R. LONGSTAFF, Chief Judge, District
The above-captioned action is before the Court for its findings of fact and conclusions of law regarding plaintiff's claim of disparate impact discrimination. The parties filed proposed findings of fact and final argument on September 24, 2004, and the matter is now considered fully submitted.
FINDINGS OF FACT
1. Plaintiff Equal Employment Opportunity Commission ("EEOC") is the agency of the United States Government charged with the administration and enforcement of Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. §§ 2000e et seq. ("Title VII").
2. Sections 2000e-5(f)(1) and (3) and 2000e-6(e) of the United States Code authorize the EEOC to bring this action, which is based upon a charge of discrimination filed by Paula Liles against Defendant Dial Corporation ("Dial").
3. Dial, an international consumer products company, employed approximately 2900 people during fiscal year 2002.
4. Dial owns and operates a plant in Ft. Madison, Iowa ("the Plant") that produces a variety of canned meat products.
5. The Fort Madison plan employs approximately 380 hourly union employees in a three-shift operation.
6. All new employees are trained at some point on the sausage making jobs because when more senior employees bid off the sausage making team, the opening is filled by the most senior new hire.
7. Thus, the first permanent job for new hires is on the sausage making team either in loading or unloading. The new hires remain on the sausage making team until they have sufficient seniority to bid off of the team. Most new hires spend a substantial period of time on the sausage making team.
8. The employees on the Frank-A-Matic, or "loading" end of the sausage making operation must lift and carry rods holding approximately 35 pounds of uncooked sausage links to a moving conveyor apparatus and place or lift the rods onto notches in the conveyor. The lowest notch is 39 inches above floor level, whereas the highest notch is more than 65 inches above floor level.
9. The employees in the Smokehouse, or "unloading" end of the operation, must lift and carry the rods of cooked sausages from the moving conveyor from either a 61 inch or 33 inch height and carry the loaded rods approximately 15 feet to the peeler table.
10. In January 2000, the Plant implemented a physical ability test, known as the Work Tolerance Screening ("WTS").
11. Two female employees, Human Resources Manager Jean Farrington, and Martha Lutenegger, an occupational health nurse, conceived of the test to pre-screen for physical ability before placing employees in the entry level sausage making jobs.
12. Both women hoped that such a test could help reduce the number of employee injuries attributed to the physically demanding nature of sausage jobs. They also viewed the test as a way of measuring applicants' capabilities in areas of strength, endurance and other physical demands.
13. Ms. Lutenegger consulted an occupational therapist, Molly Sichterman, at a hospital in Ft. Madison. Ms. Sichterman evaluated the requirements of the sausage making jobs and designed the WTS test.
14. The WTS is a strength test during which the applicant is required to lift a bar with 35 pound weights off of a table, carry it approximately ten feet to a plywood frame, place it on the frame at a height of 35 inches, return it to the table, lift it again and carry it to the frame, this time lifting it to approximately 67 inches to place it on the frame, remove it from the frame, return it to the table, and repeat the cycle on a continuous basis for seven minutes.
15. The test has at all times been administered by female occupational therapists and Ms. Lutenegger. In the first year, Ms. Lutenegger decided who passed the test; beginning in 2001 the decision was a joint decision of the occupational therapists and Ms. Lutenegger.
16. Prior to implementation of the WTS, an applicant was selected from a multi-step process including a review of his/her application, a written test, a personal interview, a group problem solving exercise and final interview. Candidates who successfully completed each step of the process were made a job offer contingent on a medical examination and a drug test.
17. After January 1, 2000, the hiring process remained the same, except for the use of the WTS, which was given after the contingent job offer had been made.
18. During the three year period from 1997-1999, before the WTS was implemented, Dial hired 162 persons to work at the Plant. Of that group, 75, or 46%, were women.
19. Since the test was implemented, between January 2000 and September 30, 2003, 162 male applicants were tested. Of these male applicants, 157, or 97%, passed and were hired by Dial. During the same period of time, 68 female applicants were tested; 26, or 38%, passed and were hired.
20. The disparity between the female hire rate and the male hire rate for the four-year period following implementation of the WTS was nearly ten standard deviations.
Facts Relevant to Content Validity
21. Employee testing is governed by EEOC-issued federal regulations known as the "Uniform Guidelines on Employee Selection Procedures."
22. The Uniform Guidelines define content validity as "data showing that the content of the selection procedure is representative of important aspects of performance on the job for which the candidates are to be evaluated." 29 C.F.R. § 1607.5(B).
23. Dial did not perform any validity studies as required by the Uniform Guidelines prior to implementation of the WTS.
24. A test that is significantly more difficult than the job is not content valid. For example, a test requiring an applicant to type 100 words per minute would not be content valid for a job that required typing 40 words per minute.
25. The loader job in the Smokehouse requires 1.25 lifts per minute.
26. Women who completed the WTS usually performed at least 6 lifts per minute.
27. Constant lifting is not required to perform the loader job; there is a break between each lift when the worker returns from loading the rod on the machine. Workers stop and talk to one another for a few seconds while performing the lifts. Furthermore, the chains on the actual machines used in the Smokehouse drop at an angle so that the top rung for the rod of sausages is approximately five feet off of the floor.
28. The WTS requires constant carrying and lifting of 35 pounds between the table and the plywood frame.
29. Certain female applicants who were marked as "pass" on the WTS score sheet were later "failed" by Ms. Lutenegger.
30. Certain male applicants whose score on body mechanics was rated as "fair" passed, whereas certain female applicants who were given the same rating failed.
31. Male applicants noted to have arched their backs while lifting nevertheless passed the WTS, whereas certain female applicants noted for the same characteristic failed.
32. Male applicants who raised up on their toes during the WTS passed the test, whereas at least one woman who raised up on her toes failed.
33. Neither Dial nor its expert tried to determine whether incumbent employees could pass the WTS as it was implemented.
34. Dial never considered any changes to the WTS that could lessen the disparate impact.
Facts Related to Criterion Validity
35. The Uniform Guidelines define criterion validity as "empirical data demonstrating that the selection procedure is predictive of or significantly correlated with important elements of job performance." 29 C.F.R. § 1607.5(B).
36. Injury data was categorized by the experts in several ways: (i) total injuries; (ii) whether the injury appeared to be strength related; (iii) whether the injury was required to be reported pursuant to the federal Occupational Safety and Health Act, and (iv) whether the injury resulted in workers compensation cost to the company.
37. Following implementation of the WTS, the overall number of injuries suffered by people working in the Smokehouse — without regard to the hire date of the injured employee — dropped.
38. Non-strength related injuries dropped more than strength related injuries.
39. There was a drop in total injuries, in strength related injuries, compensation cost injuries, and in OSHA recordable injuries for new hires after 1997.
40. Beginning in late 1996, Dial took various steps to reduce injury rates. For example, in 1996, Dial implemented an ergonomic job rotation program for various jobs in the Smokehouse; in April, 2000, it added a team approach to increase job rotation; also in 2000, Dial lowered the height of a bar on the Frank-O-Matic machines, and conducted periodic ergonomic and safety audits, including instituting an annual "back injury prevention program."
41. Individuals hired during 1997 were far more likely to be injured during each of their first three years of employment than individuals hired between 1998 and 2003.
42. Specifically, 1997 hires suffered 70 injuries during their first three years of employment, as compared to 6 for 1998 hires; 16 for 1999 hires; 16 for 2000 hires; 3 for 2001 hires; 4 for 2002 hires and 1 for 2003 hires.
43. In addition, 1997 hires suffered 48 strength related injuries during their first three years of employment, compared to 3 for 1998 hires; 10 for 1999 hires; 10 for 2000 hires; 3 for 2001 hires; 1 for 2002 hires and 0 for 2003 hires.
44. Persons hired in 1997 suffered 20 strength related injuries resulting in workers compensation costs during their first three years of employment, compared to 1 for 1998 hires; 3 for 1999 hires; 3 for 2000 hires; 3 for 2001 hires; 1 for 2002 hires; and 0 for 2003 hires.
45. The total number of injuries during the first three years of employment for persons hired in 1998 and 1999 — before the implementation of the WTS — was virtually the same as the totals for the persons hired after the implementation of the WTS.
46. The same is true for strength related injuries before and after 2000, and for compensation cost injuries before and after 2000.
47. When injury rates are compared using the exposure time of new hires in the Smokehouse, the rate dropped after 1997, but remained virtually the same from 1998 through 2003.
48. The injury rate per year of exposure for strength related injuries resulting in workers compensation costs, taking exposure data into account was 0.48 for 1997 hires.
49. Injury rates per year of exposure for strength related workers compensation injuries were 0.09 for 1998 hires; 0.14 for 1999 hires; 0.05 for 2000 hires; 0.16 for 2001 hires; 0.06 for 2002 hires and 0.00 for 2003 hires during their first three years of employment.
50. When the rates are broken down by year of hire, the rate of reduction of strength related workers compensation injuries between 1997 and 1998 is 80% (0.48 in 1997 compared to 0.09 in 1998). The average rate of injury for 1998-2003 is 0.8.
51. Injury rates for male and female employees prior to the implementation of the WTS were virtually identical.
52. The 43 men hired in 1997 suffered 40 injuries during their first three years of employment; the 45 women hired during 1997 suffered 30 injuries during the first three years. In 1998, the numbers were 4 injuries for 16 males and 2 injuries for 4 female hires; the 28 males hired in 1999 suffered 11 injuries during their first three years; the 26 females suffered 5 injuries during that period.
53. When injury rates are calculated taking into account the exposure data, males hired in 1997 had an injury rate of 0.45 per year, while women had a rate of 0.39. For 1998 the rates were 0.11 for men and 0.5 for women. For 1999 the male rate was 0.19 and the female rate was 0.16.
APPLICABLE LAW AND DISCUSSION
"Title VII forbids the use of employment tests that are discriminatory in effect unless the employer meets `the burden of showing that any given requirement [has] . . . a manifest relationship to the employment in question.'" Albemarle Paper Co. v. Moody, 422 U.S. 405, 425 (1975) (quoting Griggs v. Duke Power Co., 401 U.S. 424, 432 (1971)). As set forth above, the Court finds the EEOC has proven by a preponderance of the evidence that Dial's implementation and continued use of the WTS has a discriminatory effect on women. Specifically, between 1997 and 1999, prior to the test's implementation, 75 of the 162 persons, or 46%, hired by Dial were women. Jt. Exh. 1, Introduction at 23. Between January 2000 and September 2003, 183 persons were deemed eligible for hire after taking the WTS. Id. Only 26 of those individuals, or 14%, were women. Of the 162 males tested between January 2000 and September 2003, 157, or 97%, passed. Of the 68 women tested during the same time period, although 55, or 81% were able to complete the WTS, only 26, or 38%, were deemed by Dial to have passed the test. Id. The disparity between the percentages of women and men hired between January 2000 and September 2003 is nearly ten standard deviations. Tr. Tr. of Michael Campion, Ph.D., at 21. The Uniform Guidelines on Employee Selection Procedures, ("Uniform Guidelines") provide: "A selection rate for any race, sex, or ethnic group which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact. . . ." 29 C.F.R. § 1607.4(d). It is clear to the Court that a test which results in an 97% pass rate for male applicants and a 38% pass rate for female applicants produces a discriminatory effect.
It is not clear from the record whether all of these persons in fact went on to work for Dial.
Not all of the trial has been transcribed. References to ("Tr. Tr.") refer to the partial trial transcript of a particular witness' testimony.
These guidelines were promulgated by the EEOC to help enforce the mandates of Title VII of the Civil Rights Act of 1964, as amended, inthe context of employment selectionprocedures. 29 C.F.R. § 1607.2(A).
The Court acknowledges that the Guidelines are not entitled to the full force of law, but rather, "`entitled to great deference.'" Allen v. Entergy Corp., 181 F.3d 902, 905 (8th Cir. 1999) (quoting Griggs, 401 U.S. at 433-34). Nevertheless, the disparity evidenced by these statistics is so large that even Dial has all-but-conceded the EEOC's prima facie case.
The burden now shifts to Dial to demonstrate the WTS "is related to safe and efficient job performance and is consistent with business necessity." Firefighters' Inst. For Racial Equality v. City of St. Louis, 220 F.3d 898, 904 (8th Cir. 2000). To meet this burden, Dial has attempted to show the test is both content-valid and criterion-valid. See, e.g., Uniform Guidelines, 29 C.F.R. § 1607.5 (to establish a pre-employment test's validity under the Guidelines, "users may rely upon criterion-related validity studies, content validity studies or construct validity studies").
It is undisputed Dialdid not attempt to establish the test's validity under the Guidelines prior to implementing the WTS in January 2000. J. Exh. 10.
A. Whether WTS is Content-Valid
The Guidelines provide that "Evidence of the validity of a test or other selection procedure by a content validity study should consist of data showing that the content of the selection procedure is representative of important aspects of performance on the job for which the candidates are to be evaluated." Uniform Guidelines, 29 C.F.R. § 1607.5(B). To establish content-validity at trial, Dial's expert witnesses, Andrew Jackson, P.E.D., and David Jones, Ph.D., performed independent task analyses of the loading and unloading jobs and concluded the WTS was a close simulation of the tasks performed on the job. Specifically, Dr. Jackson testified at trial that he did not "know how [Dial] could make [the WTS] more valid," based on the fact the heights of the bars and the weight required to be lifted were identical to those used on the job. Jackson Tr. Tr. at 67. The Court finds persuasive, however, the testimony of the EEOC's expert, Dr. Campion, that a crucial aspect of the WTS is more difficult than the sausage making jobs themselves. The sausage-making jobs require only 1.25 lifts per minute, whereas applicants taking the WTS performed up to six lifts per minute. Campion Tr. Tr. at 23. This fact was underscored by testimony from Michael Kruse, a current Dial employee who has worked in the Smokehouse. Kruse testified at trial that the job is actually easier than the WTS because the lifting is not constant; workers are given a break between each lift. Kruse Tr. Tr. at 14.
Kruse also noted other differences between the WTS and actualwork inthe Smokehouse. For example, the actual machines used in the Smokehouse have cups to hold the rod of sausages., and a chain that descends at an angle so that the top rung for the rod of sausages is approximately five feet. Kruse Tr. Tr. at 13-14, 21, 27.
In its post-trial memorandum, Dial urges the Court to consider the fact that Dial employees must perform these 1.25 lifts per minute over a 5 to 7 hour period, whereas the test taker need only perform her lifts for 7 minutes. Although the Court agrees that an individual arguably should be able to perform the lifts faster over a 7 minute duration than over a 5 to 7 hour duration, this fact does not change the concerns raised by Dr. Campion that the pace at which applicants perform the WTS is not reflective of the pace they must eventually maintain on the job.
Although not addressed at trial, it appears reasonable to the Court that the uncertainty derived from Dial's directive to applicants that they perform "at their own pace," see, e.g., Jackson Tr. Tr. at 62, could infact encourage some applicants to increase their pace, fearing that Dialmaybe silently monitoring the number of lifts completed.
Furthermore, even in Dial's post-hoc attempts to validate the test, its experts did not try to determine whether current employees who had passed the test performed better than those who had never take the test. See, e.g., Allen v. Entergy Corp., 181 F.3d 902, 905 (8th Cir. 1999) (defendant presented evidence that "[p]erformance evaluations were mathematically correlated with the test results and confirmed that higher test scores indicated a likelihood of better job performance." As noted by the EEOC, Dial's experts never asked any employee hired before 2000 to try the test. Jones Tr. Tr. at 52.
Perhaps most importantly, Dial failed to show a business necessity for the WTS. Dial admits women were able to perform the sausage-making jobs successfully prior to 2000. Jones Tr. Tr. at 63. As discussed in more detail, below, evidence suggesting use of the WTS alone significantly reduced injuries is equivocal at best. The Court therefore concludes Dial has failed to prove that its use of the WTS was content-valid, and consistent with business necessity. See, e.g., Lanning v. Southeastern Pennsylvania Transportation Authority (" SEPTA I"), 181 F.3d 478, 489 (3rd Cir. 1999) ("Judicial application of a standard focusing solely on whether the qualities measured by an entry level exam bear some relationship to the job in question would impermissibly write out the business necessity prong of the Act's chosen standard.").
Dial's expert, Dr. Jackson, also cited a database known as the "Snook data," whichis used by the National Institute of Occupational Safety and Health ("NIOSH") and other organizations to estimate the weight an individual with certain characteristics can lift safely. Based on the Snook data, Dr. Jackson concluded that over 90% of males could be expected to safely perform the lifting required in the sausagemaking positions, but only 10-25% of women could perform the same lifting without injury. Defendant's App. E, Exh. U(8)). He thenopined that because approximately 40% of the women who took the WTS passed, the test may in fact be easier than would have been expected based on the Snook data. Jackson Tr. Tr. at 84. The Court is unpersuaded by this testimony. As proposed by counsel for Dial during trial, ifit were true that only 10% of females could safely perform the Smokehouse positions, the injuryrate for women hired prior to the implementationof the WTS should have been much higher than the actual figures reflect. See Jackson Tr. Tr. at 89-90.
B. Whether WTS is Criterion-Valid
Dial also attempted to meet its burden by showing the WTS is criterion-valid. "Evidence of the validity of a test or other selection procedure by a criterion-related validity study should consist of empirical data demonstrating that the selection procedure is predictive of or significantly correlated with important elements of job performance." Uniform Guidelines, 29 C.F.R. § 1607.5(B). In the present case, injury reduction is the criterion by which the WTS has been evaluated.
Dr. Jones testified at trial that the overall injury rate dropped significantly between January 2000 and September 2003, and that this decline is due almost exclusively to Dial's implementation of the WTS as a pre-employment screening tool. Specifically, Dr. Jones testified that: "If you take into account when people were hired and how much time they actually worked in the sausage team, total injury rates fell by about 75% after implementation of the WTS. Workers' comp injury rates fell about 80% after implementation of the WTS." Jones Tr. Tr. at 39. Referring to defendant's exhibit U-18, Dr. Jones explained:
So the real rate of change here is this: The second column from the right says over that three-year period the rate of injury was about 1.19 injuries per full-time employee year. In other words, if you were a full-time employee working in the sausage team before this test you could expect to get injured a little over once per year. After the test was put in place that number changes to .22. You could expect to get injured about once every five years. That comparison of 1.19 to .22 is what I referred to as about a 75 to 80 percent reduction in injury rates when you look at the data the right way.
Id. at 41. Furthermore, Dr. Jones also opined that the WTS likely had an impact not only on reducing strength-related injuries, but also on non-strength-related injuries. As clarified by Dr. Jones: "When a person works in a job where they became fatigued and became very tired, the chance of injury, of strength or nonstrength-related origin, increases. So there's no reason to believe that the WTS would address only one of these kinds of injuries." Jones Tr. Tr. at 35.
The EEOC disputes the percentage reductions cited by Dial, arguing that individuals hired in 1997 were far more likely to be injured in their initial years of employment than persons hired in 1998-99, before the WTS, or between 2000-2003. Accordingly, "any grouping of injuries by hire year which includes 1997 will show a dramatic reduction in injuries when compared with any other group that does not include them." EEOC Brief in Support of Proposed Findings of Fact and Entry of Judgment at 14.
The Court also finds convincing Dr. Campion's breakdown of the injuries suffered by individuals according to the year in which the individual was hired, as opposed to focusing solely on the year in which the injury occurred. Dr. Campion's analysis shows that males hired in 1997 had an injury rate of 0.45 per exposure rated, while women had an injury rate of 0.39. For 1998, the rates were 0.11 for men and 0.5 for women. For 1999, the male rate of injury per exposure rated was 0.19, with a female rate of 0.16. EEOC Exh. 36. In short, during two of the three years immediately preceding the WTS' implementation, female Smokehouse workers had a lower injury rate than their male counterparts.
Regardless of the persuasiveness of each parties' respective experts, two facts prevent the Court from concluding Dial met its burden as to criterion-validity. First, there is no way to adjust the data to control for other safety measures implemented by Dial during, and immediately preceding, the relevant time period. Evidence was presented at trial that, beginning in 1996, Dial made other attempts to reduce workplace injuries. In 1996, Dial implemented an ergonomic job rotation program for certain smokehouse jobs. In 2000, Dial added a "team approach" to increase job rotation in April, 2000. It also lowered the height of the bar on the Frank-O-Matic machines, and performed periodic ergonomic and safety audits, including instituting an annual "back injury prevention program" in 2000. EEOC Exh. 14. As evidence of Dial's inability to control for other variables, Dial expert Dr. Jones conceded that when injury rates were broken down by year of hire, strength-related workers compensation injuries saw an identical percentage reduction (80%) between 1997 and 1998 as Dr. Jones claimed was achieved between 2000 and September 2003. Jones Tr. Tr. 39, 73. In short, no matter how the numbers are "massaged," Dial could not show that the reduction in Smokehouse injuries following implementation of the WTS was due solely, or even primarily, to the test.
Second, although reduction in any number of injuries is always a desirable goal, the Court agrees with the EEOC that the relatively small number of non-trivial injuries occurring before implementation of the test can not justify a test with the enormous impact of WTS. As shown by Dr. Campion, only one strength-related workers compensation cost injury was suffered before September 2003 by an individual hired in 1998, resulting in an injury rate of 0.09 per full-time employee equivalent. EEOC Exh. 38. Individuals hired in 1999 suffered 3 such injuries, for an injury rate of 0.14. Id. Interestingly, individuals hired in 2001 suffered 3 strength-related workers compensation cost injuries, for an injury rate of 0.16, which is higher than the rates for 1998 or 1999. Id.
CONCLUSION
In summary, even assuming Dial had been able to prove the WTS was content-and/or criterion-valid, the Court finds Dial has failed to fulfill its burden to show it had a "compelling need" for implementation of the WTS, and that other, non-discriminatory mechanisms — namely, many of the same safety programs actually implemented by Dial — could not produce the same results. Bradley v. Pizzaco of Nebraska, Inc., 7 F.3d 795, 797 (8th Cir. 1993) ("Under the Griggs standard, the burden is on the defendant employer to prove both a `compelling need' for the challenged policy, and the lack of an effective alternative policy that would not produce a similar disparate impact."). Having completed the liability portion of the matter, the Court finds that a status conference is in order to identify the issues surrounding the potential remedies available, and to develop a briefing schedule for any additional matters. A telephone conference addressing these issues shall be held at 9:00 a.m. central time on Monday, February 28, 2005. The call shall be placed by counsel for the EEOC.
IT IS ORDERED.