Friday, April 01, 2005

GAO analyzes disparities at DOE laboratories

GAO analyzes disparities at DOE laboratories

ROGER SNODGRASS,, Monitor Assistant Editor

Following up on a 2002 study of equal employment issues at the three nuclear weapons laboratories, including Los Alamos National Laboratory, the Government Accountability Office examined six more Department of Energy laboratories on parity issues for women and minorities.

The four-year survey found mixed results but did detect statistically significant differences in salaries, merit pay increases and separation patterns for certain groups of women and minorities.

At the same time, "The GAO found that women and minorities received comparable merit pay increases at half the labs and comparable or greater increases at two of the remaining three labs," according to the report published last week.

Responding to the study, a DOE official disputed the analytical method used by GAO.

"This leads us to question the accuracy of the GAO analysis, and any conclusions drawn, therefrom," wrote Milton D. Johnson of DOE's Office of Science.

DOE has spent $57 million on equal employment opportunity litigation since 1998 and the report said the amount could increase "substantially" from pending lawsuits.

The six science laboratories from which employment statistics were gathered were Argonne National Laboratory in Argonne, Ill.; Brookhaven National Laboratory in Upton, N.Y.; Idaho National Engineering and Environmental Laboratory in Idaho Falls, Idaho; Lawrence Berkeley National Laboratory in Berkeley, Calif.; Oak Ridge National laboratory in Oak Ridge, Tennessee; and Pacific Northwest National Laboratory in Richland, Wash.

The report was requested by Rep. Judy Biggert, R-Ill., and released late last week. Biggert, the chair of the Science Subcomittee on Energy, with jurisdication over the DOE laboratory and research programs, said in a statement that she was particularly concerned about GAO findings that women were paid 2 to 4 percent less than their male counterparts at five of the six labs.

The study also looked at separation patterns, which were generally comparable, with two notable exceptions.

Women were said to be 40 percent more likely to quit at the Idaho facility, while 70 percent more minorities left Pacific Northwest, according to the study.

Despite specific findings at specific laboratories, the report stopped short of finding discrimination, but recommended renewed efforts by DOE and the Department of Labor to resolve a turf war that has limited federal enforcement of equal opportunity compliance.

Since the first report raising pay equity issues was released in 2002, the two agencies have been unable to agree on how to develop a more formal relationship without interfering with each other's legal responsibilities.

In its response, DOL agreed with both DOE and GAO that the statistical analysis was inadequate to prove unlawful employment discrimination, but affirmed its commitment to continue working on resolving the disagreements with the energy department.

A LANL spokesperson said today that, apart from the original GAO study, the laboratory had commissioned a report on pay equity by Finnis R. Welch, which became the basis for approximately 800 salary adjustments, beginning early last year.

The adjustments ranged from $140 to $10,250, and affected nearly all classes of employees, including women, minorities and white men.

"The end result is that the lab now continues to review salaries for all new and existing workers to insure that no group of employees is paid disparately on the basis of gender or ethnicity," said James Rickman of the Public Affairs Office.

Two separate pay equity lawsuits, one involving females and the other both females and Hispanics lawsuits that were brought against the laboratory, were consolidated about a year ago, said attorney Patrick Allen of Albuquerque.

Toward the end of this year, he said, a judge is expected to rule on certifying the cases as a class action that would include all female and Hispanics at the laboratory.

"We've contended from the beginning and continue to believe that the Welch report seriously underreported pay disparities and even with the adjustments they made, those didn't begin to address the disparities of paying female and Hispanics in comparison to white employees at the laboratory," said Allen.

This is always a complex and inflamatory issue. There is evidence to back up lower ORC scores and salaries for women and Hispanics.

If you take a class of employees such as TSMs or TECs and run the attendance data, you will find that the women and Hispanics in both classes will, on average, be absent (sick leave, family leave, etc) more ofte than the white males. This will result in lower productivity hence lower ORC scores followed by lower salaries. These are no doubt legitimate absences, but LANL operates a performance-based compensation system so lower pay for lower productivity is in order.

Next, if you look at the degreed professionals (TSMs), you will find that the women and Hispanics are more likely to have received their professional education at third tier universities such as UNM and Highlands. Indeed, a few of the SSMs got their MBAs from the University of Phoenix.

As far as the Welch Survey, it resulted in an algorithmic process of calculating what individuals' salaries shoulc be based on educational level, date of first degree, and ORC score. But, it was flawed. I have two employees, both Hispanic, who were within two years of each other in age (late 50's) and work experience, had the same education, and had received identical ORC scores for three years. Prior to the Welch Survey correction they were paid within a few $100 of each other. The Welch Survey correction awarded one of them an $1800 raise thereby creating a salary inequity!
Actually, NM Highlands "University" is a Tier 4 school for those schools whose highest degree is a Masters. Sounds "substantially equivalent" to a UC University to me.
I suspect that the Welch algorithm for calculating raises is reasonable, but that management interfered with the process of giving out the calculated raises. What else could explain the disparity the previous author described between the two Hispanic females?
Among cases I am familiar with, Welsh raises were not given to anyone who had an open grievance against LANL, no matter what the topic of the grievance was. Retaliation? It looks like it. I doubt that Mr. Welch incorporated a factor for open grievances into his formula. That was no doubt done by management.
As far as TSM's are concerned, I have seen no numbers showing that women and Hispanics are more likely to have received their degrees from third tier universities. Nor am I aware of any LANL policy allowing the tier of the University to be included in the awarding of performance based raises. Nor is there any precise way of evaluating the tier of a University. Some people think New Mexico Tech is a high caliber school, others do not. Evaluating the caliber of the person's university is not part of a performance based system.
Evaluating a person on performance is extremely subject to bias unless metrics are used. I would submit that days absent for health or vacation reasons is a very poor measure of performance even though it is metric based since it does not cover quality of work.
Under LANL policy, it is possible to lower people's scores based on attendence, but it is poor policy unless everyone is equally penalized for attendence. It is so easy to notice one person's absences and not another's.
The use of attendence in appraising performance would penalize women and minorities more than white males because women tend to be out with children more than men and Native Americans and Hispanics have cultural requirements to be present for cultural events, especially ceremonies related to death or funerals. It is illegal to discriminate against people for religious preference. Yet I have heard managers complain that some people went to "too many funerals". Too bad for those with large extended families and strong faith.
I think it speaks volumes that 40% more women left the Idaho facility and 70 percent more minorities leave the Pacific Northwest facility. Clearly something is going on in those places to drive them away.
And none of this addresses hiring or failure to promote which appear to be rampant and discriminatory. Women find it very difficult to be hired at LANL, both initially and in job transfers. Again, this looks like bias, but management claims they can't find enough women to hire. That does not explain, however, the many women I know who are unable to get jobs or change jobs at LANL.
I would generally agree with 10:53. Based on an informal, non-rigorous survey, there appears to be no obvious difference in salary solely based on where one attended school.

It would be informative to find out what schools the large number of TSMs with only a Bachelors degree attended (they are about 22% of the TSM population).
1. Should we evaluate people for the length of their career based only on their college or grades?
No, Grasshopper. It is prudent to judge each person on their actual accomplishments over the grading period.

2. Should we judge an entire gender or ethnic group on our impressions?
No, Grasshopper. One tends to notice extreme behaviors more readily- the vast (unnoticed) majority of any group falls well within the normal range. Also, it may not seem logical, but one identifies with one's own ethnic gender group and does not criticize... The age old human conflict of US vs.THEM.
Reading between the lines of the draft RFP, it seems that the DOE/NNSA recognize that there is a substantial performance difference between the S&T side and the administrative management side. That's why they put so much emphasis on preserving the university-style science and reforming the administration with "best commercial practices."

From a political (and probably a legal) standpoint, however, you couldn't leave the S&T TSMs in the UC system and transfer all the admin UC employees to a third party industrial contractor. AA/EEOC regulations alone (disparate impact on women/minorities) would prevent it. That's why they need a new organizational structure that is independent of the UC organization.
Guess what else is coming?
A structured series for the TSMs!

DOE has been pushing this for about 20 years. HR likes it because it will let them hire more people to administer it. BNL, LLNL, ANL and other DOE labs have it but SNL does not. Past LANL Directors have successfully fought this. SEAC has addressed this issue about every five years and has come out against it. BUT, Admiral Butthead likes it because that is was the Navy has.

Having dealt with a structured series for TSMs at my previous employer (also the GOCO), I can state that it is major source of dissatisfaction and strife in the workplace. I can see absolutely no positive aspect to it. The first thing that will happen is that we will have have "diversity" in the distribution. Lawsuits will follow.

And, as a first step, the LANL Jobs website now shows TSM openings as "TSM 00."
There is no more need to administratively classify the TSM series than there is to structure the TEC series. Just as there are technicians who have little technical ability but are a step above laborers, there are TSMs who have the title because of a college degree.

LANL has a reputation of arrogance. After almost 30 years the only arrogance I see is that of certain TSMs and TECs who think they are important because of their title. Position in a pecking order without demonstrated ability only leads to arrogance.

Management should rate people on how well they do their job not on their title
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?