Method for statistical comparison of occupations by skill sets and other relevant attributes

1. A method for computing a correlation coefficient for a pair of occupations as an indication of transferability of workers between the occupations comprising:

providing a computer to perform computations;

selecting the pair of occupations;

for each occupation in a set of occupations:

entering into a matrix or database a set of scores for a selected individual worker attribute for an occupation, the set of scores for each attribute being one or more sets S k consisting of N k scores, where S k x j (j=1, 2, . . . N k );

computing, with the computer, a product of an importance score and a level score for the attribute to produce a composite attribute value;

entering the composite attribute value into the matrix;

accessing, with the computer, data in the matrix;

computing, with the computer, a correlation coefficient for the pair of occupations by correlating the composite attribute value for the pair of occupations by the following formula:

Correl

⁡

(

X

,

Y

)

=

∑

i

N

k

⁢

⁢

{

(

x

i

-

μ

x

)

⁢

(

y

i

-

μ

y

)

}

[

∑

i

N

k

⁢

⁢

(

x

i

-

μ

x

)

2

⁢

∑

i

N

k

⁢

⁢

(

y

i

-

μ

y

)

2

]

wherein: X and Y represent two vectors in a set, S k , of scores for the x th and y th occupations in a set of occupations; where both x and y run from 1 to N, where N is the number of occupations in the set of occupations and x and y are composite attribute values assigned to each occupation; k is a value indicating a level assigned to an individual worker attribute; x i a X; y i a Y; i=1, 2, . . . N k ;

μ

x

=

∑

i

N

k

⁢

⁢

{

x

i

}

N

k

is the mean value of the X vector and

μ

xy

=

∑

i

N

k

⁢

⁢

{

y

i

}

N

k

is the mean value of the Y vector;

normalizing the correlation coefficients on a scale of 1 to 100;

adjusting the normalized correlation coefficients by

producing a ratio whereof the numerator is calculated as the sum of the differences between the members of each pair of the composite level-importance scores for two selected occupations, the two selected occupations being a source occupation and a destination occupation, for which the composite level-importance scores for the destination occupation exceed the composite level-importance scores for the source occupation, and the denominator is calculated as the raw sum of the composite level-importance scores of the destination occupation; and

adjusting the correlation coefficients, with the computer, for each composite attribute by averaging it with the produced ratios; and

producing a report of the correlation coefficients for all pairs of occupations in the set to indicate transferability of workers, indicated by a value of the correlation coefficients, between the occupations.

2. The method of claim 1 , wherein adjusting the normalized correlation coefficient by the formula (w 1 *NORMCORREL d +w 2 * A P C )/(w 1 +w 2 ) wherein:

i. the source occupation is that from which it is proposed that one or more workers should be transferred;

ii. the destination occupation is that to which said workers are to be transferred;

iii. A equals a total number of attributes for a descriptor “d”;

iv. N equals a number of positive differences, the number of attributes of the given descriptor “d” for which a standardized database score of the destination occupation exceeds a value of the same attribute for the source occupation;

v. P + is a percent of differences that are positive, N/A expressed as a percent;

vi. NORMCORREL d equals a normalized simple correlation coefficient between the scores of the attributes in the source and destination occupations, computed for the descriptor;

vii. P C equals the “Proximity Coefficient” which is a ratio of a sum of the scores of the source occupation to a sum of the scores of the destination occupation, but only for those attributes for which the score of the destination occupation exceeds that of the source occupation;

viii. Δ equals 1−P C , 1 minus the Proximity Coefficient;

ix. Ω equals 1−P + , 1 minus the percent of differences that are positive;

x. Γ equals Δ*Ω which is an addition to be made to P C ;

xi. A P C equals P C +Γ which is a Adjusted Proximity Coefficient;

xii. w 1 equals a positive numerical weight assigned by a using analyst to NORMCORREL d ;

xiii. w 2 equals a positive numerical weight assigned by the using analyst to A P C ; and

xiv. w 1 may but need not necessarily equal w 2 .