Newspaper article THE JOURNAL RECORD

Analysts Urge Census to Use Statistical Sampling Methods

Newspaper article THE JOURNAL RECORD

Analysts Urge Census to Use Statistical Sampling Methods

Article excerpt

WASHINGTON _ Jumping into what promises to be a lively debate over the shape of the next census, the National Academy of Sciences on Thursday called for the Census Bureau to use statistical sampling techniques to supplement its actual head count in the year 2000.

In a report issued Thursday, a panel of experts, including economists and demographers, said that using sampling techniques to estimate the number of people missed would result in a more accurate and less costly census count.

In the 1990 census, field workers made as many as six visits to each of the nearly 35 million households that failed to return their census forms.

Such visits were exceedingly costly, and often, after many failed attempts, counters estimated the number of people in these households by asking neighbors or merchants in the area _ a process that led to large inaccuracies.

The panel estimated that the change would reduce the costs of the next census by $300 million to $400 million. But among the Republicans who will control the 104th Congress are many who are wary of sampling techniques, asserting that the Constitution calls for an actual count, not merely an estimate of the population.

Moreover, many Republicans oppose estimating the uncounted because the people who are missed tend to be members of minority groups in districts that usually vote Democratic.

Achieving an accurate count could result in some states _ particularly those in the Northeast and the Midwest _ not losing congressional seats to Sun Belt states, which tend to be Republican.

In its report Thursday, the 15-member panel recommended that the field workers, known as enumerators, be sent only to a representative sample of households that fail to return the census form, and that statisticians use the results of that sample to estimate the total number of those who may have been missed. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.