTIONS FOR THE COMMITTEE MAYBE OUR
DISCUSSION WILL ADDRESS OR TALK ABOUT AND THEN WE’LL OPEN IT UP.
LOOKING AHEAD TO THE 2022 ECONOMIC CENSUS, ARE THERE
EMERGING FRIENDS THAT MAY NEED TO BE RESEARCHED AND DATA
COLLECTED ON SINCE THE CENSUS COLLECTS DATA IN MANY AREAS.
WE’RE LOOKING FOR FEEDBACK ON CONTENT IN THE QUESTIONNAIRE AND
THE SECOND HAS TO DO WITH THE ISSUES AND CHANGE AND
WHAT’S THE IMPACT ON DATA WITH CHANGES IN DISCLOSURE
METHODOLOGIES AND HOW SHOULD WE INFORM USERS ON THE EFFECTS OF
THE DATA PRODUCTS?
>>OKAY. THANKS FOR THE PRESENTATION.
MORE GENERALLY TO THE ECONOMIC DIRECTOR AND OTHERS THAT ARE
INVOLVED IN THE ECONOMIC CENSUS FOR GOOD WORK.
WE DON’T TALK ABOUT THE ECONOMIC CENSUS THAT MANY AT OUR MEETINGS
COMPARED TO OTHER DATA PRODUCTS AND AREAS AT THE CENSUS SO I
FOUND THIS INFO GRAPHIC WEB PAGE ABOUT THE USES OF THE ECONOMIC
CENSUS AND THOUGHT WE’D PUT THAT UP TO HAVE MORE UNDERSTANDING OF
IT. IT HAS FIVE TYPES OF INFORMATION
OFFERED BY THE ECONOMIC SENSES ALONG WITH POSSIBLE USES.
YOU’LL NOTICE THE TWO ON THE RIGHT, THE ACCURATE BENCHMARK
FOR ECONOMIC INDICATORS AND THE CONSISTENT COMPARABLE
COMPREHENSIVE MEASURES, THOSE MAY BE DISPROPORTIONATELY
RELEVANT TO ACADEM YOU CAN USERS.
THE CHARACTERISTICS OF THE U.S. BUSINESSES AND THE INFORMATION
ON BUSINESS LOCATION AND SIZE AND THE DATA TO UNDERSTAND
BUSINESS COMPETITIVE MIGHT HAVE BROADER USE IN NON-ACADEMIC
PURPOSES. IT’S WORTH NOTING THERE ARE
THREE PRODUCTS AT THE CENSUS THAT ARE ACTUALLY CALLED THE
CENSUS. THERE’S THE DECENNIAL, ECONOMIC
CENSUS AND CENSUS IN GOVERNMENT. I’LL RETURN TO THAT.
WHY DO WE NEED AN ECONOMIC CENSUS?
THE MAIN REASON IS TO PROVIDE DATA AT A DETITLE — DETAILED
AND GEOGRAPHIC LEVEL. FOR EXAMPLE, COUNTY BUSINESS
PATTERNS INCLUDES SIX-BIG SIX-DIGIT ESTABLISHMENT AND
VALUE AND PAYROLL AND THINGS LIKE ZIP CODE WE LOSE THE
EMPLOYMENT IN THE ZIP CODE. YOU CAN SEE THE TRADE OFF IN
GEOGRAPHIC VERSUS ECONOMIC SPECIFICITY THERE.
AND THERE’S THE ANNUAL SURVEY OF MANUFACTURERS TO ALLOW NATIONAL
DATA WITH SUBINDUSTRY DETAIL BUT AT THE STATE LEVEL IT’S JUST AT
THE SECTOR LEVEL. EVERY NOW AND THEN, EVERY FIVE
YEARS IN THIS CASE, YOU HAVE TO GO AND GET GOOD INDUSTRY DETAILS
AND HAVE TO GET GOOD GEOGRAPHIC DETAIL.
AND GAINING DATA ON TOTAL ACTIVITY TO BENCHMARK THE MORE
FREQUENT SURVEYS AND TO UPDATE THE CENSUS BUREAU’S MASTER LIST OF BUSINESS.
THE PRESENTATION TALKED ABOUT THE SCHEDULE FOR RELEASES.
THERE’S SOME KEY CHANGES RELATIVE TO WHAT WE SAW IN 2012
AS I THOUGHT I’D HIGHLIGHT. I’M THE MOST CLINKY USER OF THE
FACT FINDER SYSTEM AND GLAD TO SEE THE NEW PLATFORM AT
DATA.CENSUS.gov. IT WILL MAKE MY LIFE EASIER TO
THANK YOU FOR THAT. AND THE OTHER CHANGE IS THEY’RE
USING THE PRODUCT CLASSIFICATION SYSTEM FOR DETAILED DATA ON
PRODUCT LINES. AS I SHOWED, PRODUCT LINE IS
DISTINCT FROM INDUSTRY CLASSIFICATIONS.
SO FOR EXAMPLE, JUST POKING AROUND IN THE 2012 ECONOMIC
CENSUS, ONE PRODUCT CODE WAS MEALS AND BEVERAGES PREPARED AND
SERVED AND DISPENSED FOR IMMEDIATE
CONSUMPTION AND YOU FIND IT ACROSS A RANGE OF INDUSTRIES.
SOME DISPROPORTIONATELY BUT YOU’LL FIND IT IN HEALTH CARE
AND SOCIAL ASSISTANCE AND IN RECREATION WHERE IT COMPRISES A
SMALL LIKE 6% OF SALES.
BUT MOST HAVE A PRIMARY PRODUCT CODE.
WE’RE SCHEDULED TO GET THE PRODUCT BY INDUSTRY AND VICE
VERSA IN NOVEMBER 20. AND WITH ALL THE DISCUSSION
ABOUT WHAT NOW NEEDS TO BE SUPPRESSED, CERTAINLY AT THE
LEVEL OF THE ESTABLISHMENT COUNT WE HAVE TO WONDER WHAT HAVE YOU
LEFT IN WHAT YOU’RE PRESENTING. I APPRECIATE AND WE HEARD NICK
SAY YESTERDAY THAT THERE HAD BEEN A BUDGET CUT OF 15% TO 20%
IF, I’M NOT MISQUOTING. WE HEARD CURRENT BUDGETARY RISKS
PLACE A RELEASE OF THE CONTEMPLATED PRODUCTS IN SOME
DEGREE OF JEOPARDY. THE PAPER I’M MOST PROUD OF HAD
IMPERIAL WORK ON THE 1992 ECONOMIC CENSUS.
SO I THINK YOU HAVE TO OPEN UP THE CONVERSATION ABOUT FOR A
GIVEN BUDGET WHAT DIMENSION NEEDS TO CHANGE.
AND JUST PARENTHETICALLY IF IT BECAME A ONCE EVERY 10-YEAR
CENSUS BUT NOT HAVE TO RELY ON THE IRS AND OVERLY RESTRICTIVE
CONDITION HERE SO YOU CAN GO OUT AND SURVEY THE OTHER HALF OF THE
ESTABLISHMENTS, THEN I GUESS THAT’S ON THE TABLE.
I’M NOT TRYING TO MAKE THAT DECISION FOR ANYONE BUT THAT
WOULD SEEM TO BE THE MOST NATURAL GIVEN THE NATURE OF THE BUDGETARY ENVIRONMENT.
THE PRESENTATION LEFT US WITH EYE FEW QUESTIONS NOR COMMITTEE.
THE FIRST WAS LOOKING AHEAD TO THE 2022 ECONOMIC CENSUS AND ARE
THERE EMERGING TRENDS THAT NEED TO BE RESEARCHED AND DATA
COLLECTED ON AND PUT SOME OUT FOR THE GENERAL DISCUSSION.
THE FIRST IS THE GIG ECONOMY AS A POSSIBLY GROWING SHARE OF ECONOMIC ACTIVITY.
IT INCORPORATES THE IDEA OF BIG AND ADMINISTRATIVE DATA FROM OTHER SOURCES AND THE SECOND IN
THE PRESENTATION IS THERE A WAY TO INTEGRATE DATA ON
NON-EMPLOYER BUSINESSES MORE GENERALLY?
AND THE THIRD, THIS IS PROBABLY ME
SPEAKING A LITTLE BIT OUT OF TURN BUT AS I THINK ABOUT WHAT
MIGHT BE RELEVANT TO SOMEONE WONDERING ABOUT A BUSINESS
INVESTMENT DECISION FOR A LOCAL GOVERNMENT WORRIED ABOUT ITS
ECONOMIC FUTURE OF ITS JURISDICTION, I THINK AS MUCH
ABOUT OCCUPATIONS AS INDUSTRIES. INDUSTRIES MAY BE THE ANSWER TO
THE QUESTION SO WHAT DO THEY MAKE HERE.
PARTICULARLY INDUSTRIES COMBINED WITH PRODUCT CLASSIFICATION THE QUESTION IS
WHAT COULD PEOPLE DO HERE. IMAGINE YOU’RE IN AN ENVIRONMENT
WHERE AN ENTITY IN YOUR JURISDICTION WILL RELOCATE
SOMEWHERE ELSE AND YOU’RE NO LONGER GOING TO MAKE THAT THING.
THE PROCESS OF ECONOMIC GROWTH DEPENDS ON HOW YOU CAN EMPLOY
PEOPLE AND THE SENSE OF WHAT YOU NEED TO EMPLOY IS BASED ON THE
SORTS OF SKILLS PEOPLE HAVE. SO WHEN I USE THE WORD
OCCUPATION I’M USING IT TO SERVE A PROXY FOR THAT AND THERE’S A QUESTION
WHETHER THIS IS THE RIGHT PLACE TO GET THAT SORT OF DATA AND I
HAVE MORE TO SAY ABOUT THAT. I WON’T GO THROUGH THE DETAIL OF
THIS BUT I SAID THE GIG ECONOMY. THERE ARE TWO SORTS OF COMPANIES IN
THE GIG OR SHARING ECONOMY. ONE’S UBER, ONE AIRBNB.
JUST ASK YOURSELF WHERE IN THE ECONOMIC CENSUS WOUL K — WOULD UBER SHOW UP AND
THERE’S SOME AMBIGUITY. MOST THE TIME WHEN YOU SEE
PEOPLE DRIVE FOR UBER ASK WHAT INDUSTRY THEY’RE IN IS BECAUSE
THEY NEED A STATE LICENSE AND FILL OUT SOMETHING ON THE FORM.
THIS IS THE GUIDANCE. UBER WAS TWO YEARS OLD AT THIS
POINT. COMPARE THE TOP LINE TO 2007.
THERE WAS A 5% INCREASE IN ESTABLISHMENTS.
THIS IS THE SORT OF QUESTION ABOUT ARE WE GOING TO BE ABLE TO
CAPTURE THE ECONOMIC ACTIVITY OF TAXIS WHERE IT’S A BUNCH OF
INDEPENDENT CONTRACTORS. IT’S NOT EVEN CLEAR TO ME ANY OF
THAT WOULD NECESSARILY SHOW UP IN THE ECONOMIC CENSUS.
YOU’LL GET UBER’S HEADQUARTERS HALF THE PEOPLE ASSOCIATED IN AN
ECONOMIC CONTEXT WITH UBER. I’M NOT SAYING ANYTHING IS WRONG
BUT I’M SAYING YOU ASKED FOR AN EMERGING TREND, THAT’S THE ONE
THAT IS LEFT TO MIND FIRST.
THE STATEMENT ABOUT UBER IS CAN YOU PROVIDE FOR PUBLIC USE IN
ADDITION TO WHATEVER ROLE IT MIGHT BE PLAYING IN THE
BENCHMARKING PROCESS. SOMETHING THAT INTEGRATES THE
NON-EMPLOYER STATISTICS WHICH THIS GROUP HEARD ABOUT TWICE
LAST YEAR INTO WHAT YOU COLLECT FROM THE ECONOMIC CENSUS.
AND I KNOW IT WAS ENCOURAGING TO FIND ANY EXAMPLE OF EMERGE OF
THE NON-EMPLOYER STIFTISTICS WITH THE COUNTY PATTERNS AND
WHAT ARE YOU LEARNED THE
AVAILABILITY OF THE DATA THAT’S USEFUL IN A LARGER SCALE.
AND THE THINKING ABOUT OKAY PAYINGS, THERE’S LOTS OF —
OCCUPATIONS, THERE’S LOTS OF PLACES WHERE FEDERAL GOVERNMENT
COLLECTS AND THE HEAVY LEFTING IS DONE OBVIOUSLY AT THE CENSUS,
INFORMATION ABOUT OCCUPATION AND INDUSTRY WITH SOME GEOGRAPHIC DETAIL ON IT.
THE POPULATION SURVEY DOES THAT AND THE SIP DOES THAT.
YOU CAN SAY WELL, IF YOU WERE A USER AND WANTED TO KNOW THAT,
YOU’D HAVE THAT INFORMATION. IT WOULD BE REPORT THE
INDIVIDUAL RESPOND ENT IN — RESPONDENT IN THOSE CASES.
SOMETIMES IN THE MEASUREMENT OF ECONOMIC ACTIVITY YOU HAVE BOTH
A WAY OF GETTING THE INFORMATION ON THE HOUSEHOLD AND A WAY OF
GETTING THE INFORMATION FROM THE EMPLOYER OR THE ESTABLISHMENT.
THE MONTHLY EMPLOYMENT REPORT IS A GOOD EXAMPLE THAT OFTEN
CREATES A BIT OF CONFUSION THERE’S A HOUSEHOLD SURVEY AND
AN ESTABLISHMENT SURVEY. SO THE QUESTION AND MOTIVATION I
DESCRIBED EARLIER IS WHETHER YOU WOULD EVER THINK THAT IT WOULD
BE WORTH GETTING INFORMATION FROM THE ESTABLISHMENT ABOUT
SOMETHING ON THE SKILL PROXIED HERE BY OCCUPATION OF THE
WORKFORCE THERE. TAT AN OPEN QUESTION.
I DON’T KNOW THE ANSWER IS YES. TO THE EXTENT THE ANSWER IS YES,
YOU HAVE TO ASK WHAT IS THE INCREMENTAL VALUE OF DOING THAT
ABOVE AND BEYOND WHAT’S AVAILABLE IN THE HOUSEHOLD
SURVEYS I DESCRIBED. I KNOW THE BLS HAS THE
OCCUPATIONAL EMPLOYMENT STATISTICS.
I DON’T KNOW IF THAT’S A SURVEY DONE BY THE CENSUS BUREAU ON
THEIR BEHALF. I FOUND THE SURVEY CHALLENGING
TO WORK WITH AND I THINK THERE’S DISCLAIMERS ON THE WEBSITE
SAYING NOT RELIABLE FOR COMPARISONS OVER TIME.
I’M A POTENTIAL USER INTERESTED IN THAT AND IN MY FINAL SLIDE I
ALLUDED TO EARLIER, FOR THE PURPOSES OF DESCRIBING WHAT IS
LOST. PERCENTAGE OF ROWS SUPPRESSED IS
A PORTION BUT SOMETHING ABOUT THE PERCENTAGE OF REVENUE OR SHIPMENTSSHIPMENTS WEIGHTING
THAT INFORMATION WOULD BE HELPFUL AND FOR THE DISSEMINATION QUESTION YOU HAVE
TO LET THE FOCUS GROUP GUIDE YOU AND THAT IS WHERE I WOULD START.
I WOULD SAY THAT IF YOU’RE GOING TO BE
SUBJECT TO THE IRS RULES RELYING ON THE ADMINISTRATIVE DATA AND
THEY CALL THE TUNE AND THE BUDGETARY ENVIRONMENT IS NOT
SUCH YOU CAN CONTEMPLATE GETTING ALL OF IT, MORE OF IT
YOURSELVES, THEN MAYBE THERE’S A DIFFERENT ROLE FOR THE CENSUS
BUREAU TO PLAY. I THINK YOU’RE GOING TO LEARN
WHEN YOU ABIDE BY THE PROCEDURES FOR THE 2017 WHICH PARTS OF THE
USER COMMUNITY ARE IN NEED OF ASSISTANCE.
YOU ARE IN THE POSITION OF SEEING WHAT
NEEDS PEOPLE HAVE. DISASTER RECOVERY OR THE LIKE,
YOU ARE DATA EXPERTS IN ADDITION TO THE DATA IN ADDITION TO DATA
COLLECTORS AND DISSEMINATORS. AND I SUSPECT YOU’LL DISCOVER
YOU HAVE SOME VALUE TO THE USER COMMUNITY THERE.
LET ME PAUSE THERE AND SAY THANK YOU TO THE TEAM FOR THE GOOD WORK THEY’RE DOING.
>>THANK YOU, ANDREW. WAS THERE SOMETHING YOU NEEDED
TO RESPOND TO FIRST OR PRN IT UP TO OTHER PER SHARE OPEN IT UP TO
OTHER QUESTIONS? I SEE JOE HAS HIS TENT UP.
I THINK THAT’S LEFT OVER. COOL.
ONE PABLO. — JUAN PABLO.
>>FROM THE UNIVERSITY OF IOWA I HEARD THE HAWK EYE STATE IS
RANKED NUMBER ONE. GOOD TO HEAR.
I ALSO HAVE THE GIG ECONOMY LIKE ANDREW HAD TO ANSWER YOUR
QUESTION. I THOUGHT IF ANOTHER AREA THAT
MAYBE YOU’RE CAPTURING ALREADY, I DON’T KNOW, THAT’S REMOTE
WORK. PEOPLE ESPECIALLY IN THE
KNOWLEDGE ECONOMY WHERE THE ESTABLISHMENT MAY BE IN ONE
PLACE BUT THEY’RE PHYSICALLY LOCATED ELSEWHERE, SOMETIMES
THERE MAY BE U.S. CITIZENS WHO MAY LIVE ABROAD AND DO WORK
REMOTELY. I THINK IT WILL INCREASE.
IT MAY BE INTERESTING TO THINK ABOUT HOW TO CAPTURE THAT
INFORMATION IF IT’S NOT ALREADY BEEN CAPTURED.
THEN I HAVE ONE IDEA GIVEN THE BUDGET CONSTRAINT BUT IT’S TO
PERHAPS GIVE BACK TO THE ESTABLISHMENTS THAT RESPONDED
AND MAYBE SOME WHO DIDN’T RESPOND.
AND WHEN YOU RELEASE THE DATA FOR 2017, GET BACK TO THEM AND
SAY HERE’S SOME THING RELATED TO YOUR SECTOR OR YOUR GEOGRAPHY.
SO IN A WAY GIVING THEM VALUE FOR THE EFFORT THEY’VE PUT IN TO RESPOND.
OR PERHAPS WITH A LITTLE ENCOURAGEMENT SINCE 2022 IS
AROUND THE CORNER HOPEFULLY THEY’LL REMEMBER THE VALUE.
AGAIN, MAYBE NOT GOING TO HAPPEN BECAUSE OF BUDGET CONSTRAINTS
BUT WE COULD PUT THAT ONE IN THE WOULD BE NICE COLUMN.
THAT’S ALL I HAVE.
>>THANK YOU.
ANYONE ELSE? THAT’S FINE BECAUSE WE HAVE TO
GO OVER OUR RECOMMENDATIONS. WE CAN ALWAYS END THE DISCUSSION
EARLY IF FOLKS DON’T HAVE ADDITIONAL QUESTIONS OR
COMMENTS. ARE WE GOOD?
YOU DIDN’T NEED TO RESPOND TO WHAT JUAN PABLO SUGGESTED?
>>ONE QUICK RESPONSE ON THE COMMENT WITH THE REMOTE WORKERS.
THE WAY WE PROBABLY ARE CAPTURING IT AND IT’S PROBABLY
AN UNKNOWN UNKNOWN, THEY ARE PROBABLY BEING CAPTURED IN THE
LOCATION THEY REPORT BACK TO AND NOT AT THEIR ACTUAL INDIVIDUAL
ADDRESS. SO IF THEY’RE AN EMPLOYEE OF THE
COMPANY AND REPORT BACK TO THE OFFICE THAT EMPLOYMENT VALUE
PROBABLY SHOWS UP IN THE REGIONAL OFFICE.
>>OKAY. SO WE CAN GRAB OUR LUNCH AND
COME BACK AND GET GOING. DO WE NEED TO DO ANYTHING ELSE.
>>TOMMY WRIGHT. ONE EDITORIAL.
THANK YOU WILL AND BILL AND ANDREW AND THANK YOU VERY MUCH FOR THOSE PRESENTATIONS.
THIS ENDS THEIR SESSION AND WE’LL GRAB LUNCH.
>>SO WE’RE 15 MINUTES EARLY AND ANDREW RIGHTLY POINTED OUT WE
DON’T HAVE A LOT OF TIME TO DISCUSS RECOMMENDATIONS SO I
RECOMMEND EVERYBODY GET THEIR LUNCH, COME BACK AS SOON AS YOU
CAN WE’LL START TO COMPILE THINGS.
KATHY.>>ANYBODY WHO IS WORKING ON OR HAS RECOMMENDATIONS ON THE
CENSUS DATA PRODUCTS GO TO KATHY AND KRISHNA HE’S DOING THE
DISCLOSURE AVOIDANCE. ANYTHING ELSE FOR ANDREW, YOU
CAN ALSO IF YOU’VE BEEN WRITING IT YOU CAN E-MAIL IT TO EACH OF
THEM AND WE’LL TRY TO COMPILE THAT QUITE QUICKLY SO WE CAN
THEN START REVIEWING. I’VE ALREADY E-MAILED OUT TO
EVERYBODY SO MAYBE YOU CAN VIEW ON YOUR PHONES ONCE YOU GOT YOUR
INPUT YOU CAN VIEW ON YOUR PHONES WHAT WE COMPLIED
YESTERDAY TO START THINKING ABOUT HE HAD ITS AND WE WANT —
EDITS AND WE CAN EDIT ON CONTENT NOT STYLE AND THE CONSEQUENCES
OF NOT FINISHING BY 2:00 ARE THAT WE WILL HAVE TO WAIT SIX
WEEKS AND HAVE A VIRTUAL PUBLIC MEETING AND THEN THAT SLOWS DOWN
OUR GETTING THE SPEEDY.
THIS IS NOT A DOCUMENT THAT HAS TO GO THROUGH ALL KINDS OF LEGAL
REVIEW. IT’S JUST OUR RECOMMENDATIONS
AND EXPERTS HAVE ALREADY PUT A LOT OF GOOD THOUGHT INTO THEM.
SO PLEASE KEEP THAT IN MIND. OKAY.
GRAB YOUR LUNCH. YOU ALL ARE SO GOOD.
HAVE YOU HAVE ANYTHING FROM KRISHNA ON DISCLOSURE AVOIDANCE
AND DEBORAH E-MAILED SOMETHING AND KRUNAL HAD SOMETHING AND
ANDREW HAS THE ECONOMIC CENSUS. LET’S TRY TO FINISH THOSE UP IN 15 MINUTES.
>>SO ANDREW AND KRISHNA YOU CAN E-MAIL IT AT GMAIL.com.
>>WE GOT 10 PAGES TO COVER.
>>WHILE KRISHNA IS FINISHING THAT UP I’LL EXPLAIN SOMETHING TO TO THE NEW FOLKS.
THIS IS AN UNUSUAL MEETING WHERE WE DIDN’T HAVE DISCUSSANTS FOR A LOT OF TOPICS.
I ASKED THEM TO PREPARE THOUGHTS AND COMPILE THE INFORMATION BUT
IN MOST MEETINGS WE USUALLY HAVE A DISCUSSANT AND WHAT I
CERTAINLY HEARD FROM SEVERAL FOLKS IS THE AMOUNT OF TIME WAS
VERY BRIEF. OBVIOUSLY, WE HAD A LOT TO
DISCOVER BUT IN THE FUTURE WE WANT TO GO BACK TO HOW WE
USUALLY DO IT WHERE WE HAVE A DISCUSSANT AND WE HAVE MORE
DETAIL WE GET THE MATERIALS IN ADVANCE SO WE WANT TO NOT SET A
NEW BASED ON THIS MEETING THAT’S NOT PRODUCTIVE OR NOT AS
PRODUCTIVE AS WE’D LIKE OR USEFUL AS WE’D LIKE.
THE IDEA IS WE COME HERE TO HAVE A DOES DISCUSSION AND IT SHOULD HAVE GOOD INFORMATION
BEHIND IT AND WE SHOULD HAVE DISCUSSION.
IF THE CENSUS BUREAU HAS SOMETHING THEY’D LIKE TO INFORM
US ABOUT AND DON’T NEED AS MUCH TIME FOR DISCUSSION THERE OTHER
OPTIONS THE ONE-WAY BRIEFINGS. WE CAN MOVE ITEMS LIKE THAT TO
ONE-WAY BRIEFINGS. WE STARTED WITH LANGUAGE ABOUT REAFFIRMING THAT WE WOULD LIKE
TO HAVE SUFFICIENT TIME FOR DISCUSSION.
THE AGENDA ITEM HAS TO HAVE MORE THAN 30 MINUTES AND 30 MINUTES
OF THE SESSION AND THE DISCUSSANTS SHOULD HAVE TIME IN
EACH SESSION AND WE’D LIKE MATERIALS IDEALLY TWO WEEKS IN
ADVANCE. TO AFFIRM WHAT WE’VE DONE IN THE
PAST AND HOW GREAT IT WAS WHEN WE HAD DETAILS.
THAT WAS FANTASTIC.
WE GAVE DEMOGRAPHIC ANALYSIS AND TABLES
>>AND PROVIDE SCIENTIFIC INPUT. THAT’S THE INTENT OF THOSE FIRST
COUPLE OF PARAGRAPHS AND THE BACKGROUND BEHIND THEM.
DO THOSE LOOK GOOD TO FOLKS. ANY FRIENDLY AMENDMENTS? AND I WANT TO RAISE HANDS AND BE
ALL IN AGREEMENT BUT IF YOU HAVE ANY EDITS NOW’S THE TIME IDEALLY
TO SAY IT AND WE CAN SCROLL DOWN AND WHEN KRISHNA COMES BACK WE
CAN — OH, IT’S NOT BEING PROJECTED SOMEHOW.
>>IT JUST TAKES A MINUTE.>>OKAY.
HOPEFULLY YOU’VE HAD A CHANCE TO LOOK THIS OVER ON YOUR iPHONES.
AGAIN, LOOKING FOR CONTENT RATHER THAN STYLE.
WE HAVE 10 PAGES TO GO THROUGH.
ANYONE HAVE ANY EDITS HERE?>>TOMMY WRIGHT.
SOMETIMES WHEN PEOPLE SEE THE WORD CENSUS WE’RE NOT SURE IF
IT’S THE DECENNIAL AREA WHICH IS ONE PIECE OF THE CENSUS BUREAU
OR THE CENSUS BUREAU. SOMETIMES PEOPLE SAY IF YOU MEAN
THE ORGANIZATION SAY CENSUS BUREAU.
IT’S JUST A COMMENT. IT REMOVES AMBIGUITY. WE DON’T KNOW WHERE THERE’S A
DELAY. NOW I NEED HELP.
I DON’T KNOW WHAT TO DO.
>>IT JUST CAME BACK WHATEVER MAGICAL THING YOU DID.
>>THIS IS RICHELLE WINKLER AND ASKING CLARIFYING QUESTIONS ON
PROCESS SINCE I’M NEW. YES, WE USE OUR MICROPHONES AND
THE WAY WE PRESENT THESE RECOMMENDATIONS LIKE THINKING
ABOUT THE DRAFTED ONES NOR —
FOR THE UPDATE ON 2020 CENSUS IF CSAC HAS QUESTIONS AND DO WE
EXPECT THEM TO RESPOND TO THE QUESTIONS THE WAY IT’S WRITTEN
LIKE THIS?>>AND THEY WILL.
WHICH IS AWESOME.>>IF YOU THINK IT’S NOT PHRASED
CORRECTLY YOU SHOULD SPEAK UP OR IF YOU DON’T THINK WE’LL GET THE
ANSWER BASED ON HOW IT’S PHRASED.
NOT STYLISTIC BUT IF THERE’S A MEANING QUESTION.
>>KATHY, WHY DON’T YOU TAKE THIS.
THE HIGHLIGHTED NUMBER, SOMEBODY HIGHLIGHTED BECAUSE THEY
PROBABLY WEREN’T SURE WHAT THAT PERCENT WAS GOING TO BE.
DOES ANYBODY KNOW.>>JEFF HERE.
THE ESTIMATED NUMBER FOR THE SELF-RESPONSE.
>>OKAY. A BIT OF A DELAY.
>>AND IS THERE A DIFFERENCE BETWEEN OUR SCROLLING AND PROJECTING?
>>CAN THE TECHNOLOGY PEOPLE HELP US.
WE’RE HAVING A DELAY. DOES KATHY NOT NEED TO BE ON THE
MICROPHONE? THIS IS REALLY GOING TO MESS US
UP. WE’RE CONTINUING TO HAVE THIS PROBLEM.
>>ON THE QUESTIONS TRYING TO MAKE SURE I
HAVE THE WRITE NUMBERS IN MY MIND.
QUESTIONS TWO AND THREE, IT’S UNCLEAR TO ME HOW THEY’RE DIFFERENT.
>>THE CURRENT FIELD WAS PRESENTED AT THE CURRENT INFIELD
FAILURE RATE. THAT ONE?
>>THE LAST QUESTION IS OF THE 70% THAT WAS DONE IN-HOUSE, WHAT
WAS THE FAILURE RATE AND THEN THE NEXT ONE IS CSAC WAS LIKE TO
KNOW THE RESULT OF THE ACCURACY ASSESSMENT OF THE IN-OFFICE WORK
AND WOULDN’T THAT ENCOMPASS THE FAILURE RATE?
>>THANK YOU FOR CATCHING THAT. I HAVE AN EDIT FOR 3D.
D LIKE DAVID.>>THERE’S TWO NUMBERED THINGS
BUT YES.>>WHERE IT SAYS IN OTHER WORDS
COULD YOU SHOW CSAC BUT IT SAYS YOUR SHOW.
>>WHAT?>>3D.
IT SHOULD SAY COULD YOU SHOW CSAC INSTEAD OF COULD YOUR SHOW.
>>GOT IT.
SO THERE’S 1 THROUGH 5 IN THE DECENNIAL PIECE AND THEN THERE’S
A PARAGRAPH SAYING IT’S ALSO ENCOURAGING AND THEN THERE’S
ANOTHER LIST OF 1, 2 AND 3 AND 3 HAS MANY DIFFERENT PARTS.
ARE WE GO GOOD TO THE HIGHLIGHTED STATEMENTS?
>>I HAVE ONE MORE REMARK ON THE FIRST NUMBER 4.
ARE THERE LESSONS LEARNED.
IT STRIKES ME AS AN INEFFECTIVE QUESTION.
ARE WE LOOKING FOR A YES NO RESPONSE OR WHAT DO WE WANT TO KNOW?
YES, WE WOULD LIKE TO KNOW.
>>THIS IS JEFF. I THINK YOU CAN JUST CHANGE THE
WORDING.>>WILL ABOUT LESSONS LEARNED.
>>ANYTHING ELSE BEFORE THE HIGHLIGHTED SENTENCE?>>SO THE HIGHLIGHTED SENTENCE
SAYS THESE BREAK SHOULD HELP TO GUIDE THE BUREAU’S EFFORT FOR
USABILITY OF THE ISR FOR HARD-TO-COUNT GROUPS.
>>JUAN PABLO, IS THAT THE INTENT OF THE BREAKDOWNS.
UNDER 3 OR YOU CAN READ IT.>>THIS IS RICHELLE.
I HAVE A COMMENT ON THE SECTION. MY MEMORY OF THE RESPONSE WHEN
JUAN PABLO RAISED THESE QUESTIONS I THOUGHT THERE WAS AN
INDICATION THERE WOULD BE PRODUCING DETAILED ANALYTICAL
REPORT AND PRODUCING IT WITH A LOT OF INFORMATION.
MAYBE WE WANT TO ASK FOR A SEPARATE RESPONSE OR THE REPORT.
>>YEAH, SO TIMING IS A THING. WE SOMETIMES DONE GET THE
RESPONSES TO OUR QUESTIONS UNTIL JUST BEFORE OUR NEXT MEETING.
AND GIVEN THAT’S LITERALLY BEFORE CENSUS DAY IS PART OF WHY
IT’S NOT SO MUCH OUR GETTING THE ANSWERS TO THESE QUESTIONS AS IT
IS THEM UNDERSTANDING TO PLEASE USE THESE BREAKDOWNS TO INFORM.
THAT WAS A GOOD QUESTION. SO DOES THIS SENTENCE WORK? GREAT.
THANK YOU. EXACTLY WHY I WANTED TO INSERT
IT. NAY — THEY HAVE A PUBLIC
PRODUCT THEY CAN POINT US TO THAT POINT.
WE’RE SCROLLING DOWN. ANOTHER SECTION 1 THROUGH 8.
THERE’S A COUPLE QUESTIONS THAT END.
IF ANYONE HAS COMMENTS.>>THE OTHER CSAC QUESTION
RELATED TO THE 2020 UPDATE.>>THIS IS RICHELLE AGAIN I
THINK I KEEP MOVING US BACKWARDS A LITTLE BIT.
>>I’M GLAD SOMEONE’S PAYING ATTENTION.
>>SO IF I’M LOOKING AT THE REPORT OF THE 2018 END TO END
TEST IN SEVERAL PLACES WE MENTIONED BY DEMOGRAPHIC GROUP.
DO WE REALLY MEAN RACE ETHNIC GROUP.
IS THAT WHAT THEY’RE GOING FOR? IN MY MIND IT SEEMS AGE WOULD BE
A BIG FACTOR IN TERMS OF THINKING ABOUT LIKE INTERNET
RESPONSE VERSUS PAPER RESPONSE FOR INSTANCE.
>>I DON’T KNOW IF WE NEED TO CLARIFY THAT.
IT’S VAGUE TO ME.>>I’D LIKE RACE AND AGE.
HOW ABOUT AFTER THE FIRST ONE WE PUT RACE, ETHNICITY, COMMA, AGE.
ANY OTHERS?>>THIS IS JUAN PABLO.
I LIKE THE EXTRA PHRASE SENTENCE THAT CLARIFIES OUR INTERTENT TO
HELP WITH IDENTIFYING ISSUES THAT MAY BE FACED BY HARD TO
COUNT COMMUNITIES AND THAT’S THE INTENT.
SO I THINK ANYTHING WOULD HELP YOU IDENTIFY THOSE HARD TO COUNT
COMMUNITIES WITH DEMOGRAPHICS WILL BE USEFUL.
>>SO RACE, ETHNICITY, AGE, COMMA, ETCETERA.>>THANK YOU.
I HAVE MOVED THAT GENERAL STATEMENT UP TOP AND PUT RACE,
ETHNICITY, AGE, ETCETERA TO DESCRIBE IT.
THANK YOU, RICHELLE FOR CATCHING THAT.
>>LOOK DOWN THROUGH ITEMS 1 THROUGH 8.
>>THIS IS JEFF. I HAVE A COMMENT ON THE LAST TWO
CHRISTEN’S QUESTION ABOUT INCARCERATED.
I BELIEVE CATHY FOUND LINKS SHE’S GOING SHARE WE COULD
REMOVE THOSE I THINK.>>THEY’RE COUNTED IN THE PRISON
WHERE THEY ARE INCARCERATED.
I’LL SEND YOU DETAILED DOCUMENTATION ON THE OPERATIONS.
>>WHEN THEY’VE ANSWERED WE TRY NOT TO DOCUMENT IT ALL.
OKAY.>>I ADDED AFTER ALLISON SENT
THIS. I THINK IT WAS AL MENTIONED BUT
THEY OFFERED A DEMONSTRATION OF THE INTERNET SELF-RESPONSE FORUM
AND I ADDED A SENTENCE THAT WE WOULD APPRECIATE THAT.
THE NEXT IS THE COMMUNICATION.
>>ON THE LINE ABOVE THE QUESTIONS SHOULD SAY INTEGRATED
COMMUNICATIONS PROGRAM.>> YES, THANK YOU.
>>IT WASN’T WORKING SO YOU CAN TAKE IT OUT.
>>AND WE’RE CONSULT THE OTHER GROUPS TO GET MESSAGING OUTSIDE
THE CENSUS BUREAU AND I WAS GOING TO ADD
LIKE A CONTINUED OR CONTINUING TO WORK WITH GROUPS TO RECOGNIZE
THEY HAVE BEEN DOING THAT ALREADY.
>>THIS IS NOT USUALLY QUITE THIS PAINFUL.
>>AND ON NUMBER 4.
I DON’T SEE IT HIGHLIGHTED IN THE VERSION I HAVE BUT I THINK
THE VERSION I HAVE IS SOMETHING YOU E-MAILED.
>>WHAT IS THE PROCESS.>>READ US WHAT YOU’RE LOOKING
AT.>>IS THIS JUAN PABLO I THINK
THAT’S THE QUESTION ANSWERED THIS MORNING.
A QUESTION FROM YESTERDAY AND THIS MORNING WE GOT AN ANSWER TO
IT.>>THE ONE WHAT IS THE PROCESS
FOR DETERMINING SPECIFIC MEDIA APPROACHES TO REACH.
>>WE KNOW THERE ARE SPECIFIC PROGRAMS TARGETING PEOPLE ON
RESERVATIONS AS WELL. WHO HAD QUESTION?
WE CAN MAKE SURE YOU GET THE INFORMATION.
IT MIGHT HAVE BEEN SOMEBODY NO LONGER HERE.
>>THIS IS JUAN PABLO. IT WAS ALMOST A 5% UNDERCOUNT IN
2010 AND IT WAS NOT MENTIONED AT ALL IN THE PRESENTATION.
>>I SAID CSAC WOULD LIKE INFORMATION ABOUT SPECIFIC PROGRAMS.
GOT IT. ANYTHING ELSE ON COMMUNICATIONS? EXCELLENT.
MOVING TO THE PARTNERSHIP PROGRAM.
ALL RIGHT. THE PARTNERSHIP PROGRAM.>>THIS IS KEN, ON 4 ON
PARTNERSHIP THE FIFTH LINE THE AMERICAN ASSOCIATION OF
PEDIATRICS IS THAT THE ACADEMY? NOT SURE WHAT THIS MEANS?
>>I DON’T KNOW THAT GROUP.>>SO THIS IS HIGHLIGHTED
BECAUSE I THINK FOR TWO REASONS, ONE, I THINK THEY MENTIONED
THEY’RE DOING A BUNCH OF THIS AND TWO BECAUSE THEY HAVE
OFFERED TO DO A ONE-WAY BRIEFING FOR US ON THE UNDERCOUNT OF
YOUNG CHILDREN AND WHAT THEY’RE DOING TO ADDRESS IT.
SO IF YOU ALL WOULD LIKE, THAT ONE-WAY BRIEFING WOULD ADDRESS
THESE QUESTIONS AND RECOMMENDATIONS WHICH I THINK
WE’LL FIND MOST OF WHAT WE’RE DOING.
WOULD IT BE OKAY TO EDIT THIS DOWN TO SAY CSAC IS CONCERNED
ABOUT THE UNDERCOUNT OF YOUNG CHILDREN.
CSAC WOULD LIKE A BRIEFING ON THE PROGRAM AND EFFORTS TO
ADDRESS IT, PERIOD. WOULD THAT WORK?>>I HAVE AN EDIT FOR ITEM 4.
>>HOLD FOR ONE SECOND. I’M JUST COPYING THE DISCLOSURE STUFF.
I’M PASTING IT IN.
>>UNDER ITEM UNDER 4 IN THE PARTNERSHIPS SECTION.
THERE’S A MENTION OF THE AMERICAN ASSOCIATION OF
PEDIATRICS AND I THINK IT MEANT THE AMERICAN ACADEMY OF
PEDIATRICS.>>WE ELIMINATED THAT SENTENCE
SO THAT SOLVES TA PROBLEM. ANYTHING ELSE
ON PARTNERSHIP?>>THE CALL FOR EVALUATION
TALKED ABOUT IN THE CORNER.
>>SO GREAT. LAST CALL FOR PARTNERSHIP?
MOVING ON TO EVALUATION.>>IS THERE ANYBODY WHO DOESN’T
HAVE ACCESS OR ISN’T CONCERNED ABOUT IT?
OKAY. EVALUATING THE CENSUS.
>>SO THIS PARAGRAPH HIGHLIGHTED, THERE’S THINGS BIG
MOLD ELLING AND ENUMERATION AND THERE’S NO LEGAL IMPLICATIONS
BUT WE PROBABLY SHOULD STICK WITH SCIENTIFIC QUESTIONS SO IF THAT’S OKAY.
>>THIS STRUCK ME DURING THE PRESENTATION AND THEY MENTIONED
AN EVALUATION PROGRAM IS LOOKING AT STUDENT POPULATION.
THIS IS RICHELLE. AND LOOKING AT HOUSING STUDENTS
LIVE IN NOW AND THEY MENTIONED BOTH THE MAIN CONCERN WAS THE
PRIVATE OFF-CAMPUS DORM-LIKE BUILDINGS.
AND IN LEARNING COMMUNITIES UNIVERSITIES ACTUALLY OWN WHAT LOOK LIKE SINGLE-FAMILY HOUSE
MAY HAVE FIVE, 10, 20 STUDENTS AND SOMETIMES THEY’RE OFF CAMPUS
OR BORDERS OF CAMPUS. I DON’T KNOW IF THAT’S AN EXTRA
CATEGORY THEY MIGHT CONSIDER ADDING TO THAT RELATION.
>>HOW ABOUT CSAC SUGGESTIONS CONSIDERING ADDING.
>>IT’S A NEW GROUP CATEGORY.>>FOR STUDENT HOUSING.
>>AND GIVEN THE RISING TREND TOWARDS UNIVERSITY LEARNING
COMMUNITIES. THERE’S A WHOLE LITERATURE
AROUND THAT. THAT’S A PHRASE THEY MAY BE ABLE
TO USE TO FIND OUT MORE.>> EXCELLENT.
THANK YOU.>>OKAY.
THIS IS SOMEBODY ASKING ABOUT THE DETAILED OPERATION PLANNED
FOR THE POST ENUMERATION SURVEY.>>WE ACTUALLY HAVE A QUESTION DOWN BELOW ON THAT.
>>I’M JUST GROUPING STUFF. ANYTHING ELSE ON EVALUATION?
>>I THINK MAYBE PETER WROTE THIS.
DID HE MEAN TO SAY 2020 OR 2030?>>THIS IS JAY.
WHAT HE WAS SAYING IS THAT CENSUS HAS VARIOUS OPERATIONAL
ASSESSMENTS THEY’RE GOING TO USE TO TRY TO PREDICT WHAT HAPPENS
IN 2020. THIS IS ALSO INFORMATIVE FOR
2030 AND ONE WAY IT’S INFORMATIVE IS NOT ONLY
PROVIDING BASELINE BUT WHAT KINDS OF PREDICTIONS WERE YOU GOOD AT.
>>2020 IS CORRECT.>>I THOUGHT IT MIGHT BE.
>>ALSO THE QUALITY OF THE BUREAU’S ABILITY TO PREDICT
THESE METRICS.>>ANYTHING ELSE ON EVALUATION?
LAST CALL. DEMOGRAPHIC ANALYSIS.
>>THERE’S GOING TO BE SOME UNIQUE MATERIAL TO READ IN JUST A SECOND.
RICHELLE DID SUCH A GREAT JOB.>>ANYTHING ELSE IN THE
DEMOGRAPHIC ANALYSIS? ON THE PRODUCTS, I THINK —
>>WE’LL HAVE TO LET PEOPLE READ IT.
LET ME GIVE YOU THIS RECOMMENDATION.
I JUST GAVE THE RECOMMENDATIONS THEY ASKED AND GAVE DURING THE PRESENTATION.
>>DOES THIS SEEM REPRESENTED IN THE DISCLOSURE RELATED TO THE SECTION.
DISCLOSURE AVOIDANCE AND ABOUT THE DHC EFFECTIVELY SO MAYBE WE
CAN CONSOLIDATE.>>THANK YOU.
I’LL PUT THAT UP AND LET PEOPLE READ IT AGAIN.
HOLD ON.>>SO I THINK MORE INFORMATION
WITH THE TABLES THEY PROVIDED THE DETAILS OF THE TABLES.
>>THIS WAS PROVIDED ALREADY.>>THIS IS JUAN PABLO.
YOU SAID TABLE OR MEMO.
>>A FULL ARRAY OF ALL THE TABLES.
>>AND FOR DHC WE COULD SAY CSAC RECOMMENDS AND THE DHC
INFORMATION IS PUBLICLY PROVIDED AS IT RELATES TO CHILDREN AND
YOUTH. JUST TO MAKE SURE THEY’LL DO IT
FOR ANY FUTURE PRODUCTS.
>>HERE’S ONE PABLO. IN THE LAST SENTENCE OR DATA FOR CHILDREN.
>>ABOUT THE HOUSEHOLD AND CLARITY.
>>CLARITY OF DATA.>>AT WHAT LEVEL OF GEOGRAPHY OF
DETAILED DATA ABOUT CHILDREN BE TABULATED WITHIN INFORMATION
ABOUT THE HOUSEHOLD.
>>IS THERE A SECOND SENTENCE OR SOMETHING?
>>CHRIS, WE NEED TO INTERRUPT YOU.
>>THE SECOND HALF WE’RE STRUGGLING WITH.
>>WELL.>>YOU MEAN THE RELIABILITY OF THE DATA.
>>YOU CAN SAY LOW LEVELS.
>>THIS IS RICHELLE. I THINK THEY HAVE ALREADY
ANSWERED THAT QUESTION SO AT WHAT LEVELS OF GEOGRAPHY WILL
DETAILED DATA ABOUT CHILDREN BE TABULATED WITH INFORMATION ABOUT
THE HOUSEHOLDS THAT IS INCLUDED IN THAT SPREAD SHEET.
>>I HAVEN’T BEEN THROUGH THIS BUT I’M SEEING THE DOCUMENT OF
SPREAD SHEET BUT THAT’S THE CRITICAL THING THE
CHARACTERISTIC BY THE HOUSEHOLD CHARACTERISTICS BECAUSE
APPARENTLY THEY WON’T ALWAYS BE DONE TOGETHER.
MAYBE IT WILL BE SHOWN HERE BUT I HAVEN’T SEEN IT SO FAR.
>>INFORMATION ABOUT THE DETAILS OF THE FUTURE TABLES OR FUTURE PRODUCTS
>> I THINK THAT’S GOOD BECAUSE YOU’RE SUGGESTING A MEMO AND
THEY HAVEN’T PROVIDED A MEMO AND THE SPREAD SHEET HAS USABILITY DIFFICULTIES, OBVIOUSLY.
A MEMO WOULD BE HELPFUL.
>>WE RECOGNIZE THE DHC IS PUBLIC.
>>THE LAST ONE IS ONE PABLO’S RECOMMENDATION ABOUT REACHING
OUT TO THE DATA SCIENCE COMMUNITY.
ALL RIGHT. DISCLOSURE LIMITATION.>>THIS IS JOHN
>>VERY MINOR POINT THE FONT IS NOT THE SAME AS WITH YOU WERE USING.
>>I THINK THIS PIECE IS ABOUT THIS RESEARCH CENTERS WE HAD A
FEW QUESTIONS ABOUT.>>IT’S KEN SIMONSON.
THE CITIZENSHIP QUESTION MAKING A RECONSTRUCTION ATTACKS.
IS ATTACKS SUPPOSED TO BE SOME OTHER WORD?
>>MY GUESS IS IT’S NOT SUPPOSED TO BE PLURAL.>>WE’LL ASK FOR SPELL CHECK.
THE ZIT CITIZENSHIP QUESTION ONE.
>>IT’S NOT EVEN A QUESTION ABOUT DATA.
OKAY. WE HAVE 10 MINUTES LEFT.
RECOMMENDING THE FEDERAL REGISTER AND A PUBLIC PLAN THIS IS JUAN PABLO’S
PIECE ABOUT SCIENCE AND DATA PRIVACY.
>>LET’S SEE IF SOMEBODY WANTED THAT MORE.
>>LET ME ASK A QUESTION. WHAT’S THE SENSITIVITY ANALYSIS?
THIS IS JOE WHITLEY ASKING THE QUESTION.
>>YOU SEE THE HIGH AND LOW OF WHAT MIGHT COME OUT IN TERMS OF DIFFERENCE.
>>ALL RIGHT. I’LL TRY TO ZIP THROUGH THE
ECONOMIC CENSUS. THIS IS WITH THE GIG ECONOMY.
DO YOU WANT TO WALK US THROUGH?
>>SO I DIDN’T GET MUCH OF A RESPONSE TO THE VARIOUS POINTS I
RAISED IN RESPONSE TO THEIR QUESTIONS.
SO THERE’S BASICALLY A PARAGRAPH ON EACH ONE AS WELL AS SOMETHING
THAT I DIDN’T HAVE IN THE SLIDES BUT THE QUESTION ABOUT THE
AFFECT THAT HAVING TO RELY ON THE IRS FOR ADMINISTRATIVE
RECORDS FOR CLOSE TO HALF THE SAMPLE OR HALF THE POPULATION
RATHER IS HAVING. HAVING SEEN ALL THESE THINGS IN
THE SENSE THEY WERE ON THE SLIDES.
THERE SHOULD BE FOUR PARAGRAPHS BASICALLY.
>>I THINK WE HAVE A SECTION FROM KEN ON CONSTRUCTION
SPENDING SUGGESTING A WORKING GROUP.
>>RIGHT.>>IT’S A FORMAL FOR THE NEW
PEOPLE. A FORMAL BODY AND PROCESS TO
HAVE A SUBCOMMITTEE GO IN DEPTH WITH THE CENSUS BUREAU ON ROUND
OF TOPICS.>>RIGHT.
>>I WAS SUGGESTING THIS AS A WAY TO ENCOURAGE AND SUPPORT THE
BUREAU’S EFFORTS TO IMPROVE THE CONSTRUCTION DATA PARTICULARLY
THE CONSTRUCTION SPENDING DATA AS I MENTIONED IN MY REMARKS
YESTERDAY. THEY HAVE BEEN TALKING ABOUT
THIS AND MAKING EXPLORATORY EFFORTS FOR SOME TIME BUT STILL
HAVEN’T SEEN RESULTS AND ONE OF THE VALUES OF THE ADVISORY
COMMITTEE IS TO GIVE SUPPORT TO THINGS THEY WOULD LIKE TO BE
DOING INTERNALLY BUT MAYBE ARE NOT GETTING ENOUGH FEEDBACK FROM THE OUTSIDE.
AND ALLISON HAD SOMETHING TO ENCOURAGE THE CENSUS BUREAU TO
RESPONSE TO THE THREE WRITTEN PUBLIC COMMENTS WE RECEIVED.
WHO HAS SOMETHING THEY NEED TO READ.
RAISE YOUR HAND IF YOU NEED TO READ IT MORE CAREFULLY.
SOME SESSION. FEEL FREE.
>>THIS IS JOHN. DID WE LOSE THE COMMENT ON THE POST-ENUMERATION SURVEY.
>>I WANT TO GO TO MY DEMOGRAPHIC ANALYSIS QUESTION OF ONE PARAGRAPH.
>>HERE’S THE PARAGRAPH ON POST ENUMERATION VARE.
— SURVEY. ONE ASKED WHEN THE OPERATIONAL
PLAN WILL BE THERE AND YOU HAD A CORRECTION?
>>THIS ISN’T DEMOGRAPHIC.>>THE MEDICARE PARAGRAPH.
>>YES >>THIRD LINE SHOULD HAVE
DUPLICATE REPORTS AND UNDERREPORTING OF DEATHS.
THEY’RE TOTALLY DIFFERENT PROBLEMS.
AND THE LAST SENTENCE DEATH RECORDS TO ADDRESS THE LATTER ISSUE.
>> EXCELLENT, THANK YOU.
>>EXCELLENT, THANK YOU, JOHN. WHO NEEDS TO READ ANYTHING ELSE.
>>ANDREW.>>WE HAVE A COUPLE MINUTES I
THINK EVERYBODY SHOULD SEE THE RECOMMENDATIONS PUT FORTH TO
WALK US THROUGH ALL OF THEM.>>I AGREE.
LET’S GO BACK TO ANDREW’S THERE.
>>AND YOU WONDERED WHY. LOOK GOOD, EVERYBODY?
DO WE NEED TO REVIEW ANYTHING ELSE?
KEN? I WANT TO LOOK AT THE TOP OF
YOURS, KATHY, BECAUSE I DIDN’T HAVE A CHANCE TO READ THAT. SUMMARIZING SOME OF THE THINGS I
SAID BUT I WELCOME COMMENTS.
>>YEAH, GOOD. I DON’T HAVE TO READ THE REST.
I’M GOOD WITH THAT. ANYBODY NEED TO READ ANYTHING
ELSE?>>WE HAVE SIX MINUTES.>>YEAH.
>>IT’S IMPORTANT BECAUSE WE’RE ABOUT TO VOTE.
NEED TO FEEL LIKE YOU READ IT ENOUGH TO BE ABLE TO VOTE FOR
IT. I CAN’T HEAR WHAT YOU’RE SAYING.
I STILL CAN’T.>>DEBRA’S ON THE PHONE.
SHE MAY HAVE E-MAILED SOMETHING. WE’RE LOOKING FORWARD TO YOUR
BEING AT THE NEXT MEET, DEBORAH.
>>SHE E-MAILED SOMETHING TO KRISHNA.
DEBORAH CAN YOU HEAR ME.>>I CHECKED YOUR NOTES AND WOVE
THEM INTO THE REST OF THE NARRATIVES THEY DON’T SHOW UP AS
A BULLET FORMAT. I THINK ALL THE CONTENT WAS
CAPTURED.>>AND WE DISCUSSED IT A LITTLE
BIT AND IF THERE’S SOMETHING THAT’S SLIGHTLY CHANGED IT’S
BECAUSE OF A DISCUSSION WE’VE HAD.
SO ALL RIGHT. ARE WE READY?
EVERYBODY READ ENOUGH SO THEY CAN VOTE.
FEEL FREE TO SAY IF YOU NEED TO REVIEW SOMETHING ON HERE.
WE’RE GOOD. OKAY?
WELL, THEN ALL IN FAVOR OF THE RECOMMENDATIONS APPROVING THE
RECOMMENDATIONS? OKAY.
VERY GOOD. WELL DONE.
ALL RIGHT. THAT’S UNANIMOUS.
WE WILL SEND THEM TO YOU TO FIX THE FORMAT I MEAN THE FONTS AND
YOU’LL CATCH OUR LIGHT TYPOS. THANK YOU SO MUCH AND THANK YOU
ALL VERY MUCH. THAT WAS A TON.
THAT WAS AMAZING. WE USUALLY DON’T HAVE THAT MUCH
IN ONE AGENDA AND WON’T IN THE FUTURE.
A LOT OF REALLY IMPORTANT STUFF BEFORE THE DECENNIAL CENSUS WAS
COVERED TODAY. DID YOU HAVE SOMETHING TO STAY?
OKAY. THANK YOU.
>>THANK YOU TO THE STAFF. THIS IS TOMMY WRIGHT.
I THINK SOMEBODY’S WAITING ON KNOW SAY A ROYAL THANK YOU TO
EVERYBODY ON THE COMMITTEE AND THANKS TO EVERYONE ON THE CENSUS
STAFF AND AUDIO AND VISUAL AND SPECIAL ALAWS FOR ENID THE
DIRECTOR AND PRODUCER. HAVE A SAFE TRIP BACK.
I SHOULD SAY THE NEXT MEETING IS MARCH 18 AND 19.
>>AND ANYONE WHO WANTS A RIDE TO NATIONAL AIRPORT FOR THE
URBAN INSTITUTE I’M DRIVING PAST THOED THOSE PLACES AND — THOSE
PLACES AND WOULD BE HAPPY TO GIVE FOUR PEOPLE

Leave comment

Your email address will not be published. Required fields are marked with *.