Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Appendix A REVIEW OF STATISTICAL ASPECTS CONTAINED WITHIN PENDING IMMIGRATION LEGI SLATION: A LETTER REPORT TO THE INS, MAY 1 983 293
294 NATIONAL RESEARCH COUNCIL COMMISSION ON BEHAVIORAL AND SOCIAL SCIENCES AND EDUCATION 2101 Constitution Avenue Washington, D. C. 20418 COMW11EE ON NATIONAL STATISTICS PANEL ON IMMIGRATION SI AllSTICS Mr. Alan C. Nelson Commissioner Immigration and Naturalization Service Department of Justice Washington, D. C. 20536 Dear Mr. Nelson: May 26, 1983 T~NE(202)33~3066 The Panel on Immigration Statistics of the Committee on National Statistics was established by the National Academy of Sciences in late 1982, with funding support from the Immigration and Naturalization Service, to review and determine data needs in the area of international migration to and from the United States. Attachment 1 gives the names and affiliations of Panel members. Although the work of the Panel is still at an early stage, major provisions in legislation currently before Congress (S. 529 and H.R. 1510) to revise _ ____ _ _ _ Nationality Act have prompted the Panel to consider immediately two issues of a statistical nature raised be the bills. and reform the Immigration and ty Act have prompted the Panel to consider immediately two a statistical nature raised be the bills. The Panel has concluded that from a statistical point of view the legislation can be strengthened in two areas. The major points of our recommendations are summarized below. Attachment 2 provides a more detailed explanation and justification of them. 1. Visa waiver for certain visitors. The first recommendation concerns the provision to institute a pilot program to waive visa requirements for temporary visitors who are citizens of certain qualifying countries. Criteria are set forth both for a country to qualify as a pilot country in the first place and for a country so selected to continue as a pilot country after the first year. The Panel finds that the criteria established for permitting continued country participation in the visa waiver experiment fail to recognize the extent of the problems, difficulties, and errors inherent in the statistical system generating the measures to be used. It is our judgment that enactment of the 2-percent threshold as defined at present in the legislation would result in few, if any, countries continuing to qualify for the program after their first year. As a long-run solution, we propose that INS explore and research the issues and problems involved in producing the data required for threshold determination and provide Congress with its findings and recommendations concerning a statistically appropriate threshold level within one year. As part of its activities, the Panel would be pleased to assist in this undertaking. In the interim, we propose the use of a violation ratio based on the number of The National Research Council is the principal operQ!ing age icy of the National Academy of Sciences and the Nahonal Academy of Engineering to scrod government and other organizahons
295 Mr. Alan C. Nelson May 26, 1983 Page Two deportable aliens who had entered the United States under the visa waiver system and were identified by INS in a fiscal year plus those refused entry or withdrawing their applications for admission, and that the threshold be adjusted accordingly. aliens. 2. Data collection aspects of legalization of certain qualifying The proposed legalization process presents a unique opportunity to obtain much needed information concerning a group about which virtu- ally nothing is known. The Panel recommends, accordingly, the inclusion of a specific provision in the legislation authorizing the collection of demographic and socioeconomic data over a five-year period, to provide important insights into the adjustment process of this unique group. Such information should prove invaluable in evaluating and assessing the impact of this legislation and in providing a reliable basis for future debate. We hope that it may yet be possible to effect the recommended changes to strengthen the statistical aspects of the legislation . Sincerely yours' Burton H. Singer Chair Panel on Immigration Statistics Attachments Lisa Roney Doris Meissner (.~:
296 Attachment 1 Panel on Immigration Statistics BURTON H. SINGER (Chair), Department of Mathematical Statistics, Columbia University SAM BERNSEN, Fragomen, Del Rey and Bernsen, Washington, D. C. GEORGE BORJAS, Department of Economics, University of California, Santa Barbara NORMAN CHERVANY, Management.Sciences, University of Minnesota CHARLES KEELY, Population Council, New York City ELLEN KRALY, Department of Sociology, Hamilton College MILTON MORRIS, Joint Center for Political Studies, Washington, D. C. ALEJANDRO PORTES, Department of Sociology, Johns Hopkins University JACK ROSENTHAL, The New York Times, New York City MARK ROSENZWEIG, Department of Economics, University of Minnesota TERESA SULLIVAN, Department of Sociology, University of Texas MARTA TIENDA, Department of Rural Sociology, University of Wisconsin, Madison JAMES TRUSSELL, Office of Population Research, Princeton University KENNETH WACHTER, Department of Statistics, University of California, Berkeley DANIEL B. LEVINE, Study Director KENNETH H. HILL, Associate Study Director ROBERT WARREN, Research Associate ROBERTA PIROSKO, Administrative Secretary/Research Aide
297 Attachment 2 Rationale for Modifications to the Immigration Reform and Control Act of 1983 1. Visa waiver for certain visitors. Both S. 529 and H.R. 1510 contain provisions for instituting a pilot program of waiving visa requirements for nonim~igrant visitors for citizens of certain countries. The language in the two bills differs in only very minor respects. Essentially, a country cannot qualify as a pilot country unless the number of refusals to grant nonimmigrant visitor visas in the preceding fiscal year to citizens of that country is less than 2 percent of the number of nonimmigrant visitor visas granted or refused to such citizens. In order to continue to qualify as a pilot country, the number of violators from that country in the preceding fiscal year must be less than 2 percent of total applications for admission under the visa waiver scheme; for this purpose, violators are taken to include people who are excluded, those who withdraw their applications for admission, and those who violate their terms of entry under the scheme, largely by overstaying. Given the nature of the statistical system on which the test of eligibility will be based, the Panel views these requirements for continuation as a pilot country to be unreasonably restrictive. The Nonimmigrant Information System (NITS) recently introduced by INS is based upon a new form (I-94), one part of which is completed upon arrival, the other part being completed on departure; for departures by air, the airline is required to collect the form. Each part of the form contains the same unique identifying number, which permits matching the arrival and departure portions. Thus, length of stay can be determined from dates of arrival and departure, allowing in theory the identifica- tion of those who violated the length of stay condition of admission. Arrival portions for which no matching departure portions are available also may be interpreted as violations. No such strict interpretation is valid, however, because of the problems, difficulties, and errors inherent in the system that produces the data and specifically with the matching of arrival and departure portions of the I-94 form. The new NITS system has not yet generated enough data for an empirical assessment of its error level to be possible. It will no doubt represent a substantial improvement over its predecessor, but the Panel, based on its experience with similar systems, doubts whether the 2-percent threshold can be meaningfully measured. The weak points are as follows. o First, the collection of departure portions of the forms is not controlled directly by INS, and the possibility of loss is substantial. For example, for visitors leaving by air, the departure portions may either not be collected by the carrier or not forwarded to INS, particu- larly for nonscheduled flights. For persons leaving to Canada, all the forms may not be collected and duly returned to INS from Canada; for those departing to Mexico by land, the forms may not be surrendered voluntarily on departure.
298 o Second, a temporary visitor may lose the departure portion, in which case a replacement with no identification number is to be com- pleted. The enforcement of such completion is quite limited for all types of departure, so some arrival portions will not have corresponding departure portions. A further problem with such a loss is the matching that, in the absence of an identification number, has to be based on name, date of birth, and citizenship: name is a notoriously unreliable basis for matching as a result of spelling variations and different forms of presenting the same name, and a fairly high incidence of false non-matches could therefore be expected in the case of such replacement departure portions. o The third problem is that of the not insignificant error associ- ated with entering an 11-digit identification number into a data process- ing system; an error in the identification number of either the arrival or departure portion would result in a failure to match. Errors in the NITS process tend to produce false non-matches that appear as violations by visitors who in fact have not violated the terms of their admission. Further, important sources of error arise from parts of the system that are not under the direct control of INS, the responsi- bility being on individual visitors or carriers who are difficult if not impossible to monitor. Currently, measures of the extent and variability over time of such errors are not known, but in our judgment if the pro- posed requirements are put into effect, the 2-percent threshold is likely to defeat the object of the proposed visa waiver legislation, removing from eligibility countries with very low true violation rates, and indeed probably removing all countries from eligibility. The Panel believes that tests of eligibility of this sort should be based upon measures that can be reliably evaluated, with probable error levels that are small in comparison with trigger levels. Accordingly, the following courses of action are recommended: o That INS explore and research the issues, problems, and errors involved in producing the data elements necessary for threshold determi- nation and provide the Congress with its findings and recommendations as to a statistically appropriate threshold level within one year. The Panel would be pleased to assist the INS in this undertaking. o In the interim, that the threshold computation use information on persons positively identified by INS as being in violation of the terms of admission. Included as "violators" would be those overstaying, working unlawfully, or engaging in criminal activities, and those excluded from admission or withdrawing applications. "Violators" thus would represent the number of Reportable aliens who had entered the country under the visa waiver system and who were identified as violators by INS in the fiscal year plus the number of exclusions and withdrawals in the year. Data for 1979, the most recent year for which information is available, indicate a violation ratio defined in this way of 0.5 percent for all temporary visitors. The Panel therefore recommends the adoption of a threshold of 0.5 percent pending completion of the study
299 recommended above. The violation ratio as described here involves no matching, and the likely error in both numerator and denominator is 11. 2. Data collection aspects of legalization of certain qualifying aliens. The provisions of S. 529 and H.R. 1510 for legalizing the status of certain unlawfully resident aliens represent an important opportunity for obtaining much needed information concerning a group about which virtually nothing is known. The Senate bill, S. 529, already recognizes in Section 401(e) the importance of obtaining information, for those aliens legalized under the act, on their demographic characteristics, employment, and participation in social service programs. In the spirit of this provision, the bill should also recognize, first, the need for this information to be collected in a statistically sound and reliable manner, and, second, the need for information to be collected concerning changes that occur over time among those legalized under the proposed legislation. It is such changes over time that will reflect the impact of the new law on the American economy and society and on the legalized population itself. Only information collected during the time when the effects of the law are felt, not before they are felt, can provide a sound basis for continuing responsible administration of the programs initiated by the law. Objective findings on the progress of assimilation of the legalized population following enactment of the law would provide the basis for developing a consensus in regard to facts, bringing closer together the groups with highly diverse perspectives that have joined in the debate about proper immigration policy. Furthermore, the periodic updating of geographical and employment information on at least one group of former unlawfully resident aliens would be essential in realizing the potential value for law enforcement of statistical information acquired as part of the legalization program; this potential value was emphasized by the Select Commission on Immigration and Refugee Policy in its report. Modern statistical panel survey methods, using a sampling design, provide a way to obtain information on changes over time at a fraction of the cost of a routine reporting system. They also permit estimates of the reliability of the findings to accompany the resulting tabulations so that the quality of the conclusions can be assessed. The Panel therefore recommends the inclusion in the legislation of a specific provision mandating the collection of demographic and socio- economic information over a five-year period from a sample of aliens coming forward for legalization. The group to be designated for this study could be selected in two stages. First, a sample could be drawn of aliens coming forward for legalization, and information additional to that normally obtained from applicants for permanent residence could be collected from the sampled aliens, particularly concerning their entry into, and employment history in, the United States. A subsample of these aliens could then be drawn to participate in the continuing survey. It is further recommended that a workshop be convened in the near future to discuss the design, methodology, content, and cost of the proposed
300 surveys; that Section 401 of S. 529 be amended to provide explicitly for such data collection; and thee a section similar to Section 401 of S. 529 be added to H.R. 1510. It should be recognized that the proposed study will not provide information for undocumented aliens who do not come forward for legaliza- tion. Efforts to obtain information about this even more inaccessible group through other approaches should continue. The value of information about aliens seeking legalization also would be greatly enhanced by parallel studies of legal immigrants and refugees. It is recommended, therefore, that consideration be given to undertaking similar studies over a five-year period for these groups, though this recommendation is beyond the scope of the legislation currently before the Congress. 1