Defense Acquisition Research Journal Issue 95

Use of Factors in Development Estimates: Improving the Cost Analyst Toolkit

https://www.dau.edu

Contractor Type The fourth category analyzed is contractor type. The CCDR dataset consisted of prime contractor data and subcontractor data. The majority of the data—69.5%—is prime data. Because the fourth category had only two subcategories, the Steel-Dwass test is not needed. The identification of differences through the Kruskal-Wallis test is sufficient. Results are shown in Table 11. Differences in the contractor type category are found for only two WBS elements: ST&E and PSE. The small number of differences suggests that composite factor development does not require large amounts of time and effort dedicated to determining whether the data are from the prime or a sub. Rather, aggregated factor models consisting of both contractor types may be sufficient.

TABLE 11. KRUSKAL-WALLIS RESULTS FOR CONTRACTOR TYPE

Null Hypothesis Test Result

WBS Element

Alpha

N

Chi-Square

P -value

SE/PM

0.05

406 0.7777

0.3778

Do Not Reject

ST&E

0.05

374 12.064

0.0005

Reject

Training

0.05

192

0.0811

0.7759

Do Not Reject

Data

0.05

267

2.66

0.1029

Do Not Reject

PSE

0.05

149

5.3186

0.0211

Reject

CSE

0.05

50

1.6912

0.1934

Do Not Reject

Site Activation

0.05

47

0.0571

0.8111

Do Not Reject

Spares

0.05

84

0.087

0.768

Do Not Reject

56

Defense ARJ, January 2021, Vol. 28 No. 1 : 40-70