Statement of Census Director Kenneth Prewitt on the
How America Knows What America Needs Program
Press Conference, National Press Club
January 11, 2000
Whatever a decennial census is, the country now knows that it is not a dry, arcane statistical
exercise. Its methods are debated in Congress and litigated before the Supreme Court; its
operations are judged in every corner of the country and have become the focus of widespread
media attention; its results are eagerly awaited by elected officials in thousands of cities,
counties, and tribal lands. And why not -- the census counts affect power, money, group
interests, civil rights, in short, who gets how much of what.
Can the census be about something even bigger and more important in public life? Today's press
conference raises that possibility.
In preparing for the decennial census, now upon us -- the first enumeration takes place in only
10 days, in remote Unalakleet, Alaska -- the Census Bureau has been challenged by the
American people, speaking through their representatives in Congress, to prepare for the best
census in American history. We have met that challenge. The operational plan is in place, on
schedule and on budget.
Does that suggest that we will have the most successful census in history? Not at all. A sound
operational plan is the necessary, but not the sufficient, condition for the best census in history.
That determination will be made by the people of this nation.
Today we issue our challenge. We ask every person in America to accept his or her
responsibility for making the first census of the new century the most successful ever. We can
reach that goal by recognizing that Census 2000 is a civic event. It is civic in its inclusiveness --
the constitution requires that the census count every resident in the country. It is civic in its
consequences -- it underpins and ensures fair elections and our democratic system of political
representation. It is civic in its legal foundation -- by law, every American household is
expected to complete the form.
The larger challenge for the country is to transform what in its nature is a civic event into what
could and should be the nation's first major civic ceremony of the new century.
The program we announce today has that ambition.
A census serves as a barometer of the country's civic health. That barometer has measured the
troubling trend of declining levels of civic engagement.
This is clearly seen by comparing the 1980 and the 1990 censuses.
In 1980, three out of four housing units returned their census form by mail. Public cooperation
with the follow-up effort by door-to-door census-takers was sufficient to allow a final count of
98.8 percent of the population, leaving only 1.2 percent uncounted.
By 1990, the mailback response rate had dropped sharply, to 65 percent. Public cooperation
during the follow-up effort also declined, and the uncounted population increased by nearly a
half-percent to 1.6.
The 1990 census was less complete than the 1980 census not because Census Bureau operations
were less robust (in fact, they were much improved over 1980) but because the public's
indifference to, and even cynicism about, civic responsibility had grown over the decade.
Which brings us to 2000. Current estimates by the Census Bureau indicate that the mailback
response rate will be even lower than in 1990, trending down to approximately 61 percent.
Cooperation rates during the non-response follow-up period can also easily worsen compared to
1990. A lower initial response rate coupled with public indifference during the follow-up effort
will result in an enumeration with a higher undercount than 1990.
This will occur not because the Census Bureau does not want to or is not able "to count" but
because people in this country do not care about or want to be counted.
Our task, as a nation, is to use Census 2000 to do something about the current levels of public
indifference and cynicism. This will make for a better census, and also a better society.
I believe -- and I speak here not as a naively hopeful government bureaucrat but as an
empirically trained social scientist -- that Census 2000 can become a moment of national civic
engagement and pride. There are powerfully encouraging signals. The Census Bureau has been
successful in its effort to find community and business partners, with approximately 35 thousand
partner organizations now actively promoting the census. Twelve thousand cities, counties, and
tribal governments have formed complete count committees. Thousands upon thousands of
community groups, businesses, local governments, and media outlets are investing their own
money, labor and time in improving the census count.
Today the Census Bureau, joined by the impressive array of official supporters, offers added help
to that local community effort by launching How America Knows What America Needs.
This initiative is based on two simple challenges.
First, in '90 Plus Five, every community in America is challenged to improve its 1990 mail
response rate by by at least five percentage points. The vast majority of census forms will reach
households by the third week of March, and will start coming back to the Census Bureau soon
thereafter. We have designed a system that will report to all 39 thousand jurisdictions in the
country how many of their households have returned a questionnaire by March 27, and will
update that information every 24 hours until April 11, when the mailback phase of the census
operation is concluded. Each community will be measuring its performance against the baseline
it set for itself in 1990. Where a community's response rate was 50% in 1990, its 2000 goal is
55%; where it was 85%, the task is to reach 90%.
Across the country, then, the goal is 70% -- an increase of nearly 10 percent over the Census
Bureau's own estimate of what to expect in 2000.
Between March 27 and April 11 we will be taking the temperature of the civic health of the
The second challenge is even more important. Assume that '90 Plus Five meets its challenge and
70% of the housing units return their form -- we have 36 million yet to be counted. That's a lot.
These housing units have to be visited, at great cost, one by one, and often revisited and revisited
before we get a response.
Because You Count is about this phase of the census, which operates for approximately 10 weeks
starting in late April. Here the challenge in each community is to persuade the hard-to-find, the
indifferent, the fearful, or the hostile to cooperate with census-takers who will be coming door to
door. A census undercount, which has nagged every census since 1790, occurs in this phase. So
also then, does the differential undercount, with its well-known implications for whether our
system of political representation is fair, our spending of public funds is equitable, and our
investment by the private sector is effective.
Can '90 Plus Five and Because You Count be successful? If so, Census 2000 will truly be a
moment of national pride. We will celebrate a resurgence of community obligation and civic
Here is what to watch for. If proportionately fewer American households return the census form than did so in 1990 and if levels of public cooperation with door-to-door census-takers repeat the 1990 experience, the several decade long decline in civic engagement continues and no one can know where it will bottom out. If the decline is reversed, take this as a healthy sign that the country intends to start the 21st century by embracing its civic responsibility. In early April, we will start taking our temperature.
[an error occurred while processing this directive]