Upload
nguyennhi
View
220
Download
0
Embed Size (px)
Citation preview
Calhoun: The NPS Institutional Archive
Graduate School of Business and Public Policy (GSBPP) Thesis Day Programs and Documents
2013-03
Manpower System Analysis Thesis Day
Brief / Class of March 2013
Monterey, California, Naval Postgraduate School
http://hdl.handle.net/10945/39528
U.S. Navy Officers’ Attitudes on
the Repeal of “Don’t Ask, Don’t
Tell” (DADT)
LT Ryan Appleman
LTJG Pete McLaughlin
Advisors: Professor Mark Eitelberg
Professor Frank Barrett
2
Background
• Continuation of 5 previous NPS studies:
– 1993, 1996, 1999, 2004, 2010
• Literature Review:
– RAND Corporation reports, 1993 & 2010
– Comprehensive Review Working Group
report, Nov 2010
– Palm Center report, Sept 2012
– Psychological and sociological models of
acceptance 3
How have Navy officers’ attitudes on gays in
the military changed: - since 1993?
- since the repeal of DADT?
What are Navy officers’ impressions
regarding the effects of repeal on fleet
readiness?
Research Questions
4
Methodology
• Qualitative study:
• 59-question survey administered over a two-
week period
• Four focus-group interview sessions
Survey
Requests
Sent Out Respondents
Skipped
Questions Total
Response
Rate
573 334 24 358 62.5%
Total Collected Requested
Focus Group
Participants
Respondents 358 573 19
Completely
Filled Out
Surveys 334 334
Rate 93.3% 58.3% 5
Survey Demographics
Category Survey
Respondents
NPS Target
Population Category
Survey
Respondents
NPS Target
Population
Gender n=329 n=573 Pay Grade n=327
Male 85.4% 91.1%
O-1 2.4% 0.7%
n=281 n=522 n=8 n=4
Female 14.6% 8.9%
O-2 3.1% 5.1%
n=48 n=51 n=10 n=29
Race/Ethnicity n=330 O-3 55.7% 58.8%
n=182 n=337
Caucasian 77.9% 65.4%
O-4 28.4% 24.8%
n=257 n=375 n=93 n=142
African
American
3.9% 6.5% O-5
8.3% 8.0%
n=13 n=37 n=27 n=46
Hispanic 5.5% 5.1%
O-6 2.1% 1.2%
n=18 n=29 n=7 n=7
Asian/Pacific
Islander
3.9% 5.8% Enrollment n=329
n=13 n=33
Native
American
1.2% 1.4% Resident
62.6% 54.8%
n=4 n=8 n=206 n=314
Other 7.6% 15.9% Distance
Learning
31.6% 43.6%
n=25 n=91 n=104 n=250
Staff/Other 5.8% 1.6%
n=19 n=9
6
Data Picture
Data Analyzed Utilizing Four Methods:
1. Navy (NPS Survey) vs. Society (Gallup Polls)
2. Trend Analysis of Navy over Time
3. Demographic Breakdown
4. Focus-Group Analysis
7
Results
Navy officers vs. Society: Homosexuals in the
Military
Question 16. Gays and Lesbians should be allowed to serve openly in our
military. (2010-2012) [Homosexuals should not be restricted from serving
anywhere in the Navy (1994-2004)] (Percent who Strongly Agree or Agree)
Do you think Homosexuals should or should not be hired for each of the
following occupations…The Armed Forces? (Percent who agree they should
be allowed)
Year Navy Society
1992/1994 24.6% 57%
1996 35.8% 65%
1999 39.2% 70%
2004 49.7% 80%
2010 59.8% 76%
2012 73.4% N/A 8
Navy officers vs. Society: Same-Sex Marriage and
Benefits
Question 44. Same-sex spouses of homosexual service members should be
entitled to the same benefits provided to the spouses of heterosexual service
members? (2012) [If homosexuals were allowed to serve openly, their
dependents should be entitled the same benefits provided to dependents of
heterosexuals? (2004-2010)] (Percent who Strongly Agree or Agree)
Do you think there should or should not be health insurance and other
employee benefits for gay and lesbian domestic partners or spouses? (Percent
who believe there should be)
Year Navy Society
2004 69.2% N/A
2009/2010 76.5% 67%
2012 70.2% 77%
Results (cont’d)
9
Trend Analysis of Navy Officer Attitudes:
Policy
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
201220102004199919961994
Percent
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
201220102004199919961994
Percent
18. The current policy is good for
national defense.
33. On the whole, I like the current
policy better than the old policy.
Results (cont’d)
10
Trend Analysis of Navy Officer Attitudes:
Leadership
7. I would have no difficulty working for
a homosexual Commanding Officer.
21. A division officer’s sexual
preference has no effect on the
officer’s ability to lead.
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
201220102004199919961994
Percent
0
10
20
30
40
50
60
70
80
90
201220102004199919961994
Percent
Results (cont’d)
11
Trend Analysis of Navy Officer Attitudes:
Comfort and Habitability
3. I would prefer not to have
homosexuals in my command.
20. I feel uncomfortable in the
presence of homosexuals and have
difficulty interacting normally with them.
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
201220102004199919961994
Percent
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
201220102004199919961994
Percent
Results (cont’d)
12
• Major Themes from Focus Groups
– Nothing changed
– No adverse effect of repeal on unit cohesion,
morale, readiness
– Leadership and professionalism matter
– Equal benefits for equal service
Results (cont’d)
14
• Views have shifted dramatically since 1994 from strongly negative to strongly positive toward repeal of DADT and homosexuals serving openly in the military
Data Analysis: Conclusions
15
Data Analysis: Conclusions
• A vast majority of Navy officers say they have no
difficulty serving with homosexuals, even though
a number claim to feel uncomfortable sharing
living quarters with a homosexual
16
Data Analysis: Conclusions
• Higher-ranking officers and officers with 16-20 YOS are less tolerant than other YOS groups
17
Recommendations
• Further analysis into reasons behind 16-20 YOS group acceptance level
• Continue to monitor post-repeal effects on fleet readiness, particularly fairness and potential harassment
18
NAVY ENLISTED RECRUITING:
ALTERNATIVES FOR IMPROVING
RECRUITER PRODUCTIVITY
LCDR Todd C. Winn
Advisors: Professor Jeremy Arkes
Professor Steve Mehay
20
Background
• Efficient management of the Navy’s enlisted recruiter force has become increasingly important.
– 1980s Production per Recruiter (PPR) 2.4
– 1990s average PPR 1.3
– 2000s average PPR 0.86
• New strategies needed in an effort to improve recruiter productivity
• Alternative to the current 8-month “On-Boarding” Process
21
Primary:
Can NRC increase recruiter productivity by
altering the on-boarding process from eight
months to six months with minimal to no cost to
the Navy?
Research Question
22
Methodology
• Interviews conducted over the past 7 months – Navy Recruiting Orientation Unit (NORU) instructors-5
– LPOs (formerly called RINCS)-26
– LCPOs (formerly called Zone Supervisors)-12
– CRs-2
• Policy Analysis-comparing current on-
boarding timeline vs. alternative
• Focus of study (E-5/E-6) active duty enlisted
production recruiters
23
The Inverted U
Inverted U Enlisted Recruiter Production Curve
FY94-FY02 Average New Contracts
(From Samuelson et al., 2006). 24
NORU
(5 weeks)
PCS
(30 days)
NRD
Basic PQS
(45 days) SDAP$450
Recruiter
tour
starts
NRD
RDB
(45 days)
Recruiter
is fully
qualified
0 months 3.75 5.25 8.25 1.25 2.25
NRD
Advance PQS (4.5 months)
Current On-Board Process
NORU
(5 weeks)
NRD
Basic PQS
(30 days)
Recruiter
tour
starts
NRD
RDB
(45 days)
Recruiter
is fully
qualified
0 months 3.75 6.25 1.0 2.25
NRD
Advance PQS (4 months)
Alternative On-Board Process
Cost Comparison
Differences in cost
• PCS cost is approx. $12,000 per recruiter based on TEMDUINS orders followed by PCS orders
• 2010-Avg Navy PCS cost $4,500
• TEMDUINS cost $7500 (travel, per diem, lodging)
• Under the proposed system, the additional TAD cost per potential recruiter should be less than $600
27
Funding
• NPC writes PCS orders to NRD, NRD writes (not funds) TAD orders to NORU
• NRDs do not have the funding to send recruiters TAD to NORU
• Cost for TAD orders $8,100
• Alternative funding
• Reallocation of funds from NPC to NRD
• Funding code assigned that would not come out of NRDs budget
28
Costs/Benefits of Alternative On-
Boarding Process
•Cost of new approach ≈ $600/recruiter •1000 new recruiters/year total cost ≈ $1.8M/year
•Benefit will be reduced recruiters needed if recruiters sign more contracts in tour
•Depends on:
•# recruiters
•Productivity per recruiter
•Increased # contracts per tour
•Increased # contracts per tour is unknown: •More from shortening the on-boarding process
•Less from sending TAD to NORU 29
Back-of-envelope initial estimates
Results of sending a recruiter TAD to NORU and the additional productivity needed
to break even.
30
Recommendations
Conduct a Randomized Experiment:
• Proposed design: • Treatment group - 600 enlisted sailors receive PCS
orders to their NRD, complete basic PQS, then report to NORU
• Control group – All recruiters who are not in the treatment group and are randomly selected
• Recommended length of study is 3 years with intermittent results after every year
• Reallocation of funds from NPC to NRC = $4.86 million
• Analysis of data to determine results: • Compare and contrast productivity between the two
groups
• End of experiment surveys with LPOs from both groups.
31
Retention Elasticity and
Projection Model For U.S.
Medical Corps Officers
CDR Abdullah AlShehri, RSNF
LCDR Hyrum Brossard, MSC, USN
Advisors: Professor Dina Shatnawi
Professor Yu-Chu Shen
33
Background
• Medical Corps Mission Statement: “We enable readiness, wellness, and healthcare to Sailors, Marines, their families, and all others entrusted to us worldwide be it on land or at sea.”
• Environment has Changed
– GWOT
– Economic Recession
• Replicate a previous study that was conducted by
CNA in 2002 using data from FY02 through FY11
– Civilian-Military Pay Gap’s effect on Navy Medical Corps
Retention
34
Primary: 1. How does a change in the civilian-military medical
providers pay gap affect the retention of Navy medical specialists?
2. What are the projected retention rates for Navy medical providers, and how would adjusting special pay incentives influence their retention?
Secondary: 1. Has the prolonged GWOT and recent economic
downturn influenced the Navy Medical Corps retention rate from FY02 through FY11?
Research Questions
35
Methodology
• Multivariate probit model
– Dependent Variable: • Whether a person stays or leaves the Navy after their initial obligation
– Explanatory Variables: • Variable of Interest
– Civilian-Military Pay Gap
• Demographics
– Gender, race, age
• Military Experience
– Years of Service, Rank, Accession Source
• Years/Specialty Dummy Variable
• Forecasting Methods:
– Mean, Moving Average, and Exponential
Smoothing
36
Data Source
• Bureau of Medicine and Surgery (BUMED)
• BUMIS data (FY02 through FY11)
•Special Pay
• Incentive Special Pay (ISP)
• Multi-year Special Pay (MSP)
•DFAS
• Database for Base Pay, BAH, & BAS
•Medical Group Management (MGMA)
• Civilian Physician Compensation Data 37
Preliminary Data Analysis
Table 1. Number of unobligated providers at a decision point to leave the Navy
Fiscal Year
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Total
N= 634 642 726 633 534 475 377 331 324 284 4,960
Stayers= 453 484 563 449 329 303 207 200 174 164 3,326
Figure 1. MC Retention Rate vs. Civilian-Military Pay Gap
38
Description of Demographics
Figure 2. Delta between FY02 & FY11
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
FY2002
FY2011
39
Results
0%
2%
4%
6%
8%
10%
12%
Main Model Primary Care Surgical Specialties Other Specialties
% Change of Probability of staying in the Navy if Civilian-Military pay gap is reduced by $10,000
40
Results (cont’d)
USUHS= Uniformed Service University of the Health Sciences
AFHPSP Deferred= Armed Forces Health Professions Scholarship Program
FAP= Financial Assistance Program
-20%
-15%
-10%
-5%
0%
5%
10%
USUHS AFHPSP Deferred FAP
Probability of Retention in The Navy (Relative to AFHSPS)
41
Elasticity Results
*p<0.1
Specialty Elasticity
Family Practice –0.37
Internal Medicine –0.51
Pediatrics –0.19
Occupational Medicine –0.06*
General Surgery –0.76
Neurological Surgery N/A
Orthopedic Surgery –0.68
OB/GYN –1.42
Ophthalmology –0.42*
Otolaryngology –1.0
Specialty Elasticity
Urology –0.45
Anesthesiology –0.54
Dermatology –0.14*
Emergency Medicine –0.50
Neurology 0.75*
Pathology 0.07*
Physical and
Rehabilitation N/A
Psychiatry –0.46
Radiology –0.74
42
Results
• The effect of GWOT: • The retention probability decreased by 14.1 percentage points after
FY05 compared to FY02-FY04
• Probability of leaving the Navy is 23 percentage
points higher for African Americans compared
to their White counterparts • Asian and Hispanic Physicians are not statistically different from their
White Counterparts
• Military Tax Advantage showed no significant
change on retention 43
Recommendations
1. The Navy should increase or shift its funding in support of the USUHS program.
2. Increase the MSP amount & offer the MSP earlier to LT’s and LCDR’s before their initial obligation of service is over (Regardless of Specialty).
3. Increase MSP amount to primary care and surgical specialty categories (Regardless of Rank).
44
Evaluating the Tailored Adaptive
Personality Assessment System
(TAPAS) on Navy Recruits
LT Jessica Fahrman, USN
Advisors: Professor Steve Mehay
Professor Elda Pema
46
Background
• Civilian studies have documented the effect of non-cognitive attributes on
• Job performance
• Schooling decision
• Army has used TAPAS to assess the “whole-person”
– Effect of TAPAS scores on attrition
• Within Navy, attrition rates vary within education/AFQT categories
– TAPAS may help to better categorize recruits
47
1. Does TAPAS bring in new information?
– Evaluate the correlation of TAPAS with observables (demographics, ASVAB, AFQT, education).
2. Do TAPAS scores predict whether an applicant will enlist?
– Estimate the probability that an applicant enlists, controlling for TAPAS and all other observable background characteristics.
Research Questions
48
•TAPAS is a computer adaptive test designed
to capture personality characteristics.
– 15 total measured personality facets
– 5 broad categories (extraversion, conscientious,
agreeableness, emotional stability, openness to
experience)
– 2 composites
• “can-do” – designed to predict job knowledge
and training graduation
• “will-do” – designed to predict motivation,
commitment, and attrition
TAPAS Description
49
Methodology
• TAPAS administered to Navy recruits at
MEPS from April 2011 to March 2013
1.Measure correlations of TAPAS scores
with select demographics and cognitive
test scores
2.Estimate a probit model for probability to
enlist, conditional on TAPAS scores, and
other observables.
50
Composite scores by Race
•TAPAS composite scores among races compared to
Caucasians
Standard Deviation :
Will Do: 17.432
Can Do: 17.271 51
TAPAS scores by Race
•TAPAS Traits with strongest correlation among
races
Asian Black Hawiian Hispanic Std Dev
achievement -0.179*** -0.0758*** -0.0718 -0.0238 0.507
adjustment -0.0823*** 0.0191 -0.0493 0.0112 0.526
dominance -0.177*** 0.00962 -0.129** 0.0455*** 0.553
inteleff -0.148*** 0.0906*** -0.117** 0.0179 0.572
nondelinquency -0.121*** -0.0410** -0.0912* -0.0446*** 0.497
physical -0.129*** -0.118*** -0.00891 -0.0327* 0.597
selfcontrol -0.0327 0.0761*** -0.0390 -0.0175 0.545
sociability -0.0755*** -0.0666*** -0.0887 -0.0277 0.565
tolerance 0.110*** 0.0677*** 0.163*** 0.112*** 0.531
****Control variables: gender, age, marital status, number of dependents,
waiver, education, AFQT, and ASVAB subtests.
Base group: Caucasians
*** p<0.01, ** p<0.05, * p<0.10 n=11,042
52
Composite scores by Education
•TAPAS composite score correlation among
education levels compared to HSDGs.
*** p<0.01, ** p<0.05, * p<0.10
53
TAPAS scores by Education
•TAPAS Traits with strongest correlation among
education levels
NHSD College Std Dev
achievement -0.0634* 0.0397** 0.507
cooperation -0.0720** 0.00689 0.454
dominance -0.119*** 0.0208 0.553
eventemper 0.0265 -0.0322** 0.456
inteleff -0.0268 -0.0468** 0.572
nondelinquency -0.0614* -0.0339* 0.497
physical -0.0937** 0.116*** 0.597
*** p<0.01, ** p<0.05, * p<0.10 n=11,042
****Control variables: race, gender, age, marital
status, number of dependents, waiver, AFQT, and
ASVAB subtests.
Base group: HSDG
54
Non-Cognitive vs Cognitive
•TAPAS Traits with strongest correlation among
AFQT and ASVAB sub-tests
55
Probability to Enlist
•The likelihood an applicant enlists based on
composite scores
Probit controlling for
cognitive tests
Probit without controlling
for cognitive tests
Coefficient 0.00262*** 0.00168**
Std error (0.000818) (0.000795)
Partial effect [0.00103]*** [0.000660]**
Coefficient 0.000445 0.00332***
Std error (0.000758) (0.000733)
Partial effect [0.000175] [0.00131]***
Can-do
Will-do
*** p<0.01, ** p<0.05, * p<0.1
Standard errors in parentheses
56
Results
• TAPAS scores vary significantly by
– Race
– Education level
• Minimal to no correlation between cognitive and non-cognitive test scores
– It appears that TAPAS is picking up new information about a recruit
• TAPAS composites significantly predict the probability of an applicant to enlist
57
Future Work
• As the cohorts age evaluate TAPAS’s predictive effect on Navy attrition
1. DEP attrition
2. Boot camp attrition
3. First term attrition
58
Manpower Requirements
Estimation for UCLASS
Squadrons
CDR Gary Lazzaro
Advisors: Professor Bill Hatch
Professor Cary Simon
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
60
Background
• Largest cost of any new system is manpower
– Up to 70% of total system cost
• Most research is performed on technical aspects of UCLASS aircraft
– Scant research in manpower requirements
– “Autonomy” may reduce manpower
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
61
• Primary:
– What are the manpower requirements for a deployable UCLASS squadron?
• Secondary:
– How will UCLASS manpower requirements compare to F/A-18F manned squadrons?
Research Questions
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
62
Methodology
• Determine squadron manpower requirements
– Squadron support departments
– UCLASS aircraft maintenance
– UCLASS aircraft operators
– Deployable squadron vs. deployable detachment
• Assumptions must be made for some
UCLASS squadron parameters
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
63
Data Picture
• Squadron manpower requirements may be
grouped into four common areas
– Some requirements depend upon number of
aircraft, and others depend upon unit configuration.
64
• Data from squadron manpower documents
– Analyzed VFA, VAW, VRC, VP & HSL units
– Identified common manpower requirements
• Theoretical analysis to determine UCLASS
aircraft operators requirements
– Autonomy vs. ethics
– AVO/MPO requirements change during mission
Data Explained
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
65
Results of Data
• Fully deployable squadron estimate:
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
• Detachment concept squadron estimate:
Manpower comparison Officers Enlisted Total Difference
Completely deployable
UCLASS squadron36 161 197 86
One F/A-18F squadron 44 239 283 --
Manpower comparison Officers Enlisted Total Difference
UCLASS shore component
with five detachments103 540 643 772
Five F/A-18F squadrons 220 1195 1415 --
66
Data Analysis: Results
• Total annual manpower costs (FY13 $):
– UCLASS shore component
with five detachments: $60M
– Five UCLASS squadrons: $89M
– Five F/A-18F squadrons: $119M
• Manpower costs compared to five F/A-18F
squadrons:
– UCLASS shore component
with five detachments: $62M less
– Five UCLASS squadrons: $30M less
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
67
Recommendations
• Manpower planners should consider the detachment concept configuration for a future UCLASS squadron
• Research assumptions will need to be revisited as the UCLASS program matures
• UCLASS aircraft operator requirements should be allowed to vary during a mission
UCLASS = Unmanned Carrier Launched Airborne Surveillance and Strike System
68