TEN_21-10.pdf

ETA Advisory File Text

NAWS Public Access Data Fiscal Years 1989 2009 Background This version of the National Agricultural Workers Survey NAWS Public Access Data NAWSPAD contains information from 52 479 in-person interviews with hired crop farm workers. The interviews were conducted in 467 counties and 40 states between October 1 1988 and September 30 2009. The data set includes 220 questionnaire variables and 100 created variables. This document describes the data and provides analysis tips. The data are currently available in a zipped SAS file. Summary tables of the survey s key variables will be posted to the NAWS Web page this year. It is highly recommended that the questionnaire and codebook be downloaded and examined prior to analyzing data. Questionnaire variable names in the codebook match the cycle 64 questionnaire. The NAWSPAD was constructed under two guiding principles 1 to protect the privacy rights of the farm worker respondents and the agricultural employers at whose establishments the workers were interviewed and 2 to provide a data set that allows in- depth analysis of the demographic and employment characteristics of hired crop workers at both the national and regional level. Confidential Data Safeguards The names and addresses of employers and respondents do not appear in either the full NAWS data set or the NAWSPAD. Additional safeguards have been incorporated in the NAWSPAD the variables crop and task have been collapsed into five and six categories respectively the 12 sampling regions have been collapsed into six and the season of interview variable has been removed. Sampling Regions Workers are sampled from 12 regions which are aggregated from 17 U.S. Department of Agriculture-designated regions. The sampling methodology is discussed at http www.doleta.gov agworker statmethods.cfm . As noted above the 12 regions have been collapsed into six in the NAWSPAD. Table 1 lists the regions and the states each region represents Table 2 shows the number of interviews by fiscal year FY and region. Table 1. NAWSPAD Regions and States Region No. Region Name States Represented 1 EAST North Carolina Virginia Kentucky Tennessee West Virginia Connecticut Maine Massachusetts New Hampshire New York Rhode Island Vermont Delaware Maryland New Jersey Pennsylvania 2 SOUTHEAST Arkansas Louisiana Mississippi Alabama Georgia South Carolina Florida 3 MIDWEST Illinois Indiana Ohio Iowa Missouri Kansas Nebraska North Dakota South Dakota Michigan Minnesota Wisconsin 4 SOUTHWEST Arizona New Mexico Oklahoma Texas 5 NORTHWEST Idaho Montana Wyoming Colorado Nevada Utah Oregon Washington 6 CALIFORNIA California Table 2. Number of Interviews by Fiscal Year and Region 1989-2009 FY Region 1 East Region 2 Southeast Region 3 Midwest Region 4 Southwest Region 5 Northwest Region 6 California Total 1989 311 504 322 269 324 796 2526 1990 149 571 58 300 230 810 2118 1991 149 731 88 321 279 1032 2600 1992 147 552 87 235 201 674 1896 1993 423 351 442 248 265 664 2393 1994 478 320 454 221 290 611 2374 1995 347 306 411 237 354 708 2363 1996 264 371 306 242 277 591 2051 1997 338 397 257 242 264 602 2100 1998 319 428 262 236 262 592 2099 1999 569 487 546 239 489 1282 3612 2000 383 603 440 223 528 1409 3586 2001 414 548 393 177 417 1162 3111 2002 445 585 478 224 431 1198 3361 2003 464 616 499 266 477 1263 3585 2004 413 519 455 171 407 1081 3046 2005 314 337 374 118 314 770 2227 2006 176 300 196 116 196 535 1519 2007 197 294 176 100 193 551 1511 2008 316 407 211 142 265 841 2182 2009 275 392 232 162 337 821 2219 Total 6891 9619 6687 4489 6800 17993 52 479 Interview Cycles and Analysis Tips Interviews are conducted three times per year to account for the seasonality of agricultural production and employment. The cycles begin in October February and July. Due to the timing of the cycles the data are analyzable by Federal fiscal year October 1 to September 30 rather than calendar year. The data are not analyzable by interview cycle. Sampling and post-sampling weights are used in the NAWS to adjust the relative value of each interview so that population estimates may be obtained from the sample. The composite weight variable PWTYCRD is included in the NAWSPAD and it should be used in all analyses to generate unbiased percentages and means. For example to obtain the percentage of male and female crop workers in Federal fiscal years 2001-2002 the following SAS statements would be used proc freq data name of data set table GENDER where 2001 le FY le 2002 weight PWTYCRD run For a fuller explanation of the weights please consult pages 6-13 of the statistical methods document http www.doleta.gov agworker statmethods.cfm . Data Limitations Except for California the data are not available at the state-level. Due to low annual sample sizes at least four years should be combined for all regional-level analyses. When analyzing data at the national level at least two years of data should be combined. Due to the survey s complex sampling design a special analysis procedure is needed to generate survey-design-corrected standard errors. The procedure used in SAS is discussed on pages 17-18 of the statistical methods document. As the procedure requires the variable for the primary sampling units which for confidentiality safeguards is not included in the NAWSPAD researchers who need to report design-corrected standard errors are encouraged to contact Mr. Daniel Carroll via phone 202 693-2795 or e-mail carroll.daniel.j dol.gov to discuss their information needs. The health condition variables provided in the NAWAPAD represent diagnoses over the life of the crop worker and may not represent existing conditions for the farm worker at the time of interview. Work-relatedness was not assessed for these health conditions.