3 CIR 410 (1978).

IN THE COURT OF INDUSTRIAL RELATIONS

OF THE STATE OF NEBRASKA

OMAHA ASSOCIATION OF | CASE NO. 240
FIRE FIGHTERS, LOCAL 385, |
|
Plaintiff, |
|
v. | OPINION AND ORDER
|
CITY OF OMAHA, NEBRASKA, |
A Municipal Corporation, |
|
Defendant. |

Filed February 16, 1978.

Appearances:

For the Plaintiff: James F. Costello

For the Defendant: Kent N. Whinnery

Before: Wall, P.J.; DeBacker and McGinley, J.J.

WALL, P.J.:

This matter brings before us the dispute between the Omaha Fire Fighters and the City of Omaha as to the proper wage to be set for the firefighters for the calendar year 1978. We find that we have jurisdiction of the parties and of the subject matter of the dispute.

This case presents evidence of correlation or noncorrelation of factors affecting wages upon which we have speculated in the past. Omaha Police Union v. City of Omaha, 3 CIR 233-1 (1977). The evidence confirms our suspicion that the presumptions which we had previously indulged were, in fact, not based on reality. In Omaha Police Union, supra, we offered the defendant three alternatives in approaching plaintiff's array to show that where data was taken from outside the relative homogeneity of Nebraska, that is should be adjusted before we used it directly in setting wages. The defendant here attempted to do this by showing discrepancies in the percentage of workers in manufacturing and relative earnings per worker. It did not present data on the other two items approved by the Supreme Court in Lincoln Fire Fighters Assn. v. City of Lincoln, 198 Neb. 1974, 252 N. W. 2d 207 (1977), degree of unionization or median family income. Nor did it attempt to show any correlation between its figures and firefighters' wages. There is a relatively simple statistical test to show correlation or non-correlation of different sets of data, the sample correlation coefficient, which yields a number that is a measure of how much linear relationship there is between two sets of data.[1] The value of the sample correlation coefficient is always between + 1 and -1. A value of -1 shows a perfect linear relationship in inverse. A value of +1 shows a perfect linear relationship on the positive side. If there is no linear relationship, the value will be near zero, indicating the relationship is one of almost pure chance. The larger the value, the more pronounced the linear relationship of the two sets of data becomes. The plaintiff's expert testified that at a value of .743 in a universe of 10 data items, the relationship is considered significant by statisticians, showing that there is only a 5% probability that the correlation is due to random factors or chance.[2]

Defendant's economist did not do this test on his data, and was seriously impeached as to his knowledge of his data base. Plaintiff's economist-statistician did do this test for correlation on percentage of workers in manufacturing, coming up with a value of .264, indicating no correlation. We have done the test on defendant's data on earnings per worker, finding a value of .461, again indicating no meaningful correlation. Understandably, defendant's economist did not use this test on his data, though he acknowledged he was aware of it, for it would have shown his figures to be completely meaningless.

Plaintiff's expert also used the test on the data for median family income, securing a value of -.06 or almost pure chance. He used the test on data for unionization and produced a value of .554, a figure also indicating no meaningful correlation. On the other hand, in correlating population, size of residential housing stock, and amount of property protected, he found values of .826, .838 and .835, indicating a high correlation with little possibility of chance affecting the correlation.

It appears that the evidence has safely laid to rest the presumptions as to correlation that we have previously indulged. The evidence we really need is evidence of causality and of similarity of the data universe from which the statistics are taken. It may well be that nonparametric statistics are of little or no use to us.

The plaintiff and defendant have concurred in presenting the same 10 cities used by this Court in the 1975 cases between the same parties, with the addition of the City of Pittsburgh by the defendant. The defendant's Safety Director testified that in his opinion the cities of Dayton, Des Moines and Lincoln are not comparable to the City of Omaha because their populations are below 250,000. As we have pointed out before, we are not wedded to any particular array in determining cases between the same parties, especially where changes, such as the population of Dayton, may have removed a particular city from a comparable status. Murray Educ. Assoc. v. School District , 3 CIR 145-1 (1975), Nehawka Classroom Teachers Assn. v. School District , 3 CIR Adv. 248, 3 CIR 181-1 (1977). Nevertheless, where the data before the Court can be utilized to provide a reasonable basis for an order fixing the prevalent wage rate, the Court will do so. Omaha Assoc. Firefighters v. City of Omaha , 2 CIR 117-1 (1975), aff'd. 194 Neb. 436, 231 N.W.2d 710 (1975).

We also note that the data for Kansas City is effective July, 1978, far beyond the effective date for which we are to set the wage scale in Omaha. However, since we are going to construct a hypothetical universe from which to infer the proper wage, we feel that we may use the Kansas City data, relying on the premise that any upward skew caused by using it will be more than offset by the downward skew caused by three cities below the population range considered normally comparable.

The plaintiff's array results in a mean of the minimum salaries for firefighters of $12,947.00, or an 8.5% increase. The median of the minimum is $13,404.00, or a 12.4% increase. The plaintiff's mean of the maximum produces a 6.4% increase and the median of the maximum is 8.2%. If, as suggested by the defendant, we accept plaintiff's array and add Pittsburgh, the mean of the minimum is $12,876.18, or 7.9% and the median of the minimum is $13,725.00, or 15.1%. Again, adding Pittsburgh to the maximum, the mean is $15,689.00, or 5.9% and the median is $16,020.00, or 8.1%.

If we delete the four cities about which we question comparability, but add Pittsburgh, the mean of the minimum is 14.7% and the median of the minimum is 20.7% The mean of the maximum is 9.7% and the median of the maximum is 8.6%. However, since the parties have chosen other cities, and our choice results in a far higher figure, in order to assume a conservative approach, we use the parties' data and not ours.

Constructing a scatter chart of the figures we have obtained, we find a range at the minimum from 7.9% to 15.1% with an approximate center at 11.5%. In order to allow for any upward skew caused by the fact that Akron and Dayton have effective dates in April, rather than in January, in the informed exercise of our legislative judgment and expertise, we set the increase of the minimum at 9%. A scatter chart of the maximums shows a range from 5.9% to 8.2% with an approximate center at 7.8%. Again, in order to allow for the possible upward skew, in the informed exercise of our legislative judgment and expertise, we set the increase of the maximum at 7%. We spread the differential over the intermediate steps.

We note that plaintiff has tendered data showing the per hour compensation differences in the arrayed cities. In the plaintiff's array, this would result in increases ranging from 16.1% to 23.4%. As noted elsewhere, we prefer to make this type of adjustment by adjusting the work load, rather than adjusting the hourly wage, though this is not a binding limitation. Beatrice Ed. Assn. v. School Dist., 3 ClR 207-1 (1978). Here, the plaintiff has not asked for a reduction in the work week, nor do its figures demonstrate that Omaha is so far out of the parameters of the mainstream as to require adjustment. Until the evidence of actual working hours, as opposed to the work week, is more definite, we make no adjustment, nor do we utilize the nominal hourly wage in setting the salaries comparable to the prevalent.

Turning then to a determination of the proper salaries for Captains, we find the plaintiff's array to produce 19.1% at the mean and 25.1% at the median. Adding Pittsburgh as the defendant suggests, gives a mean of $19,518.73, or 18.2% and a median of $20,614.00, or 24.8%. Because the indicated increase differs so much at this rank, as compared to the other ranks, we determine to adjust for any possible upward skew in the data by setting the increase for Captains at 15.0%.

On District Chiefs, we find the Plaintiff's array to Adding Pittsburgh, we find a mean of $22,090.00, or 19.4%. We determine to set the wage level at the center point between the lower figures, or 12.8%.

We then turn to those ranks on which there is little data. Plaintiff suggests that we use percentages of other established salaries. Defendant has made no suggestion. Plaintiff has presented an array of percentages indicating that Omaha's differentials are too low. We do not deem the data sufficient to show that Omaha's practice is not comparable or is out of the mainstream. Accordingly, we maintain the current differentials as follows: EMT Technician, 102.3% of a firefighter's salary, or an increase of 7.2%; Engineer and Lieutenant, 104.7% of a firefighter's salary, or an increase of 7.3%; EMT Captain, 102.1% of a Captain's salary, or an increase of 15.3%; Fire Prevention Inspector and Drillmaster, 87.3% of a District Chief's salary, or an increase of 11.2%.

Finally, we turn to other matters of dispute.

Dues Checkoff . While not discussed by the parties, we do not find it to be an agreed article in Exhibit 12. We find no variance from the prevalent, and direct its continuance.

Group Insurance . We find no variance from the prevalent and direct the defendant to continue to pay 100% Of family medical coverage. We find the defendant to be outside the prevalent as to a#mount of group life insurance, and direct an increase to $7,000.00 per member, 100% paid by the City.

Uniform Allowance. The defendants array supports a far higher allowance here than does the plaintiff's. We adopt the center point between the two means, or $210.00 per year, Plus continuance of the provision of turn-out gear, as provided in the current contract.

Longevity . We find the defendant to be outside the mainstream in this item, and direct an increase to $150.00 per year from 7-14 years, to $360.00 per year from 15 to 21 years, and to $576.00 per year over 21 years.

Holidays . The City's array would support an increase of I day. The plaintiff's array supports an increase of 1/2 days. We would ordinarily direct an increase of 1 day to 10 days, but the defendant urges us to increase the holidays to 11, which would match the State of Nebraska, and accordingly, we direct an increase to 11.

In setting the above amounts, we have taken into account all factors required by statute and Court decision, including pension and IOD pay, even though not specifically litigated or discussed herein. The step schedule for firefighters is attached hereto as Annex "A".

Ord ered that the dispute be settled as herein provided.

_______________________________