The Labor Condition Application (LCA)
The Labor Condition Application (LCA) is the first major form which must be filed in the process of employing foreign workers in the U.S. on a nonimmigrant basis, chiefly via an H-1B visa. As stated at this link, "Employers petitioning for temporary foreign workers in specialty occupations must have a certified Labor Condition Application from the Department of Labor and then must submit a Form I-129 to USCIS." A great deal of disclosure data from these LCAs is available on the Department of Labor website at this link. Expanding the Disclosure Data tab on this page will display links to the LCA Disclosure Data from the most recent year in the section titled "OFLC Programs and Disclosures". Links to LCA Disclosure Data from 2008 to the prior year and be found further down the page, in the section titled "LCA Programs (H-1B, H-1B1, E-3)".
Key data from the LCA Disclosure Data can be searched and displayed via the R Shiny application at https://econdata.shinyapps.io/lca1/. This document shows examples of how this can be done.
Displaying Data for a Single Employer and Year
Displaying Data for Each Application
Below is the initial screen that is currently displayed at https://econdata.shinyapps.io/lca1/. The panel on the left is the input sidepanel and specifies that the 2021 CERTIFIED applications for APPLE are to be listed, sorted by TOTAL_WORKERS. The ^ and $ characters before and after APPLE specify the beginning and end of the full EMPLOYER_NAME and their usefulness in this case is explained below.
The output in the Output tab starts with 3 lines beginning with '>' and representing warning and status messages. Those are followed by the title and subtitle, the number of rows, and the mean salary of the matching data. There is one row per application and the mean salary is the average of the WAGE_RATE_FROM column (not weighted by TOTAL_WORKERS). Following the NUMBER OF ROWS and MEAN(SALARY) is a list of the number of TOTAL_WORKERS and APPLICATIONS per year.
The list of yearly totals is followed by a list containing one row per application. For each application, the variables listed in the header are listed by default. All but Year and WAGE_PWAGE are taken directly from the disclosure files and correspond to the the fields CASE_STATUS, SOC_TITLE, TOTAL_WORKER_POSITIONS (listed as TOTAL_WORKERS), EMPLOYER_NAME, WORKSITE_CITY (listed as WORK_CITY), WORKSITE_STATE (listed as WORK_STATE), WORK_RATE_OF_PAY_FROM (listed as WORK_RATE_FROM) described at the 2021 LCA Record Layout. WAGE_PWAGE is calculated and equals WORK_RATE_FROM (the lowest wage that the worker will be paid) divided by the prevailing wage. Because the worker must ordinarily be paid at or over the prevailing wage, this number should be greater than or equal to 1.
Grouping by EMPLOYER_NAME
In order to look at data for a single employer and only that employer, the "Group by" input can be used to determine all of the employer names that will match the one specified in the "Search EMPLOYER_NAME" input. For example, setting "Search EMPLOYER_NAME" to APPLE and "Group by" to EMPLOYER_NAME and then clicking the Submit button will result in the following screen:
A "Search EMPLOYER_NAME" input that allows a character after APPLE will match all items except for 1 (APPLE) and 15 (MOLLY CRABAPPLE). Hence, a trailing $ is required after APPLE to specify that no other characters are allowed. Then, in order to avoid matching 15 (MOLLY CRABAPPLE), a leading ^ is also required. Since 1 (APPLE) appears to be the only valid match, the entry of ^APPLE$ is the appropriate "Search EMPLOYER_NAME" input. It should be noted that this input is treated as a regular expression or "regex".
Cleaning the EMPLOYER_NAME
One important thing to note is that the application will remove extraneous characters and trailers in order to simplify the EMPLOYER_NAMEs and combine different names for the same employer. The settings for this processing are at the bottom of the left sidepanel as shown below.
The "Clean EMPLOYER_NAMEs" input is set to "Clean Both" by default. The checkboxes under "Ignore in Employer" specify characters which can be ignored. Checking "comma", "period", and "blanks" will generally convert those specified characters into single blanks. The checkbox "case" will ignore character case (upper or lower), the checkbox "the" will ignore the leading word "the" (regardless of case), and the checkbox "and" will convert all cases of " and " to " & ". Finally, the phrases selected in the "Delete Trailer in Employer" input specify those phrases which will be deleted if they appear at the end of the employer name.
All of the above cleaning can be turned off by setting the "Clean EMPLOYER_NAMEs" input to "Clean None". Doing so in this case will result in the EMPLOYER_NAMEs in the leftmost 4 columns below:
EMPLOYER_NAME WORKERS APPLICATIONS EMPLOYER_NAME WORKERS APPLICATIONS
1 APPLE INC. 38980 4155 1 APPLE 38980 4155
2 APPLET SYSTEMS LLC 14 14 2 APPLET SYSTEMS 14 14
3 APPLEXUS TECHNOLOGIES LLC 7 7 3 APPLEXUS TECHNOLOGIES 12 12
4 APPLESEED MONTESSORI INC. 5 5 4 APPLESEED MONTESSORI 5 5
5 APPLEXUS TECHNOLOGIES INC. 5 5
6 APPLE TREE DENTAL 2 2 5 APPLE TREE DENTAL 2 2
7 ABENDROTH'S APPLE RIDGE ORCHARD LLC 2 2 6 ABENDROTH'S APPLE RIDGE ORCHARD 2 2
8 BERRY APPLEMAN & LEIDEN LLP 2 2 7 BERRY APPLEMAN & LEIDEN 2 2
9 THE BOARD OF APPLETON MUNICIPAL HOSPITAL 1 1 8 BOARD OF APPLETON MUNICIPAL HOSPITAL & N 1 1
10 APPLECHEM INC. 1 1 9 APPLECHEM 1 1
11 APPLE AMERICAN GROUP LLC 1 1 10 APPLE AMERICAN GROUP 1 1
12 APPLECART LLC 1 1 11 APPLECART 1 1
13 HOLBERT APPLE ASSOCIATES, INC. 1 1 12 HOLBERT APPLE ASSOCIATES 1 1
14 APPLE TREE DAYCARE LLC 1 1 13 APPLE TREE DAYCARE 1 1
15 PROJECT APPLECART LLC 1 1 14 PROJECT APPLECART 1 1
16 MOLLY CRABAPPLE INC. 1 1 15 MOLLY CRABAPPLE 1 1
17 APPLE TREE INTERNATIONAL CORP 1 1 16 APPLE TREE INTERNATIONAL 1 1
The rightmost 4 columns show the resulting EMPLOYER NAMEs when the "Clean EMPLOYER_NAMEs" input is set to "Clean Both". As can be seen, the main difference is that the specified trailers are deleted from the EMPLOYER_NAMEs. Also, the 7 and 5 applications for "APPLEXUS TECHNOLOGIES LLC" and "APPLEXUS TECHNOLOGIES INC.", respectively, are combined to get 12 applications for "APPLEXUS TECHNOLOGIES". Also, "BOARD OF APPLETON MUNICIPAL HOSPITAL & N" is truncated to 40 characters because the "Maximum Column Width" input is set to 40.
Displaying Data for Multiple Employers and a Single Year
Grouping by EMPLOYER_NAME (primary employer)
If the "Clean EMPLOYER_NAMEs" input is set back to the default "Clean Both", the "Search EMPLOYER_NAME" input is set to blank, the "Group by" input is set to EMPLOYER_NAME, and the Submit button is clicked, the following screen will be displayed:
This shows all of the employers in 2021, grouped by EMPLOYER_NAME and sorted by TOTAL_WORKERS. The totals by year show that there were a total of 1,013,997 workers in 497,334 applications submitted in 2021. Both numbers are far above the 85,000 maximum H-1B slots in the annual lottery. It's not clear that those applications that request hundreds of workers are really serious about hiring that many. Hence, it may be more conservative to look at the number of applications.
In any case, changing the "Sort by" input to APPLICATIONS will cause the list to be sorted by APPLICATIONS. Following are the top 50 EMPLOYER_NAMEs in 2021, sorted by APPLICATIONS:
DISCLOSURE DATA, LCA_FY 2021-2021
(CASE_STATUS=CERTIFIED)
NUMBER OF ROWS = 497,334
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2021 1013997 497334
EMPLOYER_NAME WORKERS APPLICATIONS
1 COGNIZANT TECHNOLOGY SOLUTIONS US 12463 12463
2 TATA CONSULTANCY 12729 9797
3 GOOGLE 9047 9047
4 AMAZON.COM 83533 8786
5 MICROSOFT 19763 7329
6 ERNST & YOUNG U.S 5783 5783
7 INFOSYS 92075 5599
8 FACEBOOK 11366 5590
9 APPLE 38980 4155
10 HCL AMERICA 14059 4063
11 ACCENTURE 3817 3771
12 DELOITTE CONSULTING 5716 3715
13 INTEL 3505 3364
14 CAPGEMINI AMERICA 3228 3228
15 WAL-MART ASSOCIATES 2961 2961
16 JPMORGAN CHASE 2944 2944
17 WIPRO 13282 2785
18 AMAZON WEB 25731 2603
19 GOLDMAN SACHS 2273 2273
20 COMPUNNEL SOFTWARE GROUP 2122 2112
21 SALESFORCE.COM 2043 2043
22 TECH MAHINDRA (AMERICAS) 2008 2008
23 CISCO SYSTEMS 4817 2003
24 QUALCOMM TECHNOLOGIES 53553 1998
25 INTERNATIONAL BUSINESS MACHINES 1928 1928
26 TEKORG 1645 1645
27 DELOITTE & TOUCHE 3038 1516
28 MPHASIS 2032 1408
29 LARSEN & TOUBRO INFOTECH 1436 1368
30 MINDTREE 1282 1282
31 FIDELITY TECHNOLOGY GROUP 1538 1250
32 IBM 1221 1221
33 KFORCE 11816 1183
34 VMWARE 1171 1171
35 TESLA 1163 1163
36 ATOS SYNTEL 1152 1152
37 RANDSTAD TECHNOLOGIES 1102 1102
38 PAYPAL 1686 1077
39 LINKEDIN 2709 1043
40 ORACLE AMERICA 14386 984
41 PRICEWATERHOUSECOOPERS ADVISORY 946 946
42 CITIBANK N.A 871 871
43 UBER TECHNOLOGIES 4010 850
44 CGI TECHNOLOGIES & SOLUTIONS 7950 842
45 EBAY 1403 823
46 NTT DATA 801 801
47 CUMMINS 790 789
48 ANTHEM 777 777
49 MCKINSEY & COMPANY INC UNITED STATES 4427 751
50 ADOBE 1577 748
Of these employers, 5 of the top 10 are IT consulting firms listed at this link. Those 5 firms are Cognizant, Tata, Ernst & Young, Infosys, and HCL America. The grouping by EMPLOYER_NAMEs does not perfectly group by company. For example, #3 (AMAZON.COM) and #18 (AMAZON WEB) are both a part of Amazon. Still, the grouping appears to be fairly accurate.
Grouping by EMPLOYER_NAME2 (secondary employer)
As described at this link, EMPLOYER_NAME is the "Legal business name of the employer submitting the Labor Condition Application". Since 2019, LCAs have also included something like SECONDARY_ENTITY_BUSINESS_NAME, described here as "Name of secondary entity where the worker(s) will be placed (if applicable)". These secondary employers hire H-1B workers from their primary employers, typically an IT consulting firm, for specific assignments. They typically pay a fee to the primary employer who submitted the LCA as long as that employee is on assignment. The name of this secondary employer is denoted in this app as EMPLOYER_NAME2.
If there is no secondary employer, EMPLOYER_NAME2 will typically be set to NA in the disclosure data. The "EMPLOYER_NAME2 NAs" input determines how these NAs are handled by the app. The default setting is "Remove" which specifies that NAs are to be removed before applications are grouped by EMPLOYER_NAME2. The option "Retain" specifies that the NAs are to be retained. In this case, the largest grouping of EMPLOYER_NAME2 is likely to be NA. Finally, "Set to EMPLOYER_NAME" will set NA values of EMPLOYER_NAME2 to the same value as EMPLOYER_NAME. This will serve to ensure that every application is represented by some employer. That will be the secondary employer if there is one. Otherwise, it will be the primary employer. Following are the top 50 secondary entities obtained by grouping by EMPLOYER_NAME2 and sorting by APPLICATIONS (using the default setting of "Remove"):
DISCLOSURE DATA, LCA_FY 2021-2021
(CASE_STATUS=CERTIFIED)
NUMBER OF ROWS = 127,956
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2021 1013997 497334
EMPLOYER_NAME2 WORKERS APPLICATIONS
1 AMERICAN EXPRESS 3609 1430
2 APPLE 5976 1416
3 CAPITAL ONE 1697 1360
4 WELLS FARGO 2871 1265
5 FORD MOTOR 2537 1250
6 AT&T 2144 1247
7 CITIGROUP 1312 1025
8 FIDELITY INVESTMENTS 1517 1002
9 BANK OF AMERICA 910 905
10 VERIZON 907 869
11 ANTHEM 885 827
12 CISCO SYSTEMS 3314 790
13 CVS HEALTH 797 742
14 UNITED SERVICES AUTOMOBILE ASSOCIATION 2186 736
15 HUMANA 785 691
16 CHARLES SCHWAB 2793 654
17 FEDEX 1238 646
18 CHARTER COMMUNICATIONS 1675 638
19 GOOGLE 1819 612
20 FANNIE MAE 1073 581
21 PNC BANK 2764 580
22 WALMART 589 545
23 JOHNSON & JOHNSON 645 544
24 KAISER PERMANENTE 907 526
25 COMCAST 1762 518
26 CIGNA 2750 484
27 T-MOBILE 489 479
28 CATERPILLAR 733 453
29 NIKE 1972 442
30 MICROSOFT 1158 428
31 PAYPAL 412 403
32 FACEBOOK 418 398
33 EQUIFAX 425 392
34 MORGAN STANLEY 376 344
35 AT & T 336 336
36 CVS PHARMACY 2063 333
37 CITIBANK N.A 378 323
38 DELOITTE CONSULTING 338 317
39 ABBVIE 318 317
40 UNITED AIRLINES 610 316
41 PNC 1660 310
42 MERCK 1293 309
43 CITIBANK 385 308
44 HOME DEPOT 353 287
45 VANGUARD GROUP 343 285
46 CUMMINS 559 282
47 VERIZON SOURCING 1369 278
48 PFIZER 1099 276
49 BENEFICIARY'S RESIDENCE 273 273
50 CENTENE 367 268
As can be seen, most of the top companies are not typically thought of as tech companies as is the case with the top primary employers. Out of the top 10 secondary employers, 6 are financial companies (AMERICAN EXPRESS, CAPITAL ONE, WELLS FARGO, CITIGROUP, FIDELITY INVESTMENTS, and BANK OF AMERICA), 2 are communication providers (AT&T and VERIZON), 1 is an auto manufacturer (FORD MOTOR), and only one is a tech company (APPLE). Until 2019, these secondary employers largely avoided notice in their use of H-1B workers, at least as far as disclosure data was concerned.
Grouping by CASE_STATUS
All of the other tables in this document look only at certified LCAs. This includes the great majority of LCAs that are submitted. Clearing the "Search CASE_STATUS" input, setting the "Group by" input to CASE_STATUS, and clicking the Submit button will output the following:
DISCLOSURE DATA, LCA_FY 2021-2021
NUMBER OF ROWS = 532,201
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2021 1071503 532201
CASE_STATUS WORKERS APPLICATIONS
1 CERTIFIED 1013997 497334
2 CERT-WITHDRAWN 39621 22231
3 WITHDRAWN 14447 9944
4 DENIED 3438 2692
As can be seen, over 93 percent of the LCAs (497,334 of 532,201) were certified in 2021. As mentioned at this link, "Employers with certified LCAs may proceed with the process of obtaining an H-1B, H-1B1 or E-3 visa through USCIS and the Department of State." The other statuses are explained at this link.
Grouping by SOC_TITLE
As described at the 2021 LCA Record Layout, SOC_TITLE is the "Occupational title associated with the SOC/O*NET Code." It is similar to JOB_TITLE but is standardized and is therefore better to group by. Following are the top 50 occupational titles obtained by grouping by SOC_TITLE and sorting by APPLICATIONS:
DISCLOSURE DATA, LCA_FY 2021-2021
(CASE_STATUS=CERTIFIED)
NUMBER OF ROWS = 497,334
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2021 1013997 497334
SOC_TITLE WORKERS APPLICATIONS
1 SOFTWARE DEVELOPERS, APPLICATIONS 267735 158937
2 SOFTWARE DEVELOPERS, SYSTEMS SOFTWARE 91850 30284
3 COMPUTER SYSTEMS ANALYSTS 49895 30135
4 COMPUTER SYSTEMS ENGINEERS/ARCHITECTS 23335 16379
5 SOFTWARE QUALITY ASSURANCE ENGINEERS AND 21064 12358
6 INFORMATION TECHNOLOGY PROJECT MANAGERS 19382 10862
7 COMPUTER AND INFORMATION SYSTEMS MANAGER 24603 10397
8 BUSINESS INTELLIGENCE ANALYSTS 20919 9412
9 MECHANICAL ENGINEERS 13537 9214
10 OPERATIONS RESEARCH ANALYSTS 20017 8434
11 COMPUTER PROGRAMMERS 17997 8327
12 FINANCIAL ANALYSTS 9617 7985
13 STATISTICIANS 15536 7806
14 ELECTRONICS ENGINEERS, EXCEPT COMPUTER 61950 7791
15 COMPUTER OCCUPATIONS, ALL OTHER 76232 7646
16 DATABASE ADMINISTRATORS 13309 7311
17 ELECTRICAL ENGINEERS 12388 6781
18 MANAGEMENT ANALYSTS 18548 6303
19 ACCOUNTANTS AND AUDITORS 9183 5999
20 INDUSTRIAL ENGINEERS 7805 4608
21 NETWORK AND COMPUTER SYSTEMS ADMINISTRAT 13421 4521
22 COMPUTER AND INFORMATION RESEARCH SCIENT 16194 4090
23 MEDICAL SCIENTISTS, EXCEPT EPIDEMIOLOGIS 5908 3909
24 MARKET RESEARCH ANALYSTS AND MARKETING S 5976 3871
25 PHYSICIANS AND SURGEONS, ALL OTHER 4461 3770
26 INFORMATION SECURITY ANALYSTS 7413 3761
27 CIVIL ENGINEERS 3835 3469
28 BIOCHEMISTS AND BIOPHYSICISTS 3233 2852
29 MARKETING MANAGERS 5472 2474
30 FINANCIAL QUANTITATIVE ANALYSTS 2817 2367
31 SALES ENGINEERS 5321 2171
32 DATA WAREHOUSING SPECIALISTS 2212 1999
33 COMPUTER NETWORK ARCHITECTS 3684 1977
34 CHEMISTS 3055 1977
35 ACCOUNTANTS 2070 1864
36 LAWYERS 2144 1758
37 DATABASE ARCHITECTS 2079 1730
38 HEALTH SPECIALTIES TEACHERS, POSTSECONDA 1684 1678
39 ARCHITECTURAL AND ENGINEERING MANAGERS 4410 1668
40 MEDICAL AND CLINICAL LABORATORY TECHNOLO 2345 1646
41 PHYSICAL THERAPISTS 12411 1598
42 GRAPHIC DESIGNERS 4025 1597
43 MATERIALS ENGINEERS 3190 1499
44 FINANCIAL MANAGERS 2243 1498
45 WEB DEVELOPERS 2608 1463
46 VALIDATION ENGINEERS 1466 1422
47 MOLECULAR AND CELLULAR BIOLOGISTS 1427 1407
48 RISK MANAGEMENT SPECIALISTS 1944 1276
49 HOSPITALISTS 1326 1247
50 GENERAL AND OPERATIONS MANAGERS 1383 1229
As can be seen above, to top 7 occupational titles involve software, computers, or information technology.
Grouping by WORK_CITY and WORK_STATE
It's possible to group by more than one field. Following are the top 50 WORK_CITYs and WORK_STATEs for which applications were submitted:
DISCLOSURE DATA, LCA_FY 2021-2021
(CASE_STATUS=CERTIFIED)
NUMBER OF ROWS = 497,334
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2021 1013997 497334
WORK_CITY WORK_STATE WORKERS APPLICATIONS
1 NEW YORK NY 43376 29032
2 SAN FRANCISCO CA 26431 13865
3 SEATTLE WA 60407 11743
4 SAN JOSE CA 29473 10164
5 AUSTIN TX 39087 9963
6 CHICAGO IL 16583 9388
7 SUNNYVALE CA 22879 9084
8 ATLANTA GA 14448 7947
9 MOUNTAIN VIEW CA 14211 7514
10 HOUSTON TX 11493 7325
11 IRVING TX 9626 6993
12 PLANO TX 8791 6371
13 BOSTON MA 11973 6282
14 REDMOND WA 18053 6281
15 DALLAS TX 12431 6259
16 SANTA CLARA CA 35084 5606
17 CHARLOTTE NC 9003 5542
18 BELLEVUE WA 9814 4782
19 SAN DIEGO CA 51295 4746
20 JERSEY CITY NJ 6238 4303
21 PHOENIX AZ 7271 4075
22 ALPHARETTA GA 4896 3840
23 MENLO PARK CA 8112 3795
24 LOS ANGELES CA 7355 3559
25 PALO ALTO CA 10114 3392
26 PHILADELPHIA PA 5860 3179
27 PITTSBURGH PA 7170 3138
28 CUPERTINO CA 24655 3036
29 TAMPA FL 5949 2945
30 FREMONT CA 3597 2911
31 DURHAM NC 4644 2891
32 CAMBRIDGE MA 4717 2640
33 COLUMBUS OH 3572 2610
34 IRVINE CA 6035 2587
35 SAN ANTONIO TX 4231 2401
36 SCHAUMBURG IL 2476 2389
37 RICHARDSON TX 10005 2230
38 WASHINGTON DC 2923 2174
39 BENTONVILLE AR 3112 2107
40 HILLSBORO OR 3798 1917
41 FRISCO TX 3900 1901
42 PLEASANTON CA 3603 1850
43 PRINCETON NJ 2064 1821
44 MINNEAPOLIS MN 2943 1739
45 RICHMOND VA 2361 1733
46 MIAMI FL 2765 1671
47 JACKSONVILLE FL 2248 1659
48 MCLEAN VA 1931 1626
49 RALEIGH NC 6870 1610
50 BROOKLYN NY 3309 1563
As can be seen, 4 out of the top 10 are in the Silicon Valley area of California (SAN FRANCISCO, SAN JOSE, SUNNYVALE, and MOUNTAIN VIEW.
Grouping by WORK_STATE
Following are the top 50 WORK_STATEs for which applications were submitted:
DISCLOSURE DATA, LCA_FY 2021-2021 (CASE_STATUS=CERTIFIED) NUMBER OF ROWS = 497,334 MEAN(SALARY) = NaN YEAR WORKERS APPLICATIONS 1 2021 1013997 497334 WORK_STATE WORKERS APPLICATIONS 1 CA 305662 104061 2 TX 117446 56564 3 NY 63860 40646 4 WA 95678 26580 5 NJ 37515 26208 6 IL 37507 23575 7 MA 35355 20010 8 GA 26455 17395 9 NC 29920 16663 10 PA 26940 15820 11 FL 23587 14811 12 MI 19774 14646 13 VA 23078 13188 14 OH 17170 11228 15 AZ 17516 9247 16 MD 9941 7502 17 MN 9795 7007 18 CT 12013 6134 19 MO 8380 5894 20 TN 9414 5683 21 CO 11582 5570 22 OR 9560 4763 23 IN 8317 4697 24 WI 8595 4424 25 AR 4444 3240 26 UT 4253 2969 27 SC 3418 2635 28 DC 3660 2537 29 KY 3020 2471 30 DE 3009 2432 31 IA 2739 2192 32 KS 2640 1881 33 RI 4267 1686 34 NE 2287 1604 35 LA 2598 1411 36 AL 1861 1377 37 NH 2184 1355 38 NV 1328 1193 39 OK 1345 1157 40 NM 1297 782 41 ID 758 711 42 MS 713 589 43 ME 493 436 44 GU 380 376 45 ND 421 356 46 WV 422 349 47 HI 299 296 48 SD 297 263 49 VT 221 219 50 MT 198 157 51 AK 117 103 52 WY 117 98 53 PR 72 70 54 MP 53 53 55 VI 26 20As can be seen, California has the most requested workers and applications by far and is followed by Texas, New York, Washington, and New Jersey.
Displaying Data for Multiple Employers by Quarter
As previously mentioned, a great deal of disclosure data from these LCAs is available on the Department of Labor website at this link. As can be seen, there is a single disclosure file for each year from 2008 through 2019. Starting in 2020, however, there are 4 files with disclosure data for each year, one file for each quarter. This change was likely due to the growing size of the files. It is possible to look at a subset of a year by specifying something other than 1 and 4 for the "Start Qtr" and "End Qtr" inputs. This option turned out to be useful in looking at a problem with the 2021 data that became apparent when comparing the results to a sheet of selected statistics put out by the Department of Labor at this link. The program was simply combining the 4 quarterly files to get the yearly results for 2021, same as it had done for 2020. However, unlike 2020, the program was coming up with much higher annual totals than the Department of Labor statistics.
In order to get a look at the contents of each of the 4 files for 2021, go to the default screen, clear the "Search EMPLOYER_NAME" input, select the Distribution tab, set the Variable input to DECISION DATE, set the "Bins / Binwidth(-)" input to -30, and click the Submit button. The -30 input sets the binwidth to 30 days or about one month. This should result in the following screen:
This shows the distribution of DECISION_DATE for all of the 2021 data. The LCA data is for fiscal year 2021 which goes from October 1, 2020 to September 30, 2021. Hence, the histogram seems to make sense for this period. However, you can then look at the distribution of DECISION_DATE in the file for quarter 1 by entering 1 into both the "Start Qtr" and "End Qtr" inputs and the clicking the Submit button. Similarly, you can enter 2, 3, and 4 into both of those inputs and click the Submit button to get the distribution of DECISION_DATE in the files for each of the other 3 quarters. The following histograms are the results:
In fact, it appear that the file for Quarter 1 does properly contain the data for Quarter 1. But it then appears that the Quarter 2 file contains the data form Quarters 1 and 2 and that the Quarter 3 file contains the data for Quarters 1 through 3. The Quarter 4 file, however, then appears to correctly contain the data just for Quarter 4.
This program currently handles this problem by showing the files as requested for each quarter for 2021 except for the case where the "Start Qtr" and "End Qtr" are set to 1 and 4, respectively. In that case, it's assumed that the user wants the entire year and this if achieved by combining the Quarter 3 file (which contains Quarters 1 through 3) and the Quarter 4 file. That is why the histogram for all of 2021 in the screen above looks correct. Still, the next section compares the results from the app with the results from the Department of Labor selected statistics.
(Note: You can still force is combination of all four files for a year by setting "Start Qtr" and "End Qtr" are set to 0 and 4, respectively).
The Department of Labor (DOL) has posted selected statistics on 2021 LCA data at this link. Below is a copy of that page1:
The following table shows the workers and applications from the DOL disclosure files as displayed by the Shiny application, followed by the corresponding applications from the "Applications Processed" section of the DOL selected statistics.
---------------------------- ----------------------- ------------ -------------- LCA DISCLOSURE DATA, FY 2021 DOL FILES DOL STATS DIFFERENCE ---------------------------- ----------------------- ------------ -------------- QTR CASE_STATUS WORKERS APPLICATIONS APPLICATIONS COUNT PCT ---------------------------- ----------------------- ------------ -------------- All TOTAL 1,071,503 532,201 528,902 3299 0.6 All CERTIFIED 1,013,997 497,334 507,235 -9901 -2.0 All CERT-WITHDRAWN 39,621 22,231 All WITHDRAWN 14,447 9,944 18,975 -9031 -47.6 All DENIED 3,438 2,692 2,692 0 0.0 1 TOTAL 204,787 80,622 78,823 1799 2.3 1 CERTIFIED 197,430 75,064 74,263 801 1.1 1 CERT-WITHDRAWN 3,554 3,333 1 WITHDRAWN 3,218 1,723 4,058 -2335 -57.5 1 DENIED 585 502 502 0 0.0 4 TOTAL 214,261 126,576 126,576 0 0.0 4 CERTIFIED 190,002 116,297 121,348 -5051 -4.2 4 CERT-WITHDRAWN 20,864 7,419 4 WITHDRAWN 2,744 2,343 4,711 -2368 -50.3 4 DENIED 651 517 517 0 0.0There is some problem comparing for the CASE_STATUS of CERTIFIED and WITHDRAWN since the selected statistics don't have a CERT-WITHDRAWN (CERTIFIED-WITHDRAWN) category. It's unclear whether the selected stats consider these as CERTIFIED or WITHDRAWN. However, the TOTAL and DENIED categories should be directly comparable. For TOTAL, the file data ranges from being equal to being 2.3 percent larger than the selected statistics. For DENIED, the file data and selected statistics are identical.
The following table shows the applications and workers from the DOL disclosure files as displayed by the Shiny application, followed by the corresponding workers from the remainder of the DOL selected statistics.
---------------------------- ----------------------- ------------ -------------- LCA DISCLOSURE DATA, FY 2021 DOL FILES DOL STATS DIFFERENCE ---------------------------- ----------------------- ------------ -------------- QTR CASE_STATUS APPLICATIONS WORKERS WORKERS COUNT PCT ---------------------------- ----------------------- ------------ -------------- All TOTAL 532,201 1,071,503 1,045,621 25,882 2.5 All CERTIFIED 497,334 1,013,997 1,027,735 -13,738 -1.3 1 TOTAL 80,622 204,787 197,346 7,441 3.8 1 CERTIFIED 75,064 197,430 193,543 3,887 2.0 4 TOTAL 126,576 214,261 211,780 2,481 1.2 4 CERTIFIED 116,297 190,002 208,385 -18,383 -8.8 ---------------------------- ----------------------- ------------ -------------- QTR WORK_STATE APPLICATIONS WORKERS WORKERS COUNT PCT ---------------------------- ----------------------- ------------ -------------- All CA 111,137 318,747 309,011 9,736 3.2 All TX 59,976 122,195 118,397 3,798 3.2 All WA 27,943 101,796 96,688 5,108 5.3 All NY 43,420 68,354 64,602 3,752 5.8 All IL 26,252 40,930 38,930 2,000 5.1 All NJ 27,586 39,113 37,919 1,194 3.1 All MA 21,480 38,488 36,638 1,850 5.0 All NC 17,640 31,382 30,239 1,143 3.8 All PA 16,860 28,460 27,227 1,233 4.5 All GA 18,445 27,991 26,770 1,221 4.6 --------------------------------- ------------------ ------------ -------------- QTR SOC_TITLE APPS WORKERS WORKERS COUNT PCT --------------------------------- ------------------ ------------ -------------- All SW DEVELOPERS, APPLICATIONS 158,937 267,735 272,186 -4,451 -1.6 All SW DEVELOPERS, SYSTEMS SW 30,284 91,850 92,454 -604 -0.7 All COMPUTER OCCUPATIONS, OTHER 7,646 76,232 166,935 -90,703 -54.3 All ELECTRONICS ENG. X COMPUTER 7,791 61,950 63,737 -1787 -2.8 All COMPUTER SYSTEMS ANALYSTS 30,135 49,895 50,217 -322 -0.6 All COMPUTER & INFO SYSTEMS MGR 10,397 24,603 24,748 -145 -0.6 All COMPUTER SYSTEMS ENG./ARCH. 16,379 23,335 All SW QUALITY ASSUR. ENG./TESTR 12,358 21,064 All BUSINESS INTELLIGNC ANALYSTS 9,412 20,919 All OPERATIONS RESEARCH ANALYSTS 8,434 20,017 20,217 -200 -1.0 ---------------------------- ----------------------- ------------ -------------- QTR EMPLOYER_NAME APPLICATIONS WORKERS WORKERS COUNT PCT ---------------------------- ----------------------- ------------ -------------- All AMAZON.COM 9,302 93,247 83,963 9,284 11.1 All INFOSYS 5,634 92,796 92,100 696 0.8 All QUALCOMM TECHNOLOGIES 2,016 53,915 53,603 312 0.6 All APPLE 4,239 39,736 38,994 742 1.9 All AMAZON WEB 2,758 28,366 25,796 2,570 10.0 All MICROSOFT 7,329 19,763 19,763 0 0.0 All ORACLE AMERICA 999 14,429 14,412 17 0.1 All HCL AMERICA 4,113 14,158 14,070 88 0.6 All WIPRO 2,859 13,628 13,361 267 2.0 All NVIDIA 617 13,345 13,319 26 0.2 ---------------------------- ----------------------- ------------ -------------- LCA DISCLOSURE DATA, FY 2021 DOL FILES DOL STATS DIFFERENCE ---------------------------- ----------------------- ------------ --------------As previously explained, only the TOTAL category is directly comparable for CASE_STATUS. For TOTAL, the file data ranges from being 1.2 percent to 3.8 percent larger than the selected statistics. Similarly for the WORK_STATE totals, the file data ranges from being 3.1 percent to 5.8 percent larger than the selected statistics.
As opposed to the WORK_STATE values which are directly comparable, there's a problem with comparing the SOC_TITLE values. The selected statistics sheet says that its numbers are based on the SOC Codes. The Shiny app, however, displays the SOC titles which are related to the SOC codes. However, there does appear to be a one-to-one correspondence. For example, it appears that the "Statisticians" category in the selected statistics includes both STATISTICIANS and BIOSTATISTICIANS categories in SOC_TITLE. Still, the comparisons that could be done show that, except for the category "COMPUTER OCCUPATIONS, OTHER", the numbers were fairly close. In this case, the numbers from the selected statistics are larger by 0.6 percent to 2.8 percent. However, this relationship is not guaranteed since the numbers are not precisely comparable.
Similarly to SOC_TITLE, the EMPLOYER_NAMEs are not always precisely comparable. The selected statistics sheet says that the employers are based on unique FEINs. However, the 2021 LCA Record Layout says the following:
"The following form items are not included in the public disclosure file because they contain Personally Identifiable Information (PII): Employer’s Federal Employer Identification Number (FEIN), Attorney’s FEIN, and Attorney’s State Bar Number.
Hence, these employers are not directly comparable though both the files and selected statistics listed exactly 19,763 workers for Microsoft. Also, except for AMAZON.COM and AMAZON WEB, the file data ranges from being equal to being larger by just 2.0 percent.
In summary, 21 of the values in the files and selected statistics in the table are directly comparable. These include 9 TOTALs, 3 DENIEDs, and 12 WORK_STATEs. Of those, the file data ranges from being equal (in 4 of them) to being larger by a maximum of 5.8 percent. This would suggest that both sources are looking at the same data but that the files have been updated since the selected statistics were generated.
Displaying Data for a Single Employer and Multiple Years
If the "Start Year" input is set to 2008, the "End Year" input is set to 2021", the "Search EMPLOYER_NAME" input is set to AMAZON, the "Group by" input is set to EMPLOYER_NAME, the "Sort by" input is set to APPLICATIONS, and the Submit button is clicked, the following output will be displayed:
LCA DISCLOSURE DATA, FY 2008-2021
(CASE_STATUS=CERTIFIED, EMPLOYER_NAME=AMAZON)
NUMBER OF ROWS = 59,351
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2008 703 439
2 2009 367 367
3 2010 1678 549
4 2011 1293 670
5 2012 3811 879
6 2013 3573 1166
7 2014 4715 1660
8 2015 8717 2243
9 2016 7679 3012
10 2017 12205 4926
11 2018 15642 6289
12 2019 15260 9373
13 2020 19462 15151
14 2021 120261 12627
EMPLOYER_NAME WORKERS APPLICATIONS
1 AMAZON.COM 123556 32266
2 AMAZON CORPORATE 36908 12882
3 AMAZON WEB 35899 8669
4 AMAZON DEVELOPMENT CENTER U.S 7919 1240
5 AMAZON FULFILLMENT 2089 1128
6 AMAZON GLOBAL RESOURCES 1305 841
7 AMAZON DATA 2862 511
8 AMAZON.COM.DEDC 806 377
9 AMAZON ROBOTICS 498 300
10 AMAZON 476 233
11 AMAZON.COM.KYDC 587 168
12 AMAZON ADVERTISING 853 164
13 AMAZON DIGITAL 142 65
14 AMAZON MEDIA GROUP 74 61
15 AMAZON.COM.AZDC 189 58
16 AMAZON PAYMENTS 172 46
17 AMAZON CAPITAL 131 38
18 AMAZON.COM.INDC 195 36
19 AMAZON MECHANICAL TURK 86 33
20 AMAZON FRESH 23 23
21 AMAZON STUDIOS 78 21
22 AMAZON.COM.NVDC 88 19
23 AMAZON REGISTRY 25 19
24 AMAZON TECHNOLOGIES 32 18
25 AMAZON CORPORTATE 87 13
26 AMAZON PRODUCE NETWORK 11 11
27 AMAZON.COM.CA 125 11
28 AMAZON COPORATE 21 9
29 AMAZON.COM.KSDC 20 7
30 AMAZON COM 6 6
31 AMAZON.COM.SERVICES 5 5
32 AMAZON.COM DEDC 6 4
33 AMAZONIC VENTURES 3 3
34 AMAZON HUT 3 3
35 AMAZON WATCH 3 3
36 AMAZON PHARMACY 3 3
37 AMAZON CONSULTING 2 2
38 AMAZON FULFILLMENT SERVICES NC 2 2
39 AMAZON CONSTRUCTION 2 2
40 A2Z DEVELOPMENT CENTER INC D/B/A AMAZON 2 2
41 AMAZONIA CONSULTING GROUP 2 2
42 AMAZON FUFILLMENT 2 2
43 AMAZON RESTAURANT & BAR 2 2
44 AMAZON FULFILLMENT SVCS 2 2
45 AMAZON.COM SERVICE 2 2
46 AMAZON RETAIL 2 2
47 AMAZON WEB SERVICE 2 2
48 AMAZON DEVELOPMENT CENTER US 2 2
49 AMAZON BUILDERS HARDWARE 1 1
50 AMAZON SOCIETY 1 1
51 AMAZON PROFESSIONAL CARE 1 1
52 AMAZON HERBAL 1 1
53 AMAZON CARGO 1 1
54 AMAZON.COM.KYDC.LLC 1 1
55 AMAZON LOGISTICS GROUP 1 1
56 AMAZON CAPTAL 1 1
57 AMAZON FULFILLMENT SERVICE 1 1
58 AMAZON.COM AZDZ 5 1
59 AMAZON CONSERVATION TEAM 1 1
60 AMAZON AERO 1 1
61 AMAZON CORPORTE 1 1
62 AMAZON ROBOTICS LLC (KIVA) 1 1
63 AMAZON DEVELOPMENT CENTER 1 1
64 AMAZON WEB SERVICES INC. 1 1
65 AMAZON MASONRY 1 1
66 HAMAZON G2 1 1
67 AMAZON WEB SEERVICES 1 1
68 AMAZON CORPROATE 1 1
69 AMAZON.COM.DEDC.INC 1 1
70 AMAZON.COM SERVCES 1 1
71 AMAZON WEB SERVICES INC.(U+2020) 1 1
72 AMAZON.WEB 1 1
73 AMAZON.COM SERVICES INC.(U+2020) 1 1
74 AMAZON.COM WEB 1 1
75 AMAZON.COM SERVICES LLC (F/K/A AMAZON.CO 1 1
76 AMAZON WEB SERVICES.INC 1 1
77 AMAZON DEVELOPEMENT CENTER U.S 1 1
78 AMAZON DEV CENTER U.S 1 1
79 AMAZON.COM SERVICES LLC (HONG KONG BRANC 20 1
80 AMAZON SOLUTIONS 1 1
81 AMAZON VALLEY ENGINEERING 1 1
As can be seen, there were 81 unique employer names that contained the word AMAZON. It could take some research to determine which, if any, of these employers are not associated with the Amazon in Seattle, run by Jeff Bezos. However, it does appear that the great majority of the Applications and Workers, including all those until at least #17 (AMAZON CAPITAL) are associated with that company. Hence, it's probably not critical to filter out those that are not associated.
In any event, clicking on the "Plot Year" tab and setting the "X From, To, Step, Tick" input to 2008,2021,2,1 and setting the "Y From, To, Step, Tick" input to 0,16000,2000 will display the following screen:
This graph shows that the number of certified applications submitted by Amazon rose rapidly from 2,243 in 2015 to 15,151 in 2020 before backing off to 12,627 in 2021.
Now, clicking on the "Plot Groups" tab and setting the "X From, To, Step, Tick" input to 2008,2021,2, setting the "Y From, To, Step, Tick" input to 0,12,2, and checking the "Scale Y in thousands" checkbox will display the following screen:
The above graph shows the number of certified applications submitted for the 6 EMPLOYER_NAMEs with the most applications from 2008 through 2021. As can be seen, "AMAZON CORPORATE" was the most common EMPLOYER NAME from 2008 through 2017 and "AMAZON.COM" was the most common EMPLOYER_NAME from 2018 through 2021.
Clicking on the Distribution tab now will display the following screen:
The above is a horizontal bar chart of the 10 EMPLOYER NAMEs with the most WORKERS requested from 2008 through 2021. Unchecking the Workers checkbox and checking the Reverse checkbox will change this to showing the 10 EMPLOYER_NAMES with the most certified APPLICATIONS submitted from 2008 through 2021 and flip the bar chart such that the largest EMPLOYER_NAMEs are at the bottom as shown below:
Displaying Data for Multiple Employers and Years
If the "Search EMPLOYER_NAME" input is set blank (with "Start Year" set to 2008, "End Year" set to 2021", "Group by" set to EMPLOYER_NAME, and "Sort by" set to APPLICATIONS) and the Submit button is clicked, the following output will be displayed:
LCA DISCLOSURE DATA, FY 2008-2021
(CASE_STATUS=CERTIFIED)
NUMBER OF ROWS = 6,309,176
MEAN(SALARY) = NaN
YEAR WORKERS APPLICATIONS
1 2008 706660 404436
2 2009 483203 266230
3 2010 489177 262688
4 2011 671837 307936
5 2012 820431 352668
6 2013 882643 382951
7 2014 925398 455144
8 2015 1161218 547278
9 2016 1186143 569645
10 2017 1070024 545694
11 2018 1199745 579449
12 2019 981669 592102
13 2020 838627 545621
14 2021 1013997 497334
EMPLOYER_NAME WORKERS APPLICATIONS
1 INFOSYS 353255 199391
2 TATA CONSULTANCY 219629 128532
3 DELOITTE CONSULTING 601212 82531
4 COGNIZANT TECHNOLOGY SOLUTIONS US 104397 81139
5 WIPRO 498773 67835
6 MICROSOFT 102132 66221
7 ACCENTURE 101942 61907
8 ERNST & YOUNG U.S 207689 54506
9 GOOGLE 51293 51264
10 CAPGEMINI AMERICA 91357 49368
11 IBM INDIA PRIVATE 49384 42806
12 TECH MAHINDRA (AMERICAS) 64450 42743
13 HCL AMERICA 184593 42733
14 LARSEN & TOUBRO INFOTECH 35642 32397
15 AMAZON.COM 123556 32266
16 IBM 28281 27481
17 INTEL 40986 22651
18 FACEBOOK 43893 20925
19 COGNIZANT TECHNOLOGY SOLUTIONS U.S 466066 20876
20 APPLE 173652 20622
21 JPMORGAN CHASE 20395 19381
22 UST GLOBAL 22354 19131
23 DELOITTE & TOUCHE 121877 18359
24 GOLDMAN SACHS 19429 14606
25 WAL-MART ASSOCIATES 13326 13210
26 AMAZON CORPORATE 36908 12882
27 SATYAM COMPUTER 27432 12831
28 MPHASIS 179904 12647
29 COMPUNNEL SOFTWARE GROUP 13612 12573
30 IGATE TECHNOLOGIES 133988 12050
31 ORACLE AMERICA 115456 11742
32 NTT DATA 14417 11494
33 L&T 11071 11071
34 SALESFORCE.COM 12824 11009
35 PRICEWATERHOUSECOOPERS 206759 11000
36 HEXAWARE TECHNOLOGIES 12029 10435
37 KPMG 14623 9812
38 CISCO SYSTEMS 92355 9655
39 MINDTREE 86986 9555
40 FUJITSU AMERICA 48913 9492
41 QUALCOMM TECHNOLOGIES 164063 9428
42 BANK OF AMERICA N.A 12907 9257
43 PATNI AMERICAS 66224 9230
44 AMAZON WEB 35899 8669
45 CUMMINS 11607 8514
46 PRICEWATERHOUSECOOPERS ADVISORY 38384 8504
47 VMWARE 8861 8400
48 SYNECHRON 10821 8379
49 SYNTEL 114644 8169
50 EBAY 11395 7995
The above output shows only the top 50 of over 900 EMPLOYER_NAMEs. Clicking on the "Plot Year" tab and setting the "X From, To, Step, Tick" input to 2008,2021,2 and checking the "Scale Y in thousands" input will display the following screen:
As can be seen, the total number of appications have dropped off during COVID from their prior high in 2019. Now, clicking on the "Plot Groups" tab and setting the "X From, To, Step, Tick" input to 2008,2021,2, setting the "N Groups" input to 7, and checking the "Scale Y in thousands" checkbox will display the following screen:
This shows the 7 EMPLOYER_NAMEs that had the most applications over the entire period from 2008 to 2021. As can be seen, 6 out of the 7 are consulting companies with the lone exception being Microsoft. Changing the "Base Year (0=all)" input from 0 to 2021 will display the following screen:
This shows the 7 EMPLOYER_NAMEs that had the most applications over just the year 2021. As can be seen, now only 4 of the 7 are consulting companies with Amazon.com and Google joining Microsoft in the top 7. This makes some sense that the consulting companies which bring in more foreign workers would become less dominating during COVID. In any event, clicking on the Distribution tab now and unchecking the Workers checkbox will display the following screen:
This shows the 10 EMPLOYER_NAMEs with the most applications over the entire period from 2008 to 2021.
Using WAGE_PWAGE to Check Wage Against Prevailing Wage
This page on the U.S. Department of Labor website states the following:
The H-1B employer must pay its H-1B worker(s) at least the “required” wage which is the higher of the prevailing wage or the employer’s actual wage (in-house wage) for similarly employed workers.
Hence, it would appear that the wage offered to an H-1B worker should never be lower than the prevailing wage. For this reason, this application creates the variable WAGE_PWAGE equal to WAGE_RATE_OF_PAY_FROM (the minimum wage being offered) divided by PREVAILING_WAGE (the prevailing wage). If the wage offered is never less than the prevailing wage, then WAGE_PWAGE should never be less than 1. One problem is that WAGE_RATE_OF_PAY_FROM and PREVAILING_WAGE are not guaranteed to be for the same time period. Each can be per hourly, weekly, bi-weekly, monthly, or yearly. For that reason, this app also creates the variables SALARY and PSALARY which convert the actual and prevailing wage to yearly values.
Setting "Start Year" and "End Year" to 2021, setting "Search EMPLOYER_NAME" blank, setting "Sort by" to WAGE_PWAGE, clicking the Ascending radio button, and clicking the Submit button will display all of the applications for 2021 in the order of ascending WAGE_PWAGE. In the "Show" input section, it helps to also check PREVAILING_WAGE, PW_UNIT, and WAGE_UNIT and also uncheck SOC_TITLE (for space). Finally, adding PSALARY, and SALARY to the "Show (other)" input before WAGE_PWAGE will display the following screen:
In all, the output shows 54 lines with WAGE_PWAGE less than 1. However, all but the first 7 are above 0.9 so it is possible that these numbers are correct. Further investigation would be necessary to determine which of these numbers are correct and, if so, how it possible that they were certified when the wage is less than the prevailing wage. In any case, clicking the Descending radio button, and clicking the Submit button will display the following screen.
These values appear much stranger with the first 13 values of WAGE_PWAGE being 2000 or above. The value of WAGE_RATE_FROM for these 13 lines appear to be yearly salaries but the WAGE_UNIT is set to HOUR. Hence, it appears that this was and error and WAGE_UNIT should have been set to YEAR. The value of 2999.00 for WAGE_RATE_FROM in the 14th line is tougher to figure out. It could be that 2999.00 is a weekly salary or it could be that it was entered incorrectly and should have been 29.99 as an hourly salary. This latter error does not appear to be that uncommon. Following are the lines containing the 8 values of WAGE_PWAGE that are closest to 10:
YEAR CASE_STATUS TOTAL_WORKERS EMPLOYER_NAME WORK_CITY WORK_STATE WAGE_RATE_FROM WAGE_UNIT PREVAILING_WAGE PW_UNIT SALARY PSALARY WAGE_PWAGE
47 2021 CERTIFIED 1 OBVIOUSLEE MARKETING BROOKLYN NY 450000.00 YEAR 44179.00 YEAR 450000.0 44179.00 10.185835
48 2021 CERTIFIED 1 PAYPAL SAN JOSE CA 1480000.00 YEAR 145725.00 YEAR 1480000.0 145725.00 10.156116
49 2021 CERTIFIED 1 UNIVERSITY OF GEORGIA ATHENS GA 230000.00 YEAR 22910.00 YEAR 230000.0 22910.00 10.039284
50 2021 CERTIFIED 1 SCIENBIZIP P.C LOS ANGELES CA 865000.00 YEAR 86403.00 YEAR 865000.0 86403.00 10.011226
51 2021 CERTIFIED 1 QUANTUM TECHNOLOGIES SUNNYVALE CA 2080000.00 YEAR 208000.00 YEAR 2080000.0 208000.00 10.000000
52 2021 CERTIFIED 1 AUTOMATION TECHNOLOGIES HOUSTAN TX 919728.00 YEAR 91978.00 YEAR 919728.0 91978.00 9.999435
53 2021 CERTIFIED 1 BAUER COMPRESSORS NORFOLK VA 632632.80 YEAR 63294.00 YEAR 632632.8 63294.00 9.995146
54 2021 CERTIFIED 1 NORTHWESTERN UNIVERSITY EVANSTON IL 405000.00 YEAR 40830.00 YEAR 405000.0 40830.00 9.919177
The prevailing wages look fairly reasonable, ranging from $22,910 to $208,000. The $22,910 is for a SOC_TITLE of "BUSINESS TEACHERS, POSTSECONDARY" and the $208,00 is for a SOC_TITLE of "SOFTWARE DEVELOPERS, SYSTEMS SOFTWARE". However, the values of WAGE_RATE_FROM range from $230,000 to $2,080,000. It would seem possible that the decimal was placed one place to the right of where it should have been in these numbers. This is especially the case for $2,080,000 which happens to be exactly 10 times the prevailing wage.
Just as sorting by WAGE_PWAGE turned up errors in the data in the prior section, sorting by other variables can turn up errors as well. That's because erroneous data can often turn up at either end of a sort. For example, setting "Sort by" to WORK_CITY and clicking the "Ascending" button displays the following screen:
It's also possible to find errors using regex patterns in the Search inputs. For example, setting "Search WORK_CITY" to "[A-Z]+[0-9]+[A-Z]*" (without the quotes) will display the following screen:
This regex pattern will find all cases of WORK_CITY that begin with one or more letters, followed by one or more numbers, and then optionally followed by letters. In this way, it should find values of WORK_CITY that contain numbers but were not caught in the previous sort. This regex pattern could probably be improved to handle special characters that are not alphanumric.
Note: This application is under continuing development. If anyone should run into any issues or have any suggestions for additional features, feel free to let me know via the Contact box at the bottom of this page.
H-1B and Related Data