Beginning in 1985, the Rockwell International Corporation required all applicants for positions at its plant in Centralia, Illinois, to undergo "nerve conduction tests." The tests were designed, we are told, to confirm the presence of neuropathy--a syndrome characterized by, among other things, sensory loss and muscle weakness. Rockwell hoped the tests would identify job applicants susceptible to cumulative trauma disorders such as carpal tunnel syndrome. Accordingly, Rockwell sent each applicant to a medical facility where the median nerves in his or her arms were stimulated with electric shocks and the travel time of the electrical impulse from the shock points to the muscles was recorded. The results of the tests were given to Rockwell, which used them in making personnel decisions.

Rockwell followed this policy because its four entry level positions (trimmer, finisher, final finisher, and assembler), into which 90 percent of all new hires were placed, involved "continuing repetitive motions and/or [the use of] vibratory power tools" that put workers at risk for developing cumulative trauma disorders. Rockwell's other three nonskilled positions (molder, multi- operator, and RTM operator) did not require repetitive motion or the use of vibratory power tools, but these were more desirable bid, not entry level, positions. Thus, any new hire placed as a molder, multi-operator, or RTM operator would be "bumped" from that position by a more senior employee, as permitted by the union's collective bargaining agreement, and would end up working as a trimmer, finisher, final finisher, or assembler. Rockwell therefore rejected all nonskilled job applicants who scored outside the normal range on the nerve conduction test.1

The Equal Employment Opportunity Commission brought suit on behalf of 72 job applicants rejected by Rockwell on the basis of abnormal nerve conduction test results, alleging that the company's policy violated the Americans with Disabilities Act, 42 U.S.C. sec. 12101, et seq.2 Rockwell has stipulated that the 72 job applicants, except for flunking the test, were qualified for its entry level positions. We hasten to note that the applicants did not suffer from any impairment at the time they were turned away by Rockwell, but Rockwell merely regarded them as having an enhanced likelihood of developing impairments in the future. The bone of contention thus became whether Rockwell discriminated against the job applicants because it perceived them as suffering from a disability. See 42 U.S.C. sec. 12102(2)(C).

After 3 years of litigation, Rockwell informed the district court that it intended to move for summary judgment on the basis that the EEOC had developed no evidence that Rockwell regarded the claimants as disabled within the meaning of the ADA. According to Rockwell, the parties advised the district court at a status conference that they would need time to develop expert vocational evidence, but that expert ergonomics and medical evidence would not be relevant to resolve Rockwell's motion. The Commission disputes that it made such a representation to the court, and our record does not reflect that the Commission took any position on the matter. In any event, on August 25, 1998, the Commission filed the report of its vocational expert Dr. Michael Brethauer, and on December 18, 1998, Rockwell moved for summary judgment, asking the district court to exclude Brethauer's report under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). Briefing on Rockwell's motion for summary judgment was completed by January 29, 1999, and the motion was continued pending an April 27, 1999, Daubert hearing.

Ten days prior to the Daubert hearing the Commission sought leave of court to submit amendments to Dr. Brethauer's report, which for the first time relied upon the findings of Rockwell's ergonomics expert, Dr. Michael Smith. The district court permitted the Commission to file the amended report and rescheduled the Daubert hearing for late May in order to allow time for further discovery necessitated by the amendments. Then, a week prior to the rescheduled Daubert hearing, the Commission moved to supplement the summary judgment record with Dr. Smith's report and deposition testimony. Dr. Smith's report, which was served on the Commission on March 10, 1999, offered opinions only on the ergonomic risks posed by the four entry- level Rockwell jobs and the potential for each to cause work-related musculoskeletal disorders. Although Dr. Smith compared the repetition and duration of upper extremity movements required by the four Rockwell jobs to the repetition and duration of upper extremity movements required by a selection of jobs in other fields, he did not render an opinion on the number of jobs in the Southern Illinois job market from which the claimants would be foreclosed due to the impairments perceived by Rockwell.

The district court denied the EEOC's motion to supplement the record as untimely. Noting its previous indulgence of the Commission in granting the request to amend Dr. Brethauer's report after the completion of briefing and its previous extension of the Daubert hearing, the district court held that the Commission should have moved to supplement the record earlier, rather than sitting on Smith's report for almost 3 months. Subsequently, after hearing testimony at the Daubert proceeding, the district court excluded Dr. Brethauer's report.

Without any admissible evidence from a vocational expert, the Commission's case was doomed. On August 13, 1999, the district court granted summary judgment to the defendants, holding that the Commission could not prove Rockwell regarded the claimants as disabled because there was no evidence that Rockwell considered them foreclosed from an entire "class of jobs" or a "broad range of jobs in various classes" in the relevant geographic area. The Commission now appeals this ruling and the district court's denial of its motion to supplement the summary judgment record.3 We'll first review the scheduling decision for abuse of discretion, Arthur Pierson & Co. v. Provimi Veal Corp., 887 F.2d 837, 838-39 (7th Cir. 1989), and then the summary judgment ruling de novo. Skorup v. Modern Door Corp., 153 F.3d 512, 514 (7th Cir. 1998).

The district court denied the Commission's request to supplement the summary judgment record in an oral ruling, so its reasoning is not set out as fully as it might have been in a written order. The district judge did express concern, however, that Rockwell's motion for summary judgment had been fully briefed and awaiting decision for some time. This was a legitimate concern. The district court had already allowed the Commission to amend Dr. Brethauer's report after the completion of summary judgment briefing, which necessitated the suspension of proceedings so Rockwell could re-depose him. This delayed, of course, action on the summary judgment motion. Then, on May 20, 1999, only a week before the Daubert hearing, the Commission moved to supplement the record with Dr. Smith's report, which it received more than 2 months earlier on March 10, 1999. Permitting the Commission to supplement the record with that report would likely have provoked a second Daubert challenge from Rockwell on the ground that Smith, an ergonomist, was not qualified as a vocational expert. See, e.g., Wilson v. City of Chicago, 6 F.3d 1233, 1239 (7th Cir. 1993) (pathologist not qualified to testify on subject within the expertise of a neurologist, psychiatrist, or physiologist). This would have caused a substantial further delay in the disposition of Rockwell's summary judgment motion, which was already growing old. The district court quite reasonably decided that it didn't want to wait any longer. See Trippe Mfg. Co. v. American Power Conversion Corp., 46 F.3d 624, 629 (7th Cir. 1995) ("Federal district courts have the inherent power to administer their dockets so as to conserve scarce judicial resources.").

Moreover, Dr. Smith's report was irrelevant to the determination of Rockwell's motion for summary judgment because it expressed no opinion on vocational issues, the only point of contention in the motion. The Commission contends that Dr. Smith's ergonomic calculations are relevant to the number of alternative jobs the claimants could have performed subject to Rockwell's restrictions, and thus are relevant to whether Rockwell considered the claimants disabled. But in order to prove Rockwell considered the claimants disabled under the ADA, the Commission must have evidence that Rockwell viewed the claimants' condition as a restriction on their ability to perform "a class of jobs or a broad range of jobs in various classes" in the relevant geographic area. 29 C.F.R. sec. 1630.2(j)(3)(i)-(ii). Although Dr. Smith's report states the number of upper extremity movements per hour required of a warehouse worker at J.C. Penney and a few other selected jobs, it addresses neither how many such jobs exist in Southern Illinois nor what types of movements are required for the myriad number of alternative jobs. On this record, therefore, we cannot say the district court abused its discretion in thwarting the Commission's effort to hijack Dr. Smith's ergonomics report for use as vocational evidence.

So the EEOC is left without admissible vocational evidence. It presses on, however, contending that the district court erred in granting summary judgment to Rockwell because it could prove, based solely upon Rockwell's perception that the claimants could not perform jobs requiring frequent repetition or the use of vibratory power tools, that Rockwell considered the claimants disabled. This logical leap the Commission asks us to take--from Rockwell's perception that the claimants could not perform four specific jobs to the conclusion that Rockwell regarded them as foreclosed from many jobs in Southern Illinois--is a large one, and requires us to be extra mindful of the essential elements of the Commission's case.

In order to establish an ADA violation, the Commission must prove Rockwell discriminated against the claimants because it considered them to be disabled, or suffering from an impairment which "substantially limits" a major life activity. Sutton v. United Airlines, Inc., 527 U.S. 471, 481 (1999). When the major life activity at issue is working, as it is in this case, "substantially limits" means the claimants were "significantly restricted in the ability to perform either a class of jobs or a broad range of jobs in various classes as compared to the average person having comparable training, skills and abilities." 29 C.F.R. sec. 1630.2(j)(3)(i). A "class of jobs" is the job from which a claimant was disqualified, as well as all other jobs utilizing similar training, knowledge, and skills within "the geographical area to which the [claimant] has reasonable access." 29 C.F.R. sec. 1630.2(j)(3) (ii)(A)-(B). A "broad range of jobs in various classes," in contrast, is the job from which a claimant was disqualified, as well as all other jobs not utilizing similar training, knowledge, and skills within "the geographical area to which the [claimant] has reasonable access." 29 C.F.R. sec. 1630.2(j)(3)(ii)(A), (C). Thus, according to the Commission's own regulations, its case must include some proof of the "number and types of jobs" within the "geographical area to which the [claimant] has reasonable access." See Sutton, 527 U.S. at 492-93; Murphy v. United Parcel Serv., Inc., 527 U.S. 516, 524-25 (1999).

Rockwell argues that this proof must take the form of quantitative vocational data addressing the number of alternative jobs in the Southern Illinois area (either within the same class or across classes) from which Rockwell regarded the claimants as disqualified. In support of this contention, Rockwell cites a number of cases from other circuits that discuss the absence of such evidence in ruling against ADA plaintiffs. See, e.g., Santiago Clemente v. Executive Airlines, Inc., 213 F.3d 25, 32 (1st Cir. 2000); Broussard v. University of California, at Berkeley, 192 F.3d 1252, 1257-58 (9th Cir. 1999); Muller v. Costello, 187 F.3d 298, 313 (2d Cir. 1999); Doren v. Battle Creek Health Sys., 187 F.3d 595, 598 (6th Cir. 1999); Bolton v. Scrivner, Inc., 36 F.3d 939, 944 (10th Cir. 1994). But Rockwell points to no authority for the proposition that such quantitative data is indispensible to a claimant's case. In fact, the EEOC's interpretive guidelines make clear that the factors set out in sec. 1630.2(j)(3)(ii) "are not intended to require an onerous evidentiary showing. Rather, the terms only require the presentation of evidence of general employment demographics and/or of recognized occupational classifications that indicate the approximate number of jobs (e.g., 'few,' 'many,' 'most') from which an individual would be excluded because of an impairment." 29 C.F.R. Pt. 1630, App. sec. 1630.2(j). We therefore decline Rockwell's invitation to create a per se rule that a plaintiff cannot prevail without quantitative evidence of the precise characteristics of the local job market.

At the same time, this is not one of the rare cases in which the claimants' impairments are so severe that their substantial foreclosure from the job market is obvious. Compare DePaoli v. Abbott Lab., 140 F.3d 668, 673 (7th Cir. 1998) (medical evidence of plaintiff's inability to make any repetitive motions with her right hand sufficient to create triable issue on foreclosure from entire class of assembly line jobs) with McKay v. Toyota Motor Mfg., U.S.A., Inc., 110 F.3d 369, 373 (6th Cir. 1997) (evidence of plaintiff's inability to use vibrating power tools and perform repetitive motions with right hand insufficient to establish foreclosure from entire class of manufacturing jobs). Accordingly, the Commission had to come up with some evidence of the number and types of other jobs in Southern Illinois from which the job applicants would be excluded because of their perceived impairments. "This is not an onerous requirement, but it does require at least some evidence from which one might infer that [the claimants] faced 'significant restrictions' in [their] ability to meet the requirements of other jobs." Davidson v. Midelfort Clinic, Ltd., 133 F.3d 499, (7th Cir. 1998). Mere proof that an impairment prevented an individual from performing a particular job for a particular employer is insufficient to render him or her "disabled" under the ADA. Sutton, 527 U.S. at 491; Murphy, 527 U.S. at 525; Baulos v. Roadway Express, Inc., 139 F.3d 1147, 1151 (7th Cir. 1998).

There is no evidence in this record concerning the demographics of the Southern Illinois employment market. Rather, the Commission asks us to infer that Rockwell regarded the claimants as significantly restricted from entering that market based solely on the fact that Rockwell perceived them as unable to perform four specific jobs at Rockwell. This is precisely the type of inference we refused to make in Skorup, where we affirmed summary judgment because an ADA plaintiff failed to set out general guideposts from which we could determine if her impairment foreclosed her from few, many, or most jobs in a particular class or in a broad range of classes. 153 F.3d at 515. Similarly, the record before us in this case does not disclose whether the claimants' perceived inability to perform jobs requiring frequent repetition or use of vibratory power tools foreclosed them from any job in Southern Illinois other than the four Rockwell jobs. Although the Commission is not required to calculate an exact percentage of jobs from which Rockwell perceived the claimants as foreclosed, it cannot survive summary judgment in a case like this with no evidence of the demographics of the relevant labor market. The judgment of the district court is AFFIRMED.

DIANE P. WOOD, Circuit Judge, dissenting.

I regret that I cannot join the majority's opinion, despite the fact that there are significant points of agreement between us. The majority and I agree that, as the opinion puts it, there is no per se rule for ADA plaintiffs to the effect that they cannot prevail without quantitative evidence of the precise characteristics of the local job market. No one would quarrel with the idea that such quantitative, statistical evidence may often be useful, but the question in the case before us is whether these plaintiffs can survive summary judgment without it. No, the majority replies, and it goes on to speculate that statistical evidence is almost always necessary. I take issue with both propositions: the conclusion in this case, and the broader statement. Specifically, on this record I cannot agree that the Equal Employment Opportunity Commission had "no" evidence tending to show what class or range of jobs was at issue here--that is, the class of jobs, or broad range of jobs, from which persons with the assumed disability Rockwell had identified would be excluded. To the contrary, the Commission did have evidence, of the most concrete kind imaginable, and its evidence easily met the legal standards with respect to this element. I would find that the EEOC and the claimants are entitled to a trial on their ADA claims to resolve the remaining, and very troubling, issues in this case.

The debate we are having fits within the structure of an ADA case as follows. A person must first either have a disability, or have a history of a disability, or be regarded as having a disability, in order to be entitled to protection under the Act. See Sutton v. United Air Lines, Inc., 527 U.S. 471, 478 (1999); 42 U.S.C. sec. 12102(2). In the present case, everyone agrees that the claimants, despite the score they received on the nerve conduction tests, had no present disability. Instead, Rockwell believed that people whose nerve conduction tests produced scores above a certain level were more likely to develop cumulative trauma disorders such as carpal tunnel syndrome. (Although it is unimportant to the case in its present posture, I note that this assumption was at best highly dubious, and certainly a contested point of fact. And it is not at all clear to me that as a matter of law the ADA permits an employer to refuse to hire a person who is fully qualified to perform certain work, simply because that individual might at some unspecified time in the future develop a physical or other disability that would render her unable at that later date to meet the employer's reasonable expectations. This smacks of exactly the kind of speculation and stereotyping that the statute was designed to combat.) Rockwell then took matters one step further and decided to treat the claimants as if they already had the feared disorders; as such, it decided to exclude them from all jobs that would be unavailable to such people.

This leads us to the second step. Not every impairment is a protectible disability under the statute; the statute covers only those impairments that substantially limit a major life activity. Sutton, 527 U.S. at 481; 42 U.S.C. sec. 12102(2); 29 C.F.R. sec. 1630.2(g). When, as in this case, we are dealing with a perceived disability rather than an actual disability, the test becomes whether the employer mistakenly believes either (1) that an unimpaired person has an impairment that substantially limits a major life activity, or (2) that an actual but non- limiting impairment substantially limits a major life activity. Sutton, 527 U.S. at 489. The major life activity at issue here, as is often the case, is the activity of working. In order to show that the disability these claimants supposedly had was one that fell within the scope of the statute, the EEOC had to show that it left them "significantly restricted in the ability to perform either a class of jobs or a broad range of jobs in various classes as compared to the average person having comparable training, skills and abilities." 29 C.F.R. sec. 1630.2(j)(3)(i). The standards applicable to an actual disability apply equally in a perceived-disability case: the question is whether, if the claimants had actually had the disabilities the employer regarded them as having, the claimants would have been significantly restricted in their ability to perform either a class of jobs or a broad range of jobs in various classes. See Moore v. J.B. Hunt Transport, Inc., 221 F.3d 944, 954 (7th Cir. 2000). Rockwell, through its motion for summary judgment, imposed on the Commission the obligation to produce evidence in opposition to the motion that would have permitted a trier of fact to conclude that either a class or a broad range of jobs was so affected. In my view, the Commission met that burden.

It did so in a pragmatic way, rather than through the use of statistics. It pointed out that the record contains concrete information about the characteristics of the jobs Rockwell thought these people could not perform: that is, the job requirements of the four specific positions from which Rockwell excluded them. I certainly agree with the majority that pointing to the requirements of a specific job will not always be enough to show that a broad range of jobs or a class of jobs is at issue. If the job in question calls for highly idiosyncratic skills--the ability to work one specific kind of machine, the ability to operate one kind of computer program, or the ability to work a certain time shift, for example--the fact that a person does not have those skills would tell a court little or nothing about whether the set of qualifications for the job in question was typical of a class or broad range of jobs. On the other hand, to take an extreme example at the other end of the spectrum, if the only qualification for the job was the ability to read and write, then it is obvious that a person who was unsuited for that job would be equally unsuited for a class or broad range of jobs. In such a case, it would be absurd to require statistical evidence showing how many jobs required literacy. With this much, I believe that the majority and I are still in agreement. It is "obvious," to use the majority's word, that a broad range of jobs requires literacy, and thus it would be pointless to insist that a vocational expert prepare a report showing that 95% of the jobs in a given area would be foreclosed to a person lacking such a qualification.

The problem is that the discussion so far simply defines the end-points of a line ranging from the greatest degree of particularity to the greatest generality of job characteristics. At some point between those extremes, additional evidence beyond the criteria of the job(s) from which the applicant is excluded will become necessary to show that the exclusion affects an entire class of jobs or a broad range of jobs. The problem posed by this case is where to draw that line, and according to what criteria.

In my view, the starting point for analysis is appropriately the job or jobs from which the employer is excluding the allegedly disabled individual. A court should look at the job skills that the employer demands for that job or jobs and determine how specific to the employer's own workplace they are. In general, the more particular the job requirements, the more necessary it will be for a plaintiff to present demographic evidence that shows that those requirements are in fact found in enough jobs to matter for ADA purposes. But if the job requirements include only a list like a high school diploma, the ability to lift more than 30 pounds, the ability to use power tools, the ability to perform frequent repetitive motions (without specifying the particular motion needed), and regular attendance, they are general enough to describe a broad range of jobs, or a certain class of unskilled work. A trier of fact can tell the difference between generalized criteria that will cut across a broad range of jobs, and specialized criteria that are employer- and workplace-specific, whether or not a vocational expert tells it how many thousands of jobs in the relevant geographic area have similar requirements. It is also appropriate, in this context, to see how many jobs the particular employer thought were affected by the perceived disability: just one or two, or a broad range? The Commission introduced evidence showing that 90% of the entry-level unskilled jobs at Rockwell were off limits to the applicants here. That evidence also gave an objective view of how widely Rockwell's exclusionary practice was sweeping.

This point can be further illustrated by comparing the impairments involved in cases in which this and other circuits have allowed a claimant to go forward without requiring demographic evidence with the impairments involved in cases in which courts have suggested that demographic evidence would be necessary. This court has held that an impairment that prohibited the claimant from making repetitive motions with her right hand raised a triable issue of fact as to whether she was disabled, see DePaoli v. Abbott Laboratories, 140 F.3d 668, 673 (7th Cir. 1998), as did an impairment that prevented performing overhead work, heavy lifting, and pushing and pulling out from the body, see Cochrum v. Old Ben Coal Co., 102 F.3d 908, 911 (7th Cir. 1996). Accord Best v. Shell Oil Co., 107 F.3d 544, 548 (7th Cir. 1997) (truck driver whose employer perceived his bad knee as prohibiting him from driving trucks with a clutch or with a certain common seat configuration was regarded as disabled under the ADA); Wellington v. Lyon County School Dist., 187 F.3d 1150, 1155 (9th Cir. 1999) (plaintiff allowed to go forward based on evidence of carpal tunnel syndrome which prevented him from performing work involving metal fabrication, welding, heavy activities, carpentry, or the use of a variety of tools to do maintenance and repairs); see also 29 C.F.R. Part 1630 Appendix sec. 1630.2(j) ("an individual who has a back condition that prevents the individual from performing any heavy labor job would be substantially limited in the major life activity of working"). In contrast, the cases cited by the majority in which courts have suggested that demographic evidence is necessary have involved much more specialized restrictions. See Santiago Clemente v. Executive Airlines, Inc., 213 F.3d 25, 32 (1st Cir. 2000) (flight attendant could not fly on non-pressurized planes; could do a variety of ground jobs for employer or fly on pressurized planes); Broussard v. University of California at Berkeley, 292 F.3d 1252, 1259 (9th Cir. 1999) (animal lab technician with carpal tunnel syndrome could not perform duties of caring for mice in one lab; one of accommodations employee sought was transfer to working with different animals); Muller v. Costello, 187 F.3d 298, 313 (2d Cir. 1999) (corrections officer with asthma precluded from working in smoky environment; could still work in smoke-free county jail); Doren v. Battle Creek Health Sys., 187 F.3d 595, 598 (6th Cir. 1999) (pediatric nurse prohibited from heavy lifting or working more than 8-hour shift; no evidence that she was excluded from any pediatric job by lifting requirement, or of how many pediatric nursing jobs in area required longer than 8-hour shift); Bolton v. Scrivner, Inc., 36 F.3d 939, 943-44 (10th Cir. 1994) (employee faced "limitations" on standing, walking, and lifting overhead, but medical report gave no indication of extent of restrictions). In my view, the disability that Rockwell perceived the claimants in this case as having--a disability which, in Rockwell's estimation, prohibited these workers from performing frequent repetitive motions or from using vibratory power tools--is much more similar to a restriction on repetitive motion with one hand or a restriction on overhead work than it is to a restriction on flying in non-pressurized planes, or working in a smoky environment, or caring for certain mice.

I see nothing to be gained by having vocational experts routinely appear in ADA cases solely for the purpose of testifying that a broad range of jobs require the ability to lift 30 pounds, or the ability to perform repetitive motions. Indeed, as noted above, in DePaoli we found that a disability that prevented the plaintiff from performing repetitive motions with her right hand was protected under the Act. Obviously, in this case we cannot precisely compare the severity of the disability suffered by the plaintiffs with that of plaintiff DePaoli, for the simple reason that these people were not yet disabled at all. Rockwell merely perceived them as disabled because it thought that they might become so impaired by repetitive stress injuries in the future that they would not then be able to do the job. For analytic purposes now, as noted above, we must approach this case as if these plaintiffs were already suffering from injuries that kept them away from frequent repetitive motions and the use of vibratory power tools. Once that is done, the sufficiency of the Commission's evidence to defeat summary judgment becomes clear.

I would reverse and remand for further proceedings, and thus I respectfully dissent.