SJR 8-FAIR AND ACCURATE CENSUS MARK HODGINS, legislative aide to Chairman Ward, sponsor of SJR 8, explained the measure as follows. SJR 8 urges Congress and the Bureau of the Census to conduct the year 2000 census consistent with the ruling in the Department of Commerce v. United States House, and with the Constitution of the United States. The Bureau of the Census plans on using random sampling techniques or other statistical methods instead of an actual physical headcount of the population. A random sampling technique will dilute and violate the one person/one vote legal protection, and could expose the State to protracted and costly litigation, resulting in a court ruling invalidating the redistricting plan. SENATOR PHILLIPS asked if a random sampling technique has been used in any previous census. MR. HODGINS said it has. He explained the case he previously referred to, Department of Commerce v. United States House, was about the use of a random sampling technique. SENATOR PHILLIPS asked when that occurred. MR. HODGINS was unsure but stated in that case, the court determined the random sampling technique used was acceptable, however the current concern is that a random sampling technique could prompt the Legislature to artificially change district boundaries based on statistics, rather than a headcount, during the reapportionment process. He stated ten moose could be spotted in one square mile in Alaska, however one cannot assume that number is representative of thousands of square miles. SENATOR WILKEN asked if Alaska's congressional delegation favors this legislation. MR. HODGINS said it does. CHAIRMAN WARD stated this issue is not partisan, and that most people do not favor the random sampling technique. CHRIS MILLER, Research and Analysis Section of the Department of Labor (DOL), stated DOL has taken no position on SJR 8, but he asked the committee to consider the following information. A pending legal decision may make this issue moot. A group out of Alabama is likely to sue the Census Bureau to force it to do the same thing for state census information as it does for the national census. DOL, as the census liaison, does not know the details of the statistical sampling process the Census Bureau plans to use for the year 2000. Number 149 CHAIRMAN WARD asked Mr. Miller if he was suggesting the Legislature introduce an accompanying bill containing precise language as to how a census count should be conducted. MR. MILLER replied he would prefer to wait to see what the Census Bureau proposes. The 1990 census process began with a statistical quality control check conducted on a regional level across the country. The regional statistics were then applied to each state. DOL does not believe that method improved the numbers for Alaska. CHAIRMAN WARD asked if DOL has been notified by the Census Bureau that Alaska is being used as a test site, in which an ongoing population count would occur. MR. MILLER stated at one point, Anchorage was going to be a test site, but that is no longer the case. He explained the Census Bureau is developing a new process in which it will collect data over an eight-year period. By the eighth year, the statistics will be abundant enough to provide detailed data on small geographic areas. That process might be used in the 2010 census, but a lot of testing still needs to be done. CHAIRMAN WARD asked if the entire State of Alaska has been chosen as a test site for that ongoing procedure. MR. MILLER replied the program will be applied nationwide, and it is not a test per se. The program will occur simultaneously with the 2000 census, and the sample will be used as a base for comparison. Number 206 SENATOR ELTON asked if out-of-state military members are counted using a statistical sampling method. MR. MILLER replied the Census Bureau contracted to do a head count of out-of-state military members. SENATOR ELTON asked if enumerators from other states would go to bases overseas and in the lower 48. MR. MILLER said that was his understanding, however he would verify that information. SENATOR ELTON asked about DOL's experience with census population counts in areas of the state in which the population is very mobile, and whether that experience has led DOL to believe that statistical sampling can provide an accurate count. Number 229 MR. MILLER said DOL is working closely with the Census Bureau to ensure the Census Bureau has good information about where housing units are located. The census in Alaska will be done in two ways. In urban areas, surveys will be mailed to residents and census workers will go door-to-door. In rural Alaska, the Bureau will contract with local people to go door-to-door to collect the information. Because "break-up" occurs when the census will be taken, the door-to-door method will be used to count residents of rural areas of Alaska, and will take longer. DOL is fairly confident that the Census Bureau is "counting noses" in an efficient manner. Post-enumeration surveys are conducted after the census is complete. A 1995 post-enumeration survey was a secondary survey used for statistical purposes. It was also used as a quality control check on the 1990 census. There being no further discussion, SENATOR PHILLIPS moved SJR 8 out of committee with individual recommendations. SENATOR ELTON objected. The motion carried with Senators Phillips, Green, Wilken, and Ward voting "yea," and Senator Elton voting "nay."