A Nationally Representative Sample of Asians that is City-BasedView Presentation
*Steven Pedlow, NORC/University of Chicago
Keywords: Area-Probability, PMPE, ICP, CICPE
According to the 2010 Census, only 5.6 percent of the U.S. population is Asian (alone or in combination with another race), making it costly to obtain a representative national sample. The 2000 Partnership and Marketing Program Evaluation (PMPE) was a series of three surveys to evaluate the greatly expanded outreach and promotion campaign for the 2000 Decennial Census. In addition to a Core sample of Hispanics, non-Hispanic Blacks, and non-Hispanic Whites, separate samples were selected for Asians, American Indian and Alaska Natives (AIAN), and Native Hawaiian and other Pacific Islanders (NHOPI). The PMPE selected its sample of Asians from the five U.S. cities with the largest Asian populations. At the time, 18.8 percent of the U.S. Asian population lived within these five cities.
Ten years later, the Integrated Communications Program (ICP) promoted the 2010 Census, and the Census Integrated Communications Program Evaluation (CICPE) is a series of three surveys to evaluate the ICP. Again, separate samples for the Asian, AIAN, and NHOPI populations were added to a Core sample of Hispanics, non-Hispanic Blacks, and non-Hispanic Whites. However, a more representative area-probability design was used for the Asian sample.
Traditional area-probability sampling designs use counties or metropolitan areas as first stage units, but smaller geographical units can better target hard to reach populations. The CICPE used cities as first stage units. We chose a frame of the 1,261 cities with at least 1,000 Asian residents (alone or in combination with another race); 77.7 percent of U.S. Asians live within these 1,261 cities. This paper will show the CICPE Asian sample as a case study on the potential of using smaller geographical units in an area-probability design, and report the challenges of collecting a nationally representative sample for this hard to reach population.