“Calhoun

Institutional Archive of the Naval Postgraduate School

Calhoun: The NPS Institutional Archive DSpace Repository

Theses and Dissertations 1. Thesis and Dissertation Collection, all items

1980-06

RTC CREDO : tnitial assessment.

Hawkins, Stephen G.

Monterey, California. Naval Postgraduate School http://ndl.handle.net/10945/19046

This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.

Downloaded from NPS Archive: Calhoun

: Calhoun is the Naval Postgraduate School's public access digital repository for / (8 D U DLEY research materials and institutional publications created by the NPS community. «ist : Calhoun is named for Professor of Mathematics Guy K. Calhoun, NPS's first

NY KNOX appointed and published scholarly author.

LIBRARY Dudley Knox Library / Naval Postgraduate School 411 Dyer Road / 1 University Circle Monterey, California USA 93943

http://www.nps.edu/library

i] i 4 | “| I i Vy ; + AY av , yi vip? ya", A } ay oy anes Ny a 24 in) Wh ve nah Ase N * Mi pan af pe ee rhe! May'te et * ‘! ‘aha he ' oF \ ¢ i { Re van \ Tau tape t : 1 4 | i » aN z q “ta! Par rey Lal a EMERG Wy SL +" i a ae : te LS) \ yee a, 4 s 3 , b wi ) a , ts ; ' - ie i i 3 ni i? 4 us 4 . pet HAN iN, ay Na by, we ' Age \s ANS ve a ~ _

Mey S a, ay MW ue oe hy ah fey Alas Bo % Lb 1, LYS oy ss . Sy 7

ys : t » ao. SL esol | o ue } : x ' ' \ sf \ i iY i] \ iF t ' Vy ms A he ii ; a ny, 7 ; ry = iN yl, toe 4 i i ih i i , ae i by j%" ; 4 Ns reas % + is Nok i i 1 > i * i as. wee a i - ~ } pha a 7 %! vi mm » = . ' aKa pes \ . pho gy tee Dar a er aS . ANS 1 ; pa eo, iue if , A iN AN ay Ne AY a . iy | | ~< ON ee: SET RS eco Reete ne 8 i. , ; ' ean i) a4 i=) fe tos wT oe > ane Sa i 7 ; a : =| ; Ores nie = I \ . Ne Lea = 3 aij a hai ' +% ite 5 _ ae { Lf Fy ; = ite me | ; . i e Y : ; 1 ce J - ~ > ! i ™~ ft i - ! a - \! L i P 4 $y , ' i} ~ ' i . ' i ' i y vy i 4 1 i = P - id Lay 1 iF f s 1] 2 ? me - a _ #5! P 7 ' ad . ae - } a ra = ¥ i ee . . re y fs 1 . , ' ' é e 1h hp J 4 4 - ede = ail P= hol , i ' a a : 6 oe = ae 467i —_ 1 = = ¥ a - ' oe re = ple = .__ == . - , é ps awit : = . re Pi ae , sm a Atte wet el ees , ea . . “Th ~ Fy, me 4 wa Sas tae * a : i si ay i, 4 \ 4 ' ls pels - My oe i. om pe 30 4s 2 t i : 7 ts J5s ' fy' | ten i, La ee (Male fom Pt ee oe ; i - Pe il q i J ma”, al det . et Ae. cre » =| ae a : 0 4 4 Bs a" " é ee : r "i "Gs <A): pts i . , " 4 be” ¥ i Fi dg = so vin APES oe Lye od ie Ae ppes es GA ppm ; 2 A TS ae ee Bie ert es 6: f mI AF Oe a Fi ie Ae pigs “if. on ; , f cad rs on > ry 3 a " tap a= ae ta'= rofl i iy ] 4 s 4 ita ie 2 at 0 anf a yah } fe, pee i 4 ' e 8 st aoa a ft af} Tr sy ale i? ir - Sold Cy ie 7 i 0 Se oi 7 xy i i aps! 4 i Biel “iy bb aS A uh os reAD Yee} try ies ie th it oe lee peu J pe aE Pena" pli si ‘ier 8 2 oie ik fot ve " mF aa eeehcnes Tie: ey ‘i ee, ees 1 okt PS gett opts r - wily : a i je a a fi 7 7, ' { Akg we iis ee 4 vif ice wis Ol 4 ee ah Z A ees up oy ope © ,, HY tie He heen da eel eps ch ach ie Lt ee Lp ite of) es Cotas Weide é * Hii dtie EGE LATE YJ 1 re By me, | cal F i 4 . '¢8 4 ake of , pirat ¢ f i ; } \ if nia \ ee i yf ay Toy ee, deh in bate ts Wage e HEL. MCU rie ied yh ars h Baga Nel Stile yy Me Hage 4" i" . ty, Kae P : . 2 s A Meeigys ts es Ve , af, * ry J * vas é 44 4 ¢” Ly af Bs ae js FA on) aie d 5 el ¢ vi He 8a fe tae 4 . ‘gy LA $ oy Mes TA Teena] d | AAS MAS ALaee we ang ae AAD ee ry YY 1 Sola aaeyas ey al MOUMPOE NL AVIRA Wioat nae AA i my ee .. s es A] \ a 1) bi fi Ay 448 yp ut A Hatt i" igi : ute NT fat in My F H {! Bib i ty fy hea ay RA i, y

PUDLEY KNOX LIBRARY Rea Ba Poe OUTS SOCRAOTL Winer TERE Y Cos de 24560

NAVAL POSTGRADUATE SCHOOL

Monterey, California

THESIS

RTC CREDO: INITIAL ASSESSMENT by Stephen G. Hawkins

June 1980

Thesis Advisor: R. S. Elster

Approved for public release; distribution unlimited

T195712

a aa rr SECUMITY CLASSIFICATION OF THIS PAGE (Whan Date Entered) Ne oe ee REPORT DOCUM , READ INSTRUCTIONS = | ENTATION PAGE ) BEFORE COMPLETING FORM = - |

4. TITLE (and Subterte) . TYPE OF REPORT & PERMOD COVERED

RTC CREDO: INITIAL ASSESSMENT Master's Thesis; June 1980

6. PERFORMING ORG. REPORT NUMBER

- CONTRACT OR GRANT NUMBER(2)

7. AUTHOR(e)

Stephen G. Hawkins

EmEnwT. PROJECT, Tasx K UNIT NUMBERS

12. REPORT OATE June 1980 13. MUM@ER OF PAGES

18. SECURITY CLASS. (of tnte report)

9. PERFORMING ORGANIZATION NAME AND ADORESSE

Naval Postgraduate School Monterey, California 93940

CONTROLLING OFFICE NAME ANDO ADDRESS

Naval Postgraduate School Monterey, California 93940

MONITORING AGENCY NAME & ACORESSE/( I! differant from Cantreiting Office)

Unclassified

Naval Postgraduate School Monterey, California

Se. OECLASSIFIC ATION SCHEDULE

OCOWNGRADING 16. OF'STRIBUTION STATEMENT (of thie Repere)

Approved for public release; distribution unlimited

17. OrsSTRiBUTION STATEMENT fal the smetract entered in Bieck 20, If different frem Repert)

168. SUPPLEMENTARY NOTES

19. KEY WORDS (Continue on reverse aide if neceesary and identify by block number) CREDO, Attrition, Apprentice Training Division, Positive Motivation Unit

20. ABSTRACT (Continue an reveree eide if nececseary and identify by bieck mamber) A statistical analysis was conducted to initially assess the effectiveness of the experimental program, RTC CREDO, to counter first-term enlisted attrition in the Navy.

Cross tabulation, discriminant, and multiple regression analyses were performed to examine various experimental and con- trol groupings and their observed attrition.

The results of this study were tenative and inconclusive because of the recency of the experiment. Indications for

DO , an 1473s coi Tiow oF | wov 6813 OBSOLETE

(Page 1) l SECURITY CLASSIFICATION OF THIS PAGE (Wren Deore Entered)

Ene eee eee eee i 2 pe Secumvy CL asgrPicavTian OF Tumis PagesWren Nave Entoved:

20. (continued) future research were made, which, with the passage of time and accumulation of more data, will enable fuller and more comprehensive studies of the effectiveness of RTC CREDO to be made.

DD Form 1473

Approved for public release; distribution unlimited

RTC CREDO: INITIAL ASSESSMENT

by

Stephen G. Hawkins Lieutenant Commander, United States Navy B.A., Tulane University, 1970

Submitted in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE IN MANAGEMENT

from the

NAVAL POSTGRADUATE SCHOOL June 1980

ABSTRACT

A statistical analysis was conducted to initially assess the effectiveness of the experimental program, RTC CREDO, to counter first-term enlisted attrition in the Navy.

Cross-tabulation, discriminant, and multiple regression analyses were performed to examine various experimental and control groupings and their observed attrition.

The results of this study were tenative and inconclusive because of the recency of the experiment. Indications for future research were made, which, with the passage of time and accumulation of more data, will enable fuller and more compre-

hensive studies of the effectiveness of RTC CREDO to be made.

i,

ae.

Pease Or CONLENTS

eee DO Geren seems 6 GNc « 6 6 6 lee Be PROOIEAMG . s 8 Cw ll BRUCE CO oO EeMcriNs . + < s « « « « 8 6 © © #8 Pee REMENTAL DESEGN ..... + © © «© © © «6 « A. SUBJECT SOURCES AND SELECTION PROCESSES l. Apprentice Training Division ..... Ze eeeOcttEnyeeMOotiVatron Unit . . . « « « PMD OGREENING , . . « « « © © «© © © © « « [MemermURES AND RESULTS . . . . « « 2 «© «© «6 « « BEPROGMMURE SU gs. 6 6 6s te 8 le et lt ebbcer rr inves RESULTS . .. « « « « « » « « l. Aggregated Experimental vs. Pomgregaiea GONErOL . . 2 « 5 « » « « 2. Aggregated Early Group vs. BaGEegatcead Late Group .. . s+ « « « « 3. Early Experimental vs. Early Control . 4. Late Experimental vs. Late Control. . 5. Late ATD Experimental vs. Late ATD omMtesOll ses ss ss 6 ee ewe 6. PMU Experimental vs. PMU Control. . 7. Summary of Descriptive Results ... SPA TEREUDTON FENDINGS . . . . + +. 2s 2 «es D. REGRESSION ANALYSES RESULTS ...... -

1. Summary of Regression Analysis Results

Page al:

pe Es. say

a dey

gnc ® 24

® 34

45

255

® aie

IV. CONCLUSIONS AND RECOMMENDATIONS ..... Perrin TATIONS ON CONCLUSIONS ......« - B. CONCLUSIONS/RECOMMENDATIONS ....e.-.

APPENDIX A: INTERSERVICE SEPARATION CODE LISTING MOR@MCORES@ 6-0 4 4 5 ss. sk 6 te sl

PePPENDIX B: TYPICAL RTC CREDO WORKSHOP FORMAT .. Meee x C: SAMPLE HARD CARD ...... eo « feet xX Ds: ATD DATA SHEET .....-. «+ « « «

APPENDIX E: DESCRIPTIVE STATISTICS FOR INDIVIDUAL PieCmCREDOMNORKSGMORS ££. 2. « « « %

APPENDIX F: SUPPLEMENTAL DESCRIPTIVE STATISTICS . feeereor REFERENCES . . « . © « © © © © «© © © «© «© «

memes DISTRIBUTION LIST .....-.-+ ++. -s « «

7s

76

as

8 0

85

8 6 89

106

1C8

i...

EL .

uy .

Wi.

pil.

er .

x

mol

Poel.

Pore.

Pov.

pot wor LAB LES

NAVY MALE NPS CUMULATIVE PERCENT ATTRITION GeeebocAn YEAR . 2. 2. 2 5 8 se 8 8 le lt

INITIAL RTC CREDO PERSONNEL DISTRIBUTION Semperosa LAKES 2 5. 2. 6. 6 6 8 ew ltl ltl ll

Siam RIC CREDO PERSONNEL DISTRUBUTION .... DEMOGRAPHIC DESCRIPTION OF THE TOTAL SAMPLE .

AGGREGATE MEANS AND DISTRIBUTIONS FOR EXPERIMENTAL VS. CONTROL GROUPS ... .

DISCRIMINANT ANALYSIS RESULTS FOR AGGREGATED EXPERIMENTAL VS. CONTROL GROUPS ... oe

AGGREGATE MEANS AND DISTRIBUTIONS FOR EARLY Se GROUPS 6.2 2. 6 se tl ltl lll ll lll ll

MEANS AND DISTRIBUTIONS FOR EARLY EXPERIMENTAL See EO, GROUPS. . . . 2 «+ «© © « @ 6 « «

DISCRIMINANT ANALYSIS RESULTS FOR EARLY Pee BREMENTAL VS. CONTROL GROUPS .... .

MEANS AND DISTRIBUTIONS FOR LATE EXPERIMENTAL VS a CONTROL GROUPS ® ® 2 @ ® 2 ® 2 2 2 e 6 6 ®

DISCRIMINANT ANALYSES RESULTS FOR LATE PREM ENTAL VS. CONTROL GROUPS ......

MEANS AND DISTRIBUTIONS OF LATE ATD BXPEREIMBENTAL VS. ATD CONTROL ......... .

DISCRIMINANT ANALYSIS RESULTS FOR ATD PeeePeREMENTAL VS; CONTROL GROUPS ......

MEANS AND DISTRIBUTIONS FOR PMU EXPERIMENTAL VS. CONTROL GROUPS Sects. “= 6) 5 Ae °

DISCRIMINANT ANALYSIS RESULTS FOR PMU Pee RIMENTAL VS. CONTROL GROUPS ......

SUMMARY OF RESULTS OF RTC CREDO COMPARISON ANALYSES e 2 ® Ld e ® 2 e 2 id ® e e ® 6 e 2 o ®

Page

eZ

Zi 23

ANS

bal

a2

36

38

Sy

42

43

47

4&

ie

a)e

=yi

XIX.

2

XXI.

Peni. .

Peril.

LV .

DISTRIBUTION OF ATTRITES AMONG SEPARATION HROc aM DrolGNATORSm\oPD) CODES .... « « «

MEANS AND DISTRIBUTIONS OF ATTRITES AND TOTAL SAMPLE FOR SELECTED VARIABLES AS OF MARCH 1 9 8 0 e e e e e ® ® e e e e e S e e e

ATTRITION AS OF MARCH 1980 FOR VARIOUS Gee see Te es 2 s+ « 5 6 se ltl tll lll

RACIAL DISTRIBUTION OF ATTRITES AS OF fmen L980: WHITE” VS. NON-WHITES ......

REGRESSION ANALYSIS OF THE PROBABILITY OF SURVIVAL (NON-ATTRITION) PRIOR TO NINE M@iipnowOMwomRV ECR ~ 6 « 6+ «© 5 & 6 oe ew lw lw

REGRESSION ANALYSIS OF THE PROBABILITY OF SURVIVAL (NON-ATTRITION) PRIOR TO NINE POMUMGMORUSGERVIGE . . . © « « « « « e « « «

MEANS AND DISTRIBUTIONS FOR ATD PERSONNEL: RIL IS EVE 5

MEANS AND DISTRIBUTIONS FOR ATD EXPERIMENTALS: MMP UE os sss el lll ll ll lt

MEANS AND DISTRIBUTIONS FOR ATD CONTROLS : oe BL WAS) 5 9 TEV 5G in acm a nr a

MEANS AND DISTRIBUTIONS FOR LATE PERSONNEL: TONS) TS 9 nS

MEANS AND DISTRIBUTIONS OF LATE PERSONNEL: ATD/PMU BY EXPERIMENTAL/CONTROL ...... .

MEANS AND DISTRIBUTIONS FOR LATE EXPERIMENTALS :

ID RES 5 TTS SS

MEANS AND DISTRIBUTIONS FOR LATE CONTROLS: ATD VS ® PMU e ® ® e e e e ® d ® e ® ® e e e e

60

61

63

6 4

69

G2

93

95

oF

oN,

mee

103

105

nO .

i.

B2 .

meow eo rT GURES

Experimental Design Analyses for RTC CREDO...

Experimental Design Analysis: Aggregated meeexrimental vs. Control Groups .....e..

Discriminant Analysis Histograms for Aggregated Experimental vs. Control Groups ...

Experimental Design Analysis: Early vs. BUBEEMEGG@UDG@ 5 9 c5s 6 3s 6 6 « s+ «© © © © © 06 © #8

Experimental Design Analysis: Experimental Memmecontro! Groups (Early) .......6e-.

Discriminant Analysis Histograms for Early Meperimental vs. Control Groups ......e .

Experimental Design Analysis: Experimental Bemmmeantro! Groups (Late) ......+.+s++se-s.

Discriminant Analysis Histograms for Late Experimental vs. Control Groups .......

Experimental Design Analysis: ATD Experimental Sere wee COntrOl GroupS . . «+ »+ © »« = » « «+ « «

Discriminant Analysis Histograms for ATD meeeerimental vs. Control Groups ......e«.

Experimental Design Analysis: PMU Experimental SemmeecicecO! GrOUDS . . s « s+ « s « « © «© = «© «

Discriminant Analysis Histograms for PMU meeperimental vs. Control Groups .....e. -

Experimental Design Analysis: ATD Early vs. RRPeMEGOW OS Cecio ce 6 « 6 6 8 © © © © «© © «

Experimental Design Analysis: ATD Early vs. pp Gate Experimental Groups .....+ +. «

Experimental Design Analysis: ATD Early vs. Pembmlate Control Groups .... +. +++ « «© « »

Experimental Design Analysis: Late ATD vs. PPEMEGNaOU OSes «= 4 » © + 6 © © © 6 «© © © 6 © «

Page 25

36

32

35

a7)

40

41

44

46

49

yl

54

Sy

94

96

98

Experimental Design Analysis: ATD Experimental/ Control vs, PMU Experimental/Control Groups. .. . 100

Experimental Design Analysis: Late ATD Experimental vs.. PMU Experimetal Groups .... 102

Experimental Design Analysis: Late ATD Seino SnmmeMumecOnc TOL GrOUPS <5 «cee + «© yee LO4

1Q

I. INTRODUCTION

A. PROBLEM

A high level of first-term attrition has been a consistent problem facing the U.S. Navy in recent years. These levels of attrition are depicted in Table I, and represent the total attrition (male only) experienced in the first five years of service for each fiscal year's enlistment cohort since the beginning of the AVF. Also included in Table I is attrition attributable to failure to meet minimum behavioral or per- formance criteria as indicated by the Interservice Separation Code entries.*

With the exception of 1974, the data in Table I show a gradual’ trend of decreased attrition. However, for those years for which 48 month data are available, attrition for the first 12 months of service increases from 42 to 51 percent of the total attrition. The greatest loss experienced from each cohort during the 48 month enlistment period occurs during the first three months of service, the virtual entirety of which is devoted to recruit training and apprentice/technical training. This loss represents from 20 to 27 percent of the GOtal attrition for the 1973 to 1976 cohorts. The magnitude

of this loss may be inflated and the a degree self-inflicted.

. See Appendix A for ISC Listing.

Jig

Love el ee

ELeOC SLeLe

C8 TE 86° OP

OU 8c SiGm ce

87-LE

*PTIAaRIIO asouewrosyiad Io TeroTAePysaq aTqezdaooe TeulrurTw Jeeu o7 eInTTeJ OF enp sassoT bhurjZuasaidaiI sapoy uor7eriedas sxsOTAIaSI9azUrL

“6L6T

“OE

*3™desg Jo se AZUSeAAND

‘9TtaA FAOYOD DOWA

IOJ spueIS g-9 DSI ‘/suoseal [T[P IOJ UOTITIZAS [PROF ST [TPIOL :a70N

*qOuNnos

YWHA TWOSTH AG NOILIYLLY LNAOWdd FATLVINNNO SdN FIVW AAWN

a0) Gil S0°92

vy 'tc vc TE

8T° 82 Sc 1g

Vs Ge ie Le

GU 92 90°PE

We Te

oe oi 99° S2¢

GU CC IC.36c

Cie 9G pe) Sa At

TO°L2 SS) 7t

NG SE T6°OE

Osa c

8ST ET Gs I

8S°LT 60°C

C6 6T Se 9C

Ue BC €6° 8d

LoeEc TS Of

69°0¢ 88°9C

vVc-6T

I Wiaws

= ee 26 2 iis 88° 6 88°CT 09 Ut vy OT aie St ce SL ce. Cll Vil te Ee LY oo 91 Ee, Cr EGacc Lee On 8L°8T Co cm V9 €¢ Soe S6°8T Gee tt ol) SG L9°8T St. 9T Oc [i Te 1c CI Sa ST =e CASE

(SURUOW) BOTAASS JO YRZbusay

~) ! oO

G9 ou TeI0OL 6L6T

Sao Dot Te4F0OL 8L6T

Sag Jot TeIOL LLET

8-9 USI TeIOL I9L6T

8—=9 JSl TeIOL SLE6T

8-9 SOL Te4OL Let

8-9 OSI TROL EL6T

IL

It has been suggested that some recruits labeled as un- Suitable because of motivational or behavioral problems may have been separated from service unnecessarily [Plag, 1963]. The Navy may be depriving itself of competent manpower under the rationale of separating marginal performers is easily as possible.

The failure to meet minimum behavioral or performance criteria accounts for the majority of all Navy enlisted at- trition. With few exceptions, attrition rates for these causes follow the same patterns over time as do total at- trition rates.

Although the trends show attrition to be decreasing, the problem of first-term attrition continues to deserve high- level attention. It was estimated in 1976 that the annual cost of DOD first-term attrition was approximately one billion dollars [Defense Manpower Commission, 1976]. The task of re- ducing the amount of lost dollars as well as lost manpower is multifaceted.

The identification of personnel having a high risk of at- trition prior to service entry has been addressed. Lockman [1976], for instance, has used variables such as education level, mental ability, age, race, and number of dependents to predict enlisted attrition from the Mavy. Other research has shown that motivational and behavioral factors are prime causes of attrition [Martin, 1977; Jenkins, 1977]. Additional

psychological testing techniques have been advanced as means

3

to identify potential attrites in recruit training as early as the first day [Spielberger and Barker, 1979].

The use of realistic job preview (RJP) films as a method to facilitate the socialization processes of new recruits has been shown to be promising in reducing attrition [Horner, Mobley, and Meglino, 1979; Mogley, Griffith, Hand, and Meglino, 1979]. For those individuals who have displayed maladaptive personality traits and/or sub-satisfactory performance, interven- tions have been established to prevent their attrition.

One such effort is the Behavioral Skills Training Program (BEST) which was established at Naval Amphibious Base Little Creek, Virginia, in July 1979. This program was designed specifically for individuals who have displayed marginal per- formance and disciplinary problems. The BEST program stresses military discipline and intense physical conditioning. Person- nel to attend BEST training are designated by their commanding officers. The four week BEST program strives to instill a sense of goal achievement in each attendee in order to make him a more productive service member when he returns to his unit. Preliminary reports of BEST's effectiveness have been encouraging. * A more extensive evaluation of this program is currently being done by the Naval Personnel Research and

Development Center, San Diego, California.

From personal Communication with CDR George Anastasi, OP-=135K,

14

Another counter-attrition program in the Navy involves the Chaplains Religious Enrichment and Development Operation (CREDO). CREDO was initially established in 1971 in San Diego as a special chaplain staffed approach to the Navy's Drug Rehabilitation Program. Since then it has broadened its scope to include naval personnel and dependents with various problems, marginal performers, and disciplinary offenders. CREDO is a spiritually based program intended to promote personal growth and maturity. AN informal survey in 1978 showed that CREDO was thought to be effective in assisting fleet personnel to manage their emotional, attitudinal, and Situational problems. Survey respondents also indicated the CREDO experience reduced the likelihood of premature separa- tion from the Navy. CREDO is staffed by Navy chaplains, and comes under the auspices of the Office of the Chief of Navy

Chaplains (OP-01H).

aa PURPOSE This thesis is the initial assessment of the impact on first-term enlisted attrition of RTC CREDO, and experimental program established by the Office of the Chief of Navy Chap- lains (OP-01H) at Recruit Training Center, Great Lakes, Illinois. RTC CREDO is an extension of the existing CREDO program and is intended to offer assistancé toward personal growth and spiritual development. As in the parent CREDO, the focus

of RTC CREDO is on the individual; on his emotional, attitudinal

WE

and situational adjustment. For the Navy, success of RTC

CREDO should be reflected in decreased behavioral and marginal

performance problems, and, consequently, decreased first-term

enlisted attrition. The evaluation of RTC CREDO will focus

on its impact on attrition of first-term enlisted personnel. RTC CREDO initiated its program in May, 1979, and con-

memamea through March, 1980, The 14 workshops of the RTC

CREDO were intended to offer spiritual and motivational assis-

tance as a means toward personal growth enhancement. This

thesis will not, however, evaluate the composition or struc-

ture of RTC CREDO's internal workshop format.~ This thesis

will describe the:

* selection process by which RTC CREDO experimental and control subjects were designated

* sources from which the subjects were obtained within the Recruit Training Center

* screening of the data from the RTC CREDO population ¢ experimental and control samples ¢ preliminary data analyses

¢ findings concerning attrition to date.

: A typical format of an RTC CREDO workshop is offered in Appendix B.

iG

II, EXPERIMENTAL DESIGN

Poul JECT SOURCES AND SELECTION PROCESSES 1. Apprentice Training Division

At its inception, RTC CREDO was to use exclusively recruit personnel who had satisfactorily completed the eight week basic military training curriculum and who were subse- guently assigned to the Apprentice Training Division (ATD). Apprentice training is required for all those personnel not assigned to technical school training. These individuals are referred to as General Detail (GenDet) personnel. For the first nine RTC CREDO workshops, experimental subjects were se- lected using subjective and judgmental criteria. Responsi- for selecting ATD personnel for RTC CREDO was given to a Senior Chief Petty Officer who, on the basis of reviewing the Recruit Training Records (also known as "hard cards") * OF personnel entering ATD, would select those he felt were most in need of RTC CREDO attention. Specifically, the Recruit Performance Data and Discipline/remediation sections of the hard card were reviewed. If these sections of the hard card showed marginal performance or minor disciplinary problems for an individual, he was selected for attendance at a RTC

CREDO workshop. Academic performance was not considered in

A sample “hard card" is found in Appendix C.

Le,

the selections. For the last five workshops, the procedure for selecting ATD attendees was entirely random from the men available at the time a RTC workshop was scheduled.

The procedures for the selection of appropriate con- trol group personnel were not uniform throughout the RTC CREDO program. Retrospective selection of control for six of the first nine workshops was made in October, 1979, by the same Senior Chief Petty Officer who had selected the experimental subjects (the reasons for omission of control group selection prior to this time will not be addressed in this thesis). Without appropriate controls, however, analysis of results of six of the first nine workshops would be inconclusive. With appropriately selected experimental and control groups, causal statements can be made concerning the impact of the Mee cREDO On attrition,

To obtain control groups, the Senior Chief reviewed mamemetD) records. To select a control group for the RTC CREDO workshop held in July 1979, for example, the records of recruit graduates who entered ATD in July 1979 were re- viewed using the same performance and disciplinary criteria which had been used to select the experimental group. Con- trol groups for three of the first nine workshops were not selected because appropriate personnel could not be identi- fied (The experimental groups corresponding to these groups were also dropped.) The selection procedure of individuals for control groups for the last five workshops was entirely random.

18

The names and social security numbers of all ATD-RTC CREDO personnel were submitted on individual ATD data sheets. The data sheets provided the primary documentation for experi- mental and control personnel. An example of the ATD data sheet is provided in Appendix D.

2. Positive Motivation Unit

The Positive Motivation Unit (PMU) provided inputs to RTC CREDO only for the last five workshops. PMU personnel are recruits still undergoing basic recruit training who have displayed performance and disciplinary deficiencies of such magnitude that they require special remedial attention. The PMU program stresses military discipline and physical fitness. Counseling by Human Resource Management personnel is provided. If a recruit does not successfully complete the PMU program, he is processed for discharge from the Navy. If he does suc- cessfully complete PMU, he is returned to basic training. If he fails again to meet minimum performance and disciplinary standards, he is processed for discharge from the Navy.

The PMU was charged with providing an unscreened list of twelve PMU recruits for each RTC CREDO workshop. Indivi- duals were then to be randomly selected into either the experi- metal or control group by RTC CREDO staff personnel. Doubt has been raised by the RTC CREDO staff concerning the appro- priateness of certain PMU personnel assigned to RTC CREDO for selection. This issue cannot be fully addressed in this thesis.

It would not be appropriate here to judge whether the PMU

ABS)

meate LOLllLoOwed the spirit, intent, and letter of their instruc- tions regarding personnel selection for RTC CREDO. It is suf- ficient to say that two PMU recruits were offered to RTC CREDO who were awaiting Recruit Evaluation Board (REB) action and

discharge from the service. This was not known prior to the

Bemgauet Of the workshop in which they participated. The doubt that persists is whether other PMU-RTC CREDO personnel were in @eemeerarc Status. If so, RTC CREDO attrition results from the

PMU group are biased.

B. DATA SCREENING

Of the 638 ATD and PMU personnel for whom data were sub- mitted for analysis, the records of 552 were used. The data from the first three RTC CREDO workshops, consisting of a total of 80 people, were not used in the analysis because appropri- @ee control groups could not be established.

Six PMU personnel were dropped from the sample. Four were experimental subjects in workshop six for whom no control subjects were available, The other two individuals (workshop 11) were those mentioned earlier who were awaiting REB action and discharge prior to attending the CREDO workshop.

Table II displays the composition of the entire sample by ATD/PMU and Experimental/Control divisions and by individual Perc CREDO workshops.”

4 Workshops were renumbered commencing with the fourth work- shop conducted, All workshop numbers referenced in the text fol- low this convention. Workshops originally numbered 1-3 were

dropped, as explained above, because no control group personnel were available.

20

SOURCE :

EXPERIMENTAL CONTROL San aan ee Workshop

dates

26-29 July 1979

30 = M0, - 20-23 Aug. 1979

24-27 Sept. 1979

ov

N

Ut

NO

U1 i

v NO \O ee) © h

15-18 @OCtreo 729

26-29 Nov. 1979

~J oO N) A NO is Oy

D7] Aes. eo?

PA—T 7 Ja 90 0

WORK SHOP NUMBER Abils Tele

Wn un fel

er er)

fo) oO

ATD/PMU Ns

Experimental/ Control Ns

Zag

Grand Total N

TABLE ITI fvertAL RTC CREDO PERSONNEL DISTRIBUTION AT GREAT LAKES

e Workshops 1-6 ATD persormmel selected vita personal record revtew. Workshops 7-11 ATD personnel selected randomly.

psal

For tracking of the experimental and control personnel, Defense Manpower Data Center (DMDC) monthly submission files were used. Initially, the DMDC cohort file was to be used, but it was found to be inappropriate for this thesis. Since the cohort file is compiled at the end of each fiscal year, personnel attending RTC CREDO workshops 7-11 would not have been recorded on the file as they had not entered the service meror tO the completion of FY 79 (30 Sep 1979). The DMDC submission file, which is compiled monthly, was considered to be more appropriate. The most current submission file avail- able was that of March, 1980, so it was potentially possible to track all 552 RTC CREDO personnel. However, of the 552 total, 28 personnel were not located in the DMDC files. This represents a loss of five percent of the total sample, and can probably be attributed to administrative inconsistencies in compiling data for submission to DMDC. Table III displays the final distribution of ATD/PMU and experimental/control

personnel for RTC CREDO.

Ze

WORKSHOP NUMBER

SOURCE:

EXPERIMENTAL CONTROL —_—_——— Workshop

dates

a) a = 26-29 July 1979 eT orn sce ee Bele [el neem 7 Za¢ 1) are) 26-29 Nov. 1979 SICeeSEoe

hie 238 (13) D5 (3 )et aaa) Zs)

Experimental/

Control Ns 263 (16) Zola)

Grand Total Ns Sel esi)

*(N) tndteates mtsstng cases

TABLE III FINAL RTC CREDO PERSONNEL DISTRIBUTION

A.

ITE. PROCEDURES AND RESULTS

PROCEDURES

Examination of the data was performed using SPSS programs.

Frequencies, cross-tablulations, discriminant analyses, and

multiple regression analyses were conducted using the follow-

TIC

seven variables:

HYEC - Highest Year of Education Completed PERC - AFQT Percentile Score

AGEE - Age Upon Service Entry

Marital Status

NDEP/DEP - Absence/Presence of Dependents Regular or Reserve Status

White/Non-white - Ethnic Grouping

A flow chart diagramming the cross-tabulation analyses is

presented in Figure l.

Descriptive statistics for each RTC CREDO workshop are

presented in Appendix E. Each workshop may be considered as

providing a separate cohort for future analysis, but for this

thesis the recency of the experiment precludes examination of

the

data at that level of disaggregation. The data have been

considered for analytical purposes as coming from one cohort,

although nine months separate the first and last workshops

under study. The only separation of the data into subsets

using time is the distinction between "early" and "late" groups.

24

OGaND OLN AOF sosATeuy ubtseq TeIUsUTAedxXy

TOULNOO

. IWiLNGWIddadxd

GIdWWS TWLOL

°T FHnold

ZS

The early group corresponds to RTC CREDO workshops one through Six and the late group includes workshops seven through eleven. This distinction reflects the two methods of selection used for RTC CREDO asSignment. The early workshops were selected through subjective review of Recruit Data Records (the hard card) and the late workshops were seleeted randomly from all recruits entering the ATD at the appropriate time. The PMU inputs to RTC CREDO all fell within the late group of workshops.

Only group descriptions and comparisons central the the evaluation of RTC CREDO are reported in this chapter of this thesis. The descriptions and comparisons included in this chapter describe the experimental RTC CREDO subjects and their control counterparts from both the ATD and PMU. These groups are also contrasted with one another in this chapter. Late subjects are compared to early subjects in order to de- termine any significant difference between the two groups re- sulting from the different procedures used in their selection. The comparisons presented in this chapter will help to deter- Mine the feasibility and type of analyses that can be used to evaluate the effects of RTC CREDO.

Group descriptions and comparisons that are not central to the evaluation of RTC CREDO are presented in Appendix F. The analyses shown in Appendix F allow comparisons to be

made, for instance, between subjects from the ATD and the PMU.

26

To assist the reader, the results of the comparisions provided in Appendix F are included in the summary of the results of

all analyses given in Table XVI of this chapter.

PDO SCRIPTIVE RESULTS

The means and distributions of the previously mentioned demographic variables for the entire RTC CREDO sample are given in Table IV. For highest year of education completed (HYEC) the mode is 12 years of education. Nearly 74 percent of the total cases found had 12 years of education. Only one person had more than this amount.

The reduced sample size for the AFQT percentile score variable (PERC) reflects the number of cases for which that variable had not been recorded on the DMDC submission file. The missing data might be due to the presence of prior ser- vice personnel in the sample because ASVAB scores are not recorded for these personnel when they enlist for a second time, This cannot be verified, however, because the DMDC submission file does not have an entry for prior service status. The age range of the sample supports the belief that some prior service veterans may indeed be included. This inference is reinforced by the finding that 14 people had from 25 to 106 months of active duty service recorded on their records as of the March 1980 submission file date.

The range for the number of dependents is given in Table

IV. For presentation purposes, the figures do not differentiate

27

il

VARIABLE MEAN N MODE RANGE Ey EC juin 85 524 12 8~14 PERC Ajmay 2 517 -- ~— AGEE | en 67 524 18 16-29 DISTRIBUTION Single/Married Sn) hi Ay 524 aeere NDEP/DEP 502/22 524 0-2 Regular/Reserve 370/154 524 --- White/Nonwhite 369/154 Se ---

LYEC Highest Year of Education Completed

PERC = ATQT Percentile Score AGEE = Age at Service Entry NDEP/DEP = No Dependents/dependents

TABLE IV DEMOGRAPHIC DESCRIPTION OF THE TOTAL SAMPLE

28

between individuals claiming one or two dependents. Several times during the presentation of the statistics, a discrepancy will be noted between the marital status and dependent status distributions. Whether this discrepancy is the result of administrative inconsistencies or single personnel claiming

to have dependents is unknown.

For purposes of analysis, the race variable was condensed