DOCUMENT RESUME

ED 415 614 EC 306 093

AUTHOR Larson, Sheryl A.; Lakin, K. CharlieTITLE A Longitudinal Study of Turnover among Newly Hired

Residential Direct Support Workers in Small Community HomesServing People with Developmental Disabilities: SummaryReport, Report 50.

INSTITUTION Minnesota Univ., Minneapolis. Inst. on CommunityIntegration.

SPONS AGENCY National Inst. on Disability and Rehabilitation Research(ED/OSERS), Washington, DC.

PUB DATE 1997-08-06NOTE 27p.

CONTRACT H133B0003-90; H133D50037-96; H133B80048; H133B0072-93AVAILABLE FROM Research and Training Center on Residential Services and

Community Living, Institute on Community Integration (UAP),University of Minnesota, 214 Pattee Hall, 150 PillsburyDrive SE, Minneapolis, MN 55455; telephone: 612-624-6024;e-mail: larso072@maroon.tc.umn.edu

PUB TYPE Reports Evaluative (142)EDRS PRICE MF01/PCO2 Plus Postage.DESCRIPTORS Agencies; *Attendants; Community Programs; *Developmental

Disabilities; *Employee Attitudes; Employment Experience;*Group Homes; *Labor Turnover; Longitudinal Studies;Recruitment; Residential Programs; Surveys

IDENTIFIERS Minnesota

ABSTRACTTwo studies examined turnover among newly hired residential

direct support workers in community homes serving people with developmentaldisabilities. In Study 1, 110 community homes for people with mentalretardation in Minnesota were surveyed for facility and residentialcharacteristics, direct support worker wages and benefits, recruitment andretention outcomes, factors associated with turnover, strategies to addressstaffing challenges, and supervisor suggested changes. Study 2 involved 5surveys of 124 newly hired direct-support workers at homes that had beenpreviously surveyed. Workers were surveyed regarding worker characteristics,direct support worker outcomes, reasons for wanting to leave, workers'recommendations to agencies, the most difficult aspects of direct supportwork, and workers' recommendations for potential direct support workers.Findings of the two studies indicated an average turnover rate of 46-48percent over two years. Factors affecting recruitment and retention outcomesincluded the length of time a particular home had been in operation, severityof residents' disabilities, and the tenure of the home supervisor.Recommendations to address recruitment and retention challenges are offeredfor the application process, for the period of worker entry into theorganization, for the period of organizational socialization, and for ongoingstrategies. (Contains 37 references.) (DB)

********************************************************************************* Reproductions supplied by EDRS are the best that can be made *

* from the original document. *

********************************************************************************

A Longitudinal Study of Turnover Among

Newly Hired Residential Direct Support Workers

in Small Community Homes

Serving People with Developmental Disabilities:

Summary Report

Report 50August 6, 1997

Research and Training Center on Residential Services and Community LivingInstitute on Community Integration (UAP)

GIIIIIJ

The College of Education& Human Development

UNIVERSITY OF MINNESOTA

BEST COPY AVAILABLE

U.S. D TMENT OF EDUCATIONOffice o ucational Research and Improvement

EDU IONAL RESOURCES INFORMATIONCENTER (ERIC)

his document has been reproduced asreceived from the person or organizationoriginating it.

Minor changes have been made toimprove reproduction quality.

Points of view or opinions stated in thisdocument do not necessarily representofficial OERI position or policy.

A Longitudinal Study of Turnover Among

Newly Hired Residential Direct Support Workers

in Small Community Homes

Serving People with Developmental Disabilities:

Summary Report

Report 50August 20, 1997

By

Sheryl A. Larson, Ph.D.K. Charlie Lakin, Ph.D.

Research and Training Center on Residential Services and Community LivingInstitute on Community Integration (UAP)

University of Minnesota214 Pattee Hall, 150 Pillsbury Drive SE

Minneapolis, MN 55455612-624-6024

larso072®maroon.tc.umn.edu

ACKNOWLEDGMENTS

Dr. Robert H. Bruininks, Dr. Susan Rose, Dr. Richard Arvey, Dr. Jennifer York, Dr.Richard Weatherman, and Dr. James Brown provided advice, support and encouragement in thedevelopment and completion of this project.

Janet Bast, Stacey Tytler-Moore, Erin Simond, and Lynda Anderson collected data,updated the project's data base, organized and recorded incoming surveys, typed and organizedall of the quantitative responses to surveys, and regularly contacted the homes in the study. AmyHewitt insisted that qualitative questions would be valuable, and she was correct. Nahoon Kwakdeveloped system files for this project and provided general statistical support, and CherylMorgan, Laura LaFrenz, Jennifer Sandlin, and Linda Schaffer provided timely secretarial support.

We thank the supervisors, direct support workers, advisors, colleagues and friends whohelped with instrument development and pilot testing. We also acknowledge the agencies,supervisors, and direct support workers who participated in the study itself. They poured a greatdeal of time and effort into this effort to learn more about how to improve support for directsupport workers. Finally, we thank Dr. Marci Bryan and the Association of ResidentialResources in Minnesota for endorsing this project.

The preparation of this manuscript was supported by subcontracts (#H133B0003-90 andH133D50037-96) between the National Institute on Disability and Rehabilitation Research(NIDRR), the Research and Training Center on Residential Services and Community Living,Institute on Community Integration (UAP), University of Minnesota; and the Research andTraining Center on Community Integration, Center on Human Policy at Syracuse University.Additional support was provided by grants from NEDRR to the RRTC at the University ofMinnesota (#H133B80048, and #H133B0072-93).

The recommended citation for this report is:

Larson, S.A. & Lakin, K.C. (1997). A longitudinal study of turnover among newly hired residential direct supportworkers in small community homes serving people with developmental disabilities: Summary report. Minneapolis:University of Minnesota, Center on Residential Services and Community Living, Institute on CommunityIntegration (UAP).

Additional copies of this report may be obtained from the Publications Office, Institute on Community Integration,109 Pattee Hall, 150 Pillsbury Drive SE, Minneapolis, MN 55455, (612) 624-4512. Please call for cost andordering information

The University of Minnesota is committed to the policy that all persons shall have equal access to its programs,facilities, and employment without regard to race, color, creed, religion, national origin, sex, age, marital status,disability, public assistance status, veteran status or sexual orientation.

2

BACKGROUND INFORMATION

There are at least 305,000 paid full-timeequivalent direct support positions ininstitutional and community residentialsettings for people with developmentaldisabilities in the United States. Nationally,in 1992 direct support workers earned anaverage wage of $5.97 per hour in privateresidential programs and $8.56 per hour inpublic residential settings (Braddock &Mitchell, 1992). National studies have foundannual turnover rates for direct supportworkers in community residential settingsranging from 34% in small publicly-operatedhomes to 70% for small privately-operatedhomes, with most estimates of turnover ratesin the 50%-70% range (Braddock &Mitchell, 1992; George & Baumeister, 1981;Lakin & Bruininks, 1981; Larson & Lakin,1992). With these turnover rates, at anational average cost per hire of $1,388 fornon-exempt workers (JWT SpecializedCommunications, 1996), the cost ofreplacement staff could be as high as $2.1 to$2.9 million per year.

While staff turnover may produce bothpositive and negative outcomes, whenturnover rates exceed 50% per year as theydo in community residential settings, theproblems presented outweigh the potentialbenefits. Confirmation that turnover ratesare cause for concern for communityresidential settings comes from manysources. High turnover of direct supportworkers a concern of consumers (Whiteman& Jaskulski, 1996), parents (Jaskulski &Whiteman, 1996; Larson & Lakin, 1991),community members (Governor's PlanningCouncil on Developmental Disabilities,1992), other direct support workers (Larson& Lakin, 1992), program administrators(Bruininks, Kudla, Wieck, & Hauber, 1980;

3

Larson, 1997), human services researchers(e.g., Braddock & Mitchell, 1992; Jacobson& Ackerman, 1989; Lakin & Bruininks,1981; Zaharia & Baumeister, 1978), andpolicy makers (e.g., Department ofEmployee Relations, 1989).

Compounding the difficulties created byhigh turnover rates are increasing difficultiesin recruiting replacement staff. Severalreports have identified high vacancy ratesand related recruitment problems forcommunity residential settings (Coleman &Craig, 1981; Jaskulski & Metzler, 1990;Larson & Lakin, 1992; Larson, 1997;Legislative Budget and Finance Committee,1989; Task Force on Human ResourcesDevelopment, 1989). For example, inMinnesota an estimated 11.9% of directsupport worker positions in residentialservice settings were vacant in August 1996(Larson, 1997). Residential agencyadministrators reported that the averageposition was vacant for 2.7 weeks beforebeing filled, and that when they advertisedvacancies only 7.3 applicants on averagereplied. Those administrators also reportedthat finding qualified staff members was theirmost difficult staffing challenge (71%reported it was a problem).

In recent years, finding direct supportworkers has become more difficult due toseveral demographic and labor markettrends. The Bureau of Labor Statisticsestimates that by the Year 2005, the numberhome health aides will increase by 138% to827,000, the number of human serviceworkers will increase 136% to 445,000, thenumber of teachers aides and assistants willincrease 4.1% to 1.3 million, and the numberof personal and home care aides will increase130% to 293,000 (Leftwich, 1994). In

addition, other service related industriesranging from telemarketing to hospitality aregrowing rapidly. For every ten newlycreated jobs in the United States, eight areprojected to be service-oriented. Howeverthe proportion of the U.S. population ages18-44 (those who have historically beenmost likely to be direct support workers) is

projected to drop from 42.4% to 37.5%between 1995 and 2005 (U.S. Bureau of theCensus, 1994). Clearly recruitment andretention of community residential directsupport workers is important both inresearch and in practice.

METHODOLOGY AND STUDY RESULTS

This study examines turnover amongnewly hired residential direct supportworkers through both facility level andindividual employee level analyses. In Study1, facility level analyses describeparticipating homes, examine turnover rates,and identify variables associated with facilitylevel turnover. In Study 2, individual levelanalyses describe individual staffcharacteristics and identify individual andjob-related factors associated with turnoverdecisions. This summary reviews themethodology and results for each of thesestudies and then presents a combineddiscussion. Appendix A contains tablessummarizing the results of both studies.

Study 1: Facility Analysis

Methodology

One hundred twenty-eight of 188agencies providing residential services topeople with mental retardation in Minnesotain 1993 were screened for participation inthis study. Of the 128 agencies screened, 94(73%) were eligible for the study based oneither being part of the MinnesotaLongitudinal Study (a deinstitutionalizationstudy also conducted by the University ofMinnesota), or providing 24 hour residentialservices to people with developmental

4

disabilities in at least one home with six orfewer residents. Administrators in 83 of the94 eligible agencies (88%) agreed toparticipate in the study.

Homes were selected for the study usinga two-phase process. In the first phase,supervisors of all State-Operated CommunityServices (SOCS) homes from participatingRegional Treatment Centers's, andsupervisors of all homes in the MinnesotaLongitudinal Study (see Hayden, DePaepe,Soulen & Polister, 1995) were invited toparticipate. In all, 110 homes were invitedto participate based on being part of theMinnesota Longitudinal Study or being aSOCS home. Of these homes, 54 that servedsix or fewer people are included in thisreport.

In the second phase, homes fromagencies not in the Minnesota LongitudinalStudy that provided 24 hour supports to sixor fewer people with developmentaldisabilities were considered eligible. Onehome was randomly selected from eachagency that agreed to participate in thestudy. In all, 56 homes were selectedthrough this process.

Supervisors in 143 of the 166 homesselected in phase one and phase two (86%)agreed to participate in the study. Of the110 homes included in this analysis, 54

homes were selected in the first phase and 56homes were selected in the second phase.

Two facility surveys were completed byparticipating supervisors between December1993 and December 1996. The first surveywas administered when the home entered thestudy. The second survey was administeredtwelve months after the first survey wasreturned in homes still in the study at thatpoint. The facility surveys requestedinformation about facility characteristics,staffing patterns, general staff characteristics,recruitment and retention challenges andcharacteristics of the people living in thehome. A short form of the facility surveywas used to gather basic information fromsupervisors unwilling or unable to completethe regular Time 1 or Time 2 facility survey.Supervisors in 143 homes competed the

Time 1 survey and supervisors in 101 of 108homes still in the study at Time 2 completedthe Time 2 survey (94%). Among the 110homes included in this analysis, Facility Time1 surveys were available for 110 homes,Facility Time 2 surveys were available for 80homes, and Facility Short Surveys wereavailable at Time 2 for 16 homes.

Results

Facility Characteristics. Of the 110small group homes included in this analysis,43 (39%) were funded by the MedicaidIntermediate Care Facilities for [personswith] Mental Retardation program, and 67(61%) were funded by the Medicaid Homeand Community Based Waiver program.The average home opened in 1990 andserved 4.7 people with developmentaldisabilities. The average cost per day perresident in these homes was $170 includingall room and board charges. Just over half ofthe homes (52%) were located in the

5

Minneapolis/St. Paul metropolitan area whilethe rest 'were in out-state Minnesota.

Resident Characteristics. The peopleliving in the participating homes could becharacterized several ways. Most had severeor profound mental retardation (64%) andmore than half (55%) had moved from astate institution to this home directly. Morethan half had a specific intervention programto address challenging behavior, and onefourth (26%) had a formal diagnosis ofmental illness in addition to mentalretardation.

Direct Support Worker Wages andBenefits. Direct support workers earned, onaverage $7.07 per hour starting wages, withthe highest wage in each home averaging$9.27. Of all direct support workers in thesehomes, 43% were considered to be full-time,58% were eligible for medical or dentalbenefits, and 72% were eligible for paidleave time (holidays, vacations, paid leave).

Recruitment and Retention Outcomes.Supervisors reported that recruitingqualified workers was the most commonstaffing problem (reported by 57% ofsupervisors), followed by staff turnover(44%), and staff motivation (37%). Annualturnover rates averaged 46% for directsupport workers and 27% for supervisors.Among the direct support workers who leftthe home during a 12 month period, 45% leftwithin six months of hire, and another 23%left between six and twelve months afterhire. Only 32% of those who left had morethan 12 months tenure in the home.Turnover and difficulties caused by turnoverwere measured twice in each home. Whilethe average turnover rate did not change

significantly from Year 1 to Year 2 (46%versus 48%), the difficulty caused byturnover increased significantly during thatperiod (F = 4.19,p < .05) from 2.8 to 2.6 (1= very much, 4 = very little).

Factors Associated with Turnover.While the rate at which workers left theorganization did not differ over time orbased on the type of facility, several otherfactors were associated with facility levelturnover rates. Variables with significantcorrelations with turnover at Time 1 includedthe population of the county in which thehome was located (r = .20, p < .05),supervisor tenure in the home (r = -.23,p <.05), the proportion of direct supportworkers eligible for paid leave (r = -.34, p <.01), and the number of direct supportworkers promoted in the previous year (r =.34, p <.01). Turnover at Time 2 wassignificantly correlated with starting pay fordirect support workers at Time 1 (r = -.28, p< .01), and supervisor tenure in the home atTime 1 (r = -.24, p < .05).

Two blocks of variables were tested todetermine their contribution to explaining thevariability in turnover rates. The first blockincluded context, facility and residentcharacteristics that could affect turnover.The second block included staffing patternsand strategies. A multiple regressionanalysis accounted for 34% of the variability(26% adjusted) in turnover rates at Time 1using variables including: unemploymentrate, county population, average cost perresident, years home was open, ICF-MRstatus, resident case mix score, starting payfor direct support workers, whether live-instaff were used, the tenure of the supervisorin the home, and the percent of directsupport workers eligible for paid leave (F =4.07, p <.001). Unique contributions to

6

explaining the variability in turnover rates atTime 1 were made by resident supportneeds, starting pay for direct supportworkers, supervisor tenure in the home, andthe proportion of direct support workerseligible for paid leave.

Strategies to Address StaffingChallenges. Supervisors were asked to rankthe importance of strategies they used toaddress staffing problems in their homes byidentifying the five most important strategiesthey used. The most frequently identifiedstrategies included encouraging team workamong staff members (80% of all homes),managing fairly/treating staff members fairly(65%), communicating clear, understandableprogram objectives and agency philosophy(61%), establishing effective communicationamong staff members (45%), and using clearand understandable job roles andresponsibilities (34%). Providing realisticjob information was a priority strategy foronly 25% of the supervisors.

An exploratory analysis was conductedto learn whether the strategies consideredmost important for addressing recruitmentand retention challenges were associatedwith actual turnover rates in the homes atTime 1 and Time 2. Differences in turnoverrates between homes in which thesupervisors did or did not select each of themost popular strategies were tested usingOne-way Analysis of Variance. The use ofdirect observation to provide realisticinformation to recruits, and the selection ofproviding realistic information about the jobto applicants as an important strategy werealso tested. Of the top managementpractices identified by supervisors, only onewas related to turnover. Supervisors whovalued managing in a fair manner /treatingworkers fairly worked in homes with

significantly lower turnover rates (40%) thansupervisors who did not select fairness as apriority strategy (56%) (F = 6.16, p < .05).

Interestingly, homes that used directobservation RJPs and homes that reportedproviding realistic information to recruits asan important technique had significantlyhigher turnover rates than the other homes(51% vs 38% for homes using directobservations vs homes that did not, F = 3.84,p < .05; and 61% vs 40% for supervisorsreporting realistic information as animportant management strategy versussupervisors that did not, F = 7.54, n < .01).It could be that supervisors in homes withhigher turnover rates were more focused onthe recruitment process and directobservations because recent leavers said thejob was not what they expected, or it couldbe that an emphasis on realistic job previewsis associated with less focus on strategiesassociated with lower turnover rates.

Changes Impacting Outcomes.Supervisors were asked to describe anychanges that had taken place in the agency orin their house that might have had an impacton staffing outcomes. Of the supervisorswho identified specific changes, the mostcommon changes reported were a newsupervisor was introduced into the house(12% of all homes), the house had staffrecruitment or retention problems such as alack of qualified applicants (10% of homes),the support needs of residents changed (9%),and hiring or recruitment practices such aschanging advertising strategies, usingrealistic job previews, or changing the personresponsible for hiring new workers changed(8%). Other changes mentioned includedimprovements in staff training, improvementsin management practices, changes in wagesand benefits, and agency expansion.

7

Factors Related to Turnover(Supervisor Reports). Supervisorsidentified several factors that they feltinfluenced staffing outcomes. By far themost common factor was wages and benefitsfor workers (reported by 32% of allsupervisors). Other important factorsincluded flexible or fluctuating hours (14%),picblems with team work or workerparticipation (14%), having mature,dependable workers (13%), providing goodtraining (13%), providing consistent,effective communication for workers (12%),providing a fun or positive workenvironment (11%) and using innovatingrecruitment practices (11%). Other factorsmentioned by more than one supervisorincluded the skills and characteristics ofresidents, fair treatment of employees,support and recognition for workers,supervisor training, qualifications, and style,clear expectations, and agency practices suchas opportunities for advancement, location,and retention experience.

Supervisor Suggested Changes.Supervisors suggested ways the agencycould make their job better. By far the mostcommon response was to provide more orbetter training for supervisors (reported by21% of all supervisors). Other commonresponses included improve agencycommunication (13%), use supportivemanagement practices (12%), and improvewages and benefits for workers andsupervisors (9%). Other responses offeredby more than one supervisor included helpwith time management, hire additionalsupport stafC support supervisor decisionmaking, reduce documentation requirements,address funding and budgeting issues,improve training for direct support workers,

provide incentives for workers, and improvework space for supervisors

Study 2: Study of Newly Hired DirectSupport Workers

Methodology

Five surveys were administered to up tothree newly hired direct support workers inparticipating homes supporting six or fewerresidents. The first survey was completed athire and gathered information about personalcharacteristics, education and experience, jobexpectations, employment contextinformation (such as quality of other joboffers), and job characteristics. The secondsurvey was completed 30 days after hire andgathered information about jobcharacteristics (such as hours worked andsalary), work-related characteristics(organizational commitment and jobsatisfaction), supervisor characteristics,employment context, training needs andopen-ended information about the job andhow it could be improved. Twosupplementary surveys were completed atthe time of the second survey. One was theLeader Behavior Description Questionnaire,and the other was the OrganizationalSocialization Survey.

The third and fourth surveys gatheredupdated information about job characteristics(such as hours worked and salary), work-related characteristics (organizationalcommitment and job satisfaction), supervisorcharacteristics, employment context, trainingneeds and open-ended information about thejob and how it could be improved. The thirdsurvey was administered after six months onthe job and the fourth was administered after12 months on the job. The final surveyrequested information from exiting

8

employees about the person's leaving (e.g.,was it voluntary, where did they go), andabout the good and bad aspects of the job.These surveys were administered in writingand were returned in sealed envelopesdirectly to the investigator. Supervisorscompleted an exit survey for direct supportparticipants who left their position during thestudy.

A total of 174 of 333 direct supportworkers who were invited to participate inthe study agreed (52%). Of those, 124 wereincluded in this analysis. Those included inthis analysis were new to the home and tothe agency at the time they started. Alldirect support worker participants workedregularly scheduled shifts (as opposed to on-call) at the home, and worked in homessupporting six or fewer residents.

Results

Direct Support WorkerCharacteristics. Newly hired workers inthis study were predominantly female (81%)and were an average of 28.8 years old athire. Newly hired workers wereoverwhelmingly white (96%), unmarried(75%), with no financial dependents (76%).New hires had an average of 1.9 years ofexperience in developmental disabilities andhad completed one and a half years of post-secondary education at hire. More than half(51%) of new hires had never worked indevelopmental disabilities prior to taking thisjob. Approximately one-third (37%) of allnew hires had taken a course on mentalretardation, and 20% were currently enrolledin a post-secondary educational program.

Direct Support Worker Outcomes. Ofthe 124 new hires in this study, 33%

10

remained stayed in the same position for 12months, 3% were promoted, 11% moved toanother home in the agency, 38% leftvoluntarily and 15% were terminated within12 months of hire. An analysis of 58workers who stayed 12 months and 47workers who left revealed that stayers weresignificantly more likely than leavers to haveheard about the job from inside sources(such as current or past staff members,family members or friends) than outsidesources (such as newspaper advertisementsor employment agencies). Stayers were alsosignificantly more likely to think they had achance for a promotion than were leavers.Stayers had significantly lower intent toleave, higher organizational commitment,fewer unmet expectations, and higherextrinsic satisfaction than leavers. Overall,26% of the variability in staying or leavingwas accounted for by organizationalcommitment, met expectations, alternativejob opportunities, current salary, jobsatisfaction, supervisor structure, recruitmentsource, hours per week, months in field, ageat hire, and intent to leave.

Reasons for Wanting to Leave.Workers identified several types of incidentsthat made them want to leave the home orquit the job. More than half of the workers(51%) said there were no incidents that madethem want to leave, and another 23% did notrespond to the question. Of those who dididentify issues, the most common incidentswere problems with co-workers such as stafftalking behind each other's back (17%),inadequate pay, benefits, or incentives(16%), problems with supervisors (13%),and scheduling problems (13%).

Recommendations for Agencies. Whennewly hired direct support workers were

9

asked what agencies could do to make theirjob better, the most common response wasto increase or improve pay, benefits or otherincentives (37%). Another 33% of workerssaid no changes were needed because theemployer or supervisor was doing a good jobalready. Other common suggestionsincluded asking the supervisor to be morepersonable and attentive and to do a betterjob managing the home (17%), or to give theworkers more, better or different hours(17%).

Most Difficult Aspects of DirectSupport Work. New workers alsoidentified the hardest part of starting theirjob. Workers reported they had difficultygetting to know the people in the home andtheir behaviors and traits. This response wasreported by 45% of all new hires. Anothercommon response, reported by 43% of newhires, was that learning the routines andduties was difficult. A smaller, but stillsubstantial minority of new workers reportedhaving difficulty getting to know and getalong with other staff members (20%) andadjusting to the work schedule, particularlyfor those who worked overnight or earlymorning shifts (14%).

What Potential Direct SupportWorkers Should Know. The final questionasked of new workers was what he or shewould tell their best friend if the friend werethinking about applying for a job at thishome. Workers mentioned several positiveand challenging job features. Among thechallenging job characteristics mentionedwere challenging behavioral or medical needs(25% of workers), the need to work varyinghours that included weekends and evenings(25%), and difficult or different duties such

BEST COPY AVAILABLE

as cooking, giving medications, andproviding transportation in the companyvehicle (23%). The most common positivecharacteristics mentioned included that thejob was rewarding (19%), the workenvironment was good (19%), you have tobe responsible and mature but can have fun

(17%), and you will need lots of patience(17%). Study participants also advisedpotential workers to learn what duties areinvolved, be responsible and mature, bepatient, and treat each person with dignityand respect.

DISCUSSION

The direct support worker turnover ratesin 110 small group homes for people withdevelopmental disabilities in Minnesota,averaging 46-48% over two years, wereslightly lower than the reported nationalaverages of between 50% to 70% annually(Larson, Hewitt & Lakin, 1994). They weremuch lower than the crude separation rate of67% reported for 25 randomly selectedMinnesota community residences in 1992(Braddock & Mitchell, 1992). But by thestandards of virtually any industry, the ratesare still very high, and they remain at levelsthat preclude adequately stable directsupport for persons in small residentialsettings. In addition, many new hires werefired (15%). This termination rate remainsidentical to the one first reported forcommunity residents nearly a generation ago(Lakin, 1981). Furthermore, recruitmentproblems, which were mentioned as "a"problem in past research, were viewed bythese supervisors as "the" problem.Supervisors reported that difficulty findingnew workers was their number one agencymanagement problem.

This study identified several factors thatmake a difference in recruitment andretention outcomes for small group homes.One factor was the length a particular homehas been in operation. For example, ICF-MR certified homes in this study, whichopened much earlier than the HCBS Waiver

10

funded homes had significantly less turnoveramong direct support workers within thefirst six months of employment, hadsignificantly higher average tenure and hiredsignificantly fewer new workers in a one yearperiod. In addition, the average tenure ofworkers increased significantly overallbetween Time 1 and Time 2 in both types ofsettings. Both findings are consistent withprevious studies, which identified length ofoperation as an important factor influencingturnover rates (e.g., Lakin, 1981).

It takes time for an agency to recruit andtrain a stable cadre of workers for a newhome. Since the number of small grouphomes nationally continues to expand rapidly(from 41,826 homes in 1992 to 78,365homes in 1996; Prouty & Lakin, 1997),turnover challenges associated with openingnew group homes and building a cadre ofstable direct support workers are likely tocontinue as community supports continue toexpand. Consequently, as new smallcommunity residential options are developed,it continues to be important to identify andimplement strategies to reduce turnover,especially during the first few years ofoperation of a new program. For individualorganizations, pacing new developments maybe important to avoid experiencing such"growing pains" at large numbers of anagency's sites. Spacing the development ofnew services appears particularly important

12

in areas where unemployment is low, otherjob opportunities are high, and where wagesand benefits in other service industries arehighly competitive. Other strategies toreduce the effects of initially high turnoverrates on agencies and the people theysupport suggested by this research includeusing some of the agency's established coreof long-term employees from other sites(both supervisors and direct supportworkers) in new sites, increasing theproportion of positions offering full-timehours and benefits, and integrating acomprehensive program of recruitment andretention strategies into an agency'spersonnel practices. Elements of suchcomprehensive programs are identified in thisreport.

Resident characteristics appearedassociated with direct support workerturnover when both were measured at thesame time. Homes that served individualswith more challenging needs (in level ofmental retardation, challenging behavior,mental health issues, or assistance withactivities of daily living) tended to havehigher turnover rates. As community homesare planned for people with more substantialsupport needs, particular attention should bepaid to factors that minimize the turnover ofworkers. But personnel practices thatadequately support direct support workersare needed irrespective of the needs of thepeople they support.

The tenure of supervisors in the home isa third factor related to turnover andretention of staff. Homes that had newersupervisors had higher turnover rates andlower average tenure than homes with moretenured supervisors. This is partiallyexplained in that the maximum time asupervisor could have been in many homeswas limited because many homes had opened

11

within the last five years. However,supervisor tenure maintained its associationwith direct support worker turnover evenafter taking into consideration how long thehome had been open. When supervisorswere asked to identify factors that influenceddirect support worker retention, the mostcommonly mentioned was supervisorturnover. In the facilities surveyed, theturnover rate for supervisors was 27% over a12-month period.

The role of supervisors in affectingretention of direct support workers appearedvery important in this study. Both stayingand leaving direct support workers identifiedtheir supervisors as a key factor in leaving orwanting to leave the agency. Fairmanagement practices were the second mostcommon strategy identified by supervisors toaddress staffing problems. Direct supportworkers reported that having a competentsupervisor was a very important expectationwhen they started their new jobs. Turnoverrates were significantly lower in homeswhere the supervisors considered managingfairly to be one of their top five managementpractices. When supervisors were askedhow the agency could help them do a betterjob they requested training, improvedcommunication, fair management practicesby the agency, and support from the agencyfor staffing and recruitment issues. Indeveloping interventions to address directsupport worker recruitment and retention,the tenure, skills, and performance ofsupervisors were all importantconsiderations.

It is, of course, impossible to overlookthe importance of pay, benefits, paid leaveand promotional opportunities for therecruitment and retention of direct supportworkers. Not only did starting pay accountfor a significant portion of the variability in

turnover rates at the agency level, but stayerswere significantly more likely to report theythought they could get a promotion thanleavers. Both direct support workers andtheir supervisors identified pay as a topfactor influencing recruitment, retention, andplans to stay in the home. The availability ofother jobs with better pay, workingconditions or other conditions ofemployment was a significant predictor ofwhether direct support workers would stayor leave. Improving salaries, promotionalopportunities and benefits for direct supportworkers is fundamental to increasing thestability of directs support workers.

The relationship between direct supportworkers and their colleagues emerged as animportant issue in the qualitative data

collection. Both supervisors and directsupport workers reported that a sense ofteam work and positive relationships amongdirect support workers were important tooverall staff retention. Among supervisors,it was tied for second among the factorsviewed as influencing successful recruitment,retention and training. Stayers identifiedproblems with coworkers as the mostcommon type of incident that made themwant to.leave. Leavers also reported thatproblems with coworkers had influencedtheir decision to stay or leave. Severalworkers complained that their coworkersgossiped about them, that competitionbetween workers in different shifts was aproblem, or that poor performing coworkersmade their job more difficult.

RECOMMENDATIONS

Based on the results of this study and areview of 1,000 other studies, strategies toaddress recruitment and retention challengeswere reviewed and discussed. Suggestedstrategies included selection and recruitmentchanges, orientation and socializationpractices, mentoring and training programs,and ongoing strategies such as enhancing thestatus of workers, training supervisors, andevaluating recruitment and retentionoutcomes (see Table 1). This summarybriefly reviews strategies in each of thesecategories. A complete literature review andextended discussion of these strategies canbe found in the full-length report (Larson &Lakin, 1997).

12

Table 1Strategies to Address Recruitment andRetention ChallengesDuring the Application Process

Selection strategiesRecruitment sourcesRecruitment strategies: Realistic Job

PreviewsDuring Organizational Entry

Orientation strategies: Structuredorientations

Initial socialization: Realistic orientationprogramsDuring Organizational Socialization

Peer mentoringCompetency-based staff training

Ongoing StrategiesEnhancing the status of and opportunities

available to direct support workersDeveloping links with higher educationStaff development for supervisorsOngoing internal evaluation of efforts

14

Strategies for the Application Period

Selection Strategies. Several strategiescan be implemented before a worker is hiredto improve recruitment and retentionexperiences. Selection is the process used byorganizations to improve matches betweenemployee skills and organizational jobrequirements (Wanous, 1992). In this study,the rate of firings for new hires of 15% bythe end of one year is an indication thatimprovements in selection practices could behelpful. Potential employees need to knowspecifically the criteria for disqualification foremployment so that they do not invest theirtime applying for a job they will not beallowed to keep. Selection practices need toexclude workers who will be subsequentlydisqualified from employment because oftheir background check. Additional expensecould be saved by recruiting new workers inanticipation of openings so that thebackground checks could be completed bythe time the person is needed. This strategyhas the additional benefit of reducingovertime costs incurred when positions arevacant for long periods.

Another selection strategy that may behelpful in helping recruits to assess their ownsuitability for a job and in assessing therecruit are structured interviews. Structuredinterviews are based on a thorough jobanalysis that specifies critical factors insuccess on the job. One type of structuredinterview, patterned behavior descriptioninterviews, uses critical incidentmethodology to identify important jobdimensions and to develop a scoring guidebased on examples that illustrate the range ofexcellent to poor performance. Then sets ofquestions are developed to probe theapplicant's background and experience forspecific examples of how that person handled

13

situations similar to those he or she mightface in the position that is open (Wanous,1992). Once a specific example has beenidentified, the interviewer probes to gatherthe details about the situation, how theperson responded, and the outcome.Responses are scored with guidelinesdeveloped using the critical incidentreporting technique. Structured interviewsmay be especially helpful in identifying how arecruit handles conflict with coworkers orsupervisors.

Recruitment Sources. The pool ofpotential applicants for direct support workis not growing rapidly enough to provide anadequate supply of qualified workers. Moreeffective recruitment efforts are clearlyneeded. Possible strategies includedeveloping a volunteer program for studentsto introduce them to human services work;developing consortia of service providers ina geographic area to join recruitment effortsso that the field becomes more visible in thecommunity; and developing specificrecruitment materials such as brochures orvideotapes that could be viewed by targetedpools of potential recruits in high school andcollege classes, job centers, employmentagencies and community centers (e.g.,Hewitt & Larson, 1996); and developingpublic service announcements.

Recruitment incentives can also behelpful. Using incentive programs that paybonuses when a new hire finishes apredetermined number of months on the jobor a per recruit signing bonus for currentworkers can increase recruitment from insidesources. Besides increasing the number ofrecruits, incentive programs involvingrecruitment by current employees have theadded benefit of recruiting people who were

15

more likely to stay for at least 12 months(i.e., those recruited from "inside sources").

Recruitment Strategies: Realistic JobPreviews. Realistic job previews are arefined recruitment technique that assist inrecruiting people who will stay on the joband perform it with personal satisfactionbecause they have a "realistic" impression ofthe job before they accept it. Developingand instituting a realistic job previewinvolves several components including:developing a good concise job description,examining turnover history in thehome/agency, gathering house-specificinformation about the organizational climateincluding positive and negative jobcharacteristics, summarizing the information,identifying a format to present theinformation to applicants, and implementingand evaluating the effectiveness of the RJP.The challenges for supervisors in small grouphomes are to systematically identifyinformation not already being effectivelycommunicated, to incorporate new methodsto distribute that information, and to evaluatethe effectiveness of their efforts.Presentation formats for RJP's includeinviting prospective employees to a meal orrecreational activity at the home, showinginterviews of consumers, parents and staffmembers about the home, showing avideotape of the regular house routines,scripted oral presentations, tours or worksamples when people are home, andvolunteer or internship opportunities

This study provided information that canbe used to guide the development of RJP's.For example, direct support workersreported that the things that made them wantto stay were the people in the home - boththe residents and their coworkers, and therewards of being needed. Some employees

14

found the ability to tailor hours to their needsa positive aspect of the work. Othersappreciated being a valued member of ateam. Among the challenging aspects of thejob for some workers were physicallydemanding resident behavior, low pay andinadequate benefits, problems with theircoworkers and supervisor, and limitedopportunities for advancement. Providingaccurate information about these issues tonew recruits early in the application processis an essential part of the recruitmentprocess.

Strategies for Organizational Entry

Orientation Strategies. The mostdifficult job components for new workers inthis study as they started their jobs werebecoming acquainted with the residents,learning the routines, developingrelationships with coworkers, rememberingtraining information, and adjusting to theschedule. Many expressed concern aboutfulfilling the substantial responsibilities givento them. The experience of entering a neworganization is stressful for all workers, butis made more so when responsibility is highand direct expert support and supervision islimited as is increasingly the case in smallcommunity service settings. Agencies canhelp by communicating that the strugglesfacing the newcomer are typical and byproviding specific suggestions about how tohandle the stress they may experience. Asuccessful orientation program will reducethe anxiety of new employees and make themfeel a part of the organization, promotepositive attitudes toward the job and theorganization, establish open communicationbetween the organization and the employee,communicate the expectations theorganization has regarding performance and

1i

behavior, acquaint new employees withorganizational background, goals,philosophies, management styles, structure,products and services, and presentinformation on organizational policies,procedures, compensation practices, andbenefits (Holland & George, 1986).Providing planned opportunities for newworkers to get to know other workers andthe people they will be supporting before thefirst solo shift can be helpful in theorientation process. Pacing the informationprovided during orientation can also help toreduce the likelihood that a new worker willbecome overwhelmed with the information.

Initial Socialization.Deinstitutionalization has led to widespreaddecentralintion of services and supports.Workers who previously would have hadmany coworkers at the same site now may bethe only worker at the site at certain times ormay have only one co-worker. However,this shift has produced new demands,challenges, and stressors for direct supportworkers. It is a particular challenge todevelop strategies to support workers whoare in scattered sites. People need to knowhow to get help and to feel confident that thehelp they need will be available. Providingsuch critical information and comforts is animportant part of the initial socializationprocess. Some agencies enhance initialsocialization efforts by introducing newworkers to all of the homes in a geographicarea so that there is always someone to callfor advice or assistance.

The findings of this study also suggestthat it is important during the initialsocialization period to provide team buildingopportunities so that newcomers can feelintegrated into the social environment of thehome and agency. Considering that workers

in some homes enjoyed their colleagueswhile others wanted to leave because of theircoworkers, attention to team building anddealing with differences and disagreementsamong coworkers is important at thebeginning of employment and on an ongoingbasis.

Strategies for OrganizationalSocialization

Peer Mentoring. Peer mentoringstrategies link new employees with moresenior direct support workers to help insocialization to the job, and in developing orpracticing specific skills needed for the job(Hewitt, Larson, Ebenstein & Rose, 1996).Peer mentoring can reduce isolation of directsupport workers and increase supports. Itcan also allow supervisors to delegate thetasks of answering routine questions aboutthe job. Peer mentors benefit by developinga broader understanding of the work and oftheir co-workers.

Competency-Based Training. Trainingof new direct support workers is importantbecause it is a direct regulatory mandate formost human services agencies (Larson,Hewitt & Lakin, 1994), it is considered a keyelement in achieving higher quality services(Alpha Group, 1990; Fiorelli, Margolis,Heverly, Rothchild & Krasting, 1982), itprovides the opportunity to learn critical jobfunctions, develop new skills, and cope withjob roles (Camp, Blanchard & Huszczo,1986), and it develops attitudes and skillsamong employees that affect the quality oflife for individuals with developmentaldisabilities (Jones, Blunden, Coles, Evens &Porterfield, 1981). A study of 1,736 newhires in different organizations suggested

15

17

that workers who complete more weeks oftraining (Mean = 4.5 weeks of training forworkers who stayed more than sevenmonths) had significantly lower turnoverthan workers in agencies with fewer weeksof training (Mean = 1.9 weeks of training forworkers who voluntarily resigned) (Wanous,Stumpf & Bedrosian, 1979). Voluntaryleavers also were less likely to receiveinformal job training (19.6% for leaversversus 43.1% for stayers).

Ongoing Strategies

Enhancing the Status of andOpportunities Available to DirectSupport Workers. Some efforts to enhancethe status of and opportunities available todirect support workers can be made at theagency level through reorganization thatflattens agency hierarchy, throughrestructured wage packages that offer atleast pro-rated paid leave time for allworkers, and through flexible paid leave timeand benefits policies that allow workers touse those benefits as needed in their ownparticular circumstances. Other effortsrequire systemic change. New employmentbenefits may be needed. Some of thesecould be developed in conjunction withpublic agencies. These might include tuitioncredits at public colleges, universities andtechnical schools. Alternatively, tax creditscan be developed to allow retirees on SocialSecurity to benefit from employment insupporting people with disabilities. Evenwith these changes, however, public attitudesabout the value of direct support workers asexpressed in public policies about funding forresidential services must change if large scaleimprovements are to be made.

Another way the status of direct supportworkers can be improved is through

16

providing staff development and careeradvancement opportunities. In this study,stayers were significantly more likely to thinkpromotional opportunities were available tothem than were leavers. Providing employeebonuses for skill development, promotingworkers from within, providing educationalbenefits, and developing career ladders areall important to improving the stability andquality of the direct support workforce.

Developing Links with HigherEducation. The Kennedy FellowsMentoring Program, developed by John F.Kennedy Jr., provides scholarships andcareer mentoring to direct support workersenrolled for at least six credits in two NewYork colleges (Hewitt, Larson & Ebenstein,1996). This program encourages directsupport workers to complete a two-yeardegree, and helps them become eligible forpromotions to positions with greaterresponsibility. Fifteen other states havefederally funded Training Initiative Projectsdesigned to identify, develop and disseminatestate-of-the-art training curriculums, providetechnical assistance and training to directsupport workers, supervisors, agencies,consumers and families, develop careerladder opportunities for direct supportworkers, complete training needs-assessments, and facilitate collaborationamong key stakeholders regarding directsupport issues (AAUAP, 1996). Theseprojects reflect the growing national interestin and concern for developing, respectingand supporting direct support workers.They also reflect examples of programs thatcan increase the visibility of careers in directsupport work, and can assist in recruitingnew workers to the industry.

1 8

Staff Development for Supervisors.This study demonstrated an associationbetween supervisor behavior and recruitmentand retention outcomes. Homes with lesstenured supervisors had significantly higherturnover rates. Addressing turnover amongdirect support workers may well begin withsuccess in increasing stability amongsupervisors of the settings in which theywork. Common practices such as rotatingsupervisors through settings may bedetrimental to stability.

Staff development for supervisors is alsoimportant. Direct support workers identifiedhaving a good supervisor as an importantissue, and reported that problems withsupervision influenced their decisions to stayor leave. Supervisors also requestedassistance. Developing training forsupervisors on recruitment and retentionstrategies, and leadership and supervisionskills are important strategies to addressrecruitment and retention issues.

Ongoing Internal Evaluation ofEfforts. It is not sufficient for an agency tohave a general idea that they have a problemwith recruitment or retention. Agencies needmany different types of information tomonitor recruitment and retention outcomesand to design effective interventionstrategies. The components of a workplaceassessment include: developing an accuratejob description, examining retentionoutcomes and recruitment practices,gathering specific information about positiveand negative job features, describing anychanges or special incentives that may haveinfluenced recruitment or retention, andsummarizing the information gathered.More detailed information about developingand using workplace assessments areavailable from the author.

CONCLUSIONS

This society has made a clearcommitment to the presence andparticipation of people with developmentaldisabilities in its communities, schools, andwork places. That commitment is injeopardy. Demographic shifts depleting thenumbers of young adults, economic growthresulting in increasing jobs and wages, andhuman service expansion and other factorsare making it increasingly difficult tomaintain current levels of staff much less toexpand the number of staff available to meetneeds stemming from future growth. Thereis a crisis in the community that derives fromwhat has been inadequate attention to theintractable connection between communityliving for people with disabilities and

17

S

community supports provided by directsupport workers.

The problems of recruiting and retainingdirect support workers will continue todemand concerted and creative efforts bypublic officials, advocates, service providers,and others who care about the well-being ofpersons with developmental disabilities.Areas of particular focus include increasedamounts and attractive options incompensation, more comprehensive andmore effective recruitment initiatives,improved quality, recognition andtransferability of training, expanded careeropportunities, more effective supervision,better matching of employees to work roles,and more effective team building. Success in

these efforts are among the most importantcomponents to assuring that communityliving is a real and viable option for allAmericans with developmental disabilities.

We must not encourage the respect anddignity of one group of people (those withdevelopmental disabilities) at the expense ofanother group of people (those paid to

support them). Americans talk often of theimportance of having good teachers for ourchildren. We value those who would helpchildren to become productive citizens ofthis nation. We must also respect those whohelp and support people with mentalretardation and other developmentaldisabilities.

REFERENCES

Alpha Group (1990). Report to thecommissioner of the Minnesota Department ofHuman Services by the task force on thecompensation and training of direct careemployees. St. Paul: Minnesota Department ofHuman Services Task Force on Compensationand Training of Direct Care Staff.

American Association of University AffiliatedPrograms for Persons with DevelopmentalDisabilities (1996). Community servicestraining initiatives. Silver Springs, MD:AAUAP.

Braddock, D., & Mitchell, D. (1992).Residential services for persons withdevelopmental disabilities in the United States:A national study of staff compensation, turnoverand related issues. Washington, D.C: AmericanAssociation on Mental Retardation.

Bruininks, RH., Kudla, M.J., Wieck, C.A., &Hauber, F.A. (1980). Management problems incommunity residential facilities. MentalRetardation, 18, 125-130.

Camp, RR., Blanchard, N.P., & Huszczo, G.E.(1986). Toward a more organizationallyeffective training structure and practice.Engelwood Cliffs, NJ: Prentice Hall.

Coleman, T.E., & Craig, C. (1981). Thecommunity personnel study: Turnover issues inmental retardation community programs.

18

Boston, MA: Massachusetts Department ofMental Health, Division of Mental Retardation.Department of Employee Relations (1989).Study of employee wages, benefits and turnoverin Minnesota direct care facilities servingpersons with developmental disabilities. St.Paul, MN: State of Minnesota, Department ofEmployee Relations for the Department ofHuman Services.

Fiorelli, J.S., Margolis, H., Heverly, M.A.,Rothchild, E., & Krasting, D.J. III. (1982).Training resident advisors to provide communityresidential services: A university-based program.Journal of the Association for Persons withSevere Handicaps, 7, 13-19.

George, M.J., & Baumeister, A.A. (1981).Employee withdrawal and job satisfaction incommunity residential facilities for mentallyretarded persons. American Journal ofMentalDeficiency, 85, 639-647.

Governor's Planning Council on DevelopmentalDisabilities (1992). Minnesotans speak out: Asummary of town meetings held throughoutMinnesota on developmental disabilities issues.St. Paul: Author.

Hayden, M.F., DePaepe, P., Soulen, T., &Polister, B. (1995). Deinstitutionalization andcommunity integration of adults with mentalretardation: Summary and comparison of theone-year follow-up residential data for the

Minnesota Longitudinal Study. Minneapolis:Research and Training Center on ResidentialServices and Community Living, Institute onCommunity Integration (UAP), University ofMinnesota.

Hewitt, S., & Larson, S.A. (ExecutiveProducers). (1996). Helping hand: Anintroductory video on a career in direct service.Minneapolis, MN: Institute on CommunityIntegration; St. Cloud, MN: St. Cloud TechnicalCollege.

Hewitt, A., Larson, SA., & Ebenstein, W.(1996). State initiatives to address direct supportworker issues. In T. Jaskulski, & W. Ebenstein(Eds.). Opportunities for excellence:Supporting the frontline workforce (pp. 19-39).Washington, DC: President's Committee onMental Retardation.

Hewitt, A., Larson, S.A., Ebenstein, W., & Rose,J. (1996). Future directions: Creating asufficient, competent and stable direct supportworkforce for the 21' Century. In T. Jaskulski,& W. Ebenstein (Eds.). Opportunities forexcellence: Supporting the frontline workforce(pp.120-137). Washington, DC: President'sCommittee on Mental Retardation.

Holland, J.E., & George, B.W. (1986).Orientation of new employees. In J.J. Famularo(Ed.). Handbook of human resourcesadministration (2n4 Edition)(pp.1-24,35). NewYork: McGraw-Hill.

Jacobson, J.W., & Ackerman, L J (1989). Staffattitudes and the group home as a work place.Albany, NY: New York State Office of MentalRetardation and Developmental Disabilities.

Jaskulski, T. & Metzler, C. (1990). Forging anew era: The 1990 reports on people withdevelopmental disabilities, Appendix.Washington, D. C.: National Association ofDevelopmental Disabilities Councils.

19

21.

Jaskulski, T. & Whiteman, M. (1996). Familymember perspectives on direct support workers.In T. Jaskulski, & W. Ebenstein (Eds.).Opportunities for excellence: Supporting the

frontline workforce (pp. 56-75). Washington,DC: President's Committee on MentalRetardation.

Jones, A.A., Blunden, R., Coles, E., Evens, G.,& Porterfield, J. (1981). Evaluating the impactof training, supervision, feedback, self-monitoring, and collaborative goal setting onstaff and client behavior. In J. Hogg & P. Mittler(Eds.), Staff training in mental handicap (pp.213-299). Combridge, MA: The MIT Press.

JWT Specialized Communications (1996). Costper hire statistics. Figures based on reportsfrom the Employment Management Association.Http://www.jwtworks.com.

Lakin, K.C. (1981). Occupational stability ofdirect-care staff of residential facilities formentally retarded people. Doctoral Dissertation.Minneapolis: University of Minnesota.

Lakin, K.C., & Bruininks, R.H. (1981).Occupational stability of direct-care staff ofresidential facilities for mentally retardedpeople. Minneapolis: University of Minnesota,Department of Psycho - educational Studies,Developmental Disabilities Project on ResidentialServices and Community Adjustment.

Larson, S.A. (1997). Recruitment issues forMinnesota agencies serving people withdevelopmental disabilities or severe andpersistent mental illness: Final Report. Reportprepared for the Alliance for Consumer Options.Minneapolis: Center on Residential Services and

Community Living, University of Minnesota.

Larson, S.A, & Lakin, K C (1997). Alongitudinal study of turnover among newlyhired residential direct support workers in small

community homes serving people withdevelopmental disabilities: A monograph.Monograph submitted to the AmericanAssociation on Mental Retardation, WashingtonDC.

Larson, S.A., Hewitt, A., & Lakin, K.C. (1994).Residential services personnel: Recruitment,

training and retention. In M. Hayden & B.Abery (Eds.). Challenges for a service systemin transition: Ensuring quality communityexperiences for persons with developmentaldisabilities (pp. 313-341). Baltimore: Paul H.Brookes.

Larson, S.A., & Lakin, K C (1991). Parentattitudes about residential placement before andafter deinstitutionali7ntion: A research synthesis.Journal of the Association for Persons withSevere Handicaps, 16, 25-38.

Larson, S.A., & Lakin, K.C. (1992). Direct-carestaff stability in a national sample of small grouphomes. Mental Retardation, 30, 13-22.

Leftwich, K. (1994). Job outlook 2005: Whereto find the good jobs. Vocational EducationJournal, 6, 27-29.

Legislative Budget and Finance Committee(1989). Report on salary levels and theirimpact on quality of care for client contactworkers in community-based MH/MR and childcare programs. Harrisburg, PA: PennsylvaniaGeneral Assembly.

Prouty, R. & Lakin, K.C. (1997). Residentialservices for persons with developmentaldisabilities: Status and trends through 1996.Minneapolis: University of Minnesota, Researchand Training Center on Community Living.

Task Force on Human Resources Development(1989). Human resources in community humanservice programs: Report to the Secretaries of

20

Aging, Health, and Public Welfare. Harrisburg,PA: Author.

U.S. Bureau of the Census (1994). Statisticalabstract of the United States: 1994 (114th Ed.).Washington, D. C.: U. S. Government PrintingOffice.

Wanous, J.P. (1992). Organizational entry:Recruitment, selection, orientation andsocialization of newcomers (2nd Edition). NewYork: Addison Wesley.

Wanous, J.P., Stumpf, S.A., & Bedrosian, H.(1979). Job survival of new employees.Personnel Psychology, 32, 651-662.

Whiteman, M., & Jaskulski, T. (1996).Consumer perspectives on direct supportworkers. In T. Jaskulski, & W. Ebenstein (Eds.)Opportunities for excellence: Supporting the

frontline workforce (pp. 40-55). Washington,DC: President's Committee on MentalRetardation.

Zaharia, E.S., & Baumeister, A. A. (1978).Technician turnover and absenteeism in publicresidential facilities. American Journal ofMental Deficiency, 82, 580-593.

62

BEST COPY AVAILABLE

APPENDIX A: STUDY RESULTS PRESENTATION HANDOUT

21

23

A Longitudinal Study of TurnoverAmong Newly Hired Direct Support

Workers: Final Results andImplications

July 1997Sheryl A Larson, Center on Residential Services and

Community Living, University d MN, 214B PatiosHall, 150 Pillsbury Drive SE, Minneapolis, MN55455

Phone: 612-624-6024 E-Mail: larso0722tc.umn.eduWeb site: woww.icicoled.umn.oclulici

Workforce Trends in ResidentialSettings

Turnover rates Number of smallaverage 50%-70% group homesannually increased fromVacancy rates 10% 41,826 in '92 to

Recruitment is top 78,365 in '96

staffing problem for # Human serviceadministrators workers will grow

$1,388 per hire (Ntl.) 136% by 2005

Methodology: Selection ofHomes

Included 83 agenciesSelected all homes in the MNdeinstitutionalization study and all stateoperated small group homes (54 of 110included in analysis)

Selected 56 homes randomly from across thestate

Supervisors in 143 of 166 homes agreed toparticipate (86%)

Support Provided By

Preparation of this presentation wasfunded in part by grants #H133B30072and #H133D50037 from the NationalInstitute on Disability and RehabilitationResearch (NIDRR), US Department ofEducation.

MN Longitudinal Staff Study

Facility Analysis

Facility Characteristics

# of homes 110

- ICF-MR 43- HCBS Waiver 67Year opened 1990

Daily per diem $170Located in Twin Cities 52%# residents 4.7

44

BEST COPY AVAILABLE

Resident Characteristics

Severe or profound mental retardation 64%

Moved from state institution 55%Intervention for challenging behavior 53%

Formal diagnosis of mental illness 26%

Walk w/o assistance 67%

Independent toileting (<1accident/mo.)53%

Staffing Problems

Problem % of Homes

Finding qualified staff 57%

Staff turnover 44%

Staff motivation 37%

Wage and hour considerations 29%

Staff training and development 23%

Factors Associated w/TumoverTurnover at Time 1 Tuna. at Time 2Difficulty d support needs ( +) Starting pay at Time 1 (-)

Staffing pay (4Supervisor tenure In home (-) R2 = .27, N . 90% dsw eligible paid leave (-)

Other Time 1 variables InR2 ..3e", N 90 equation: Difficulty d

support needs, Supervisorintenure home, % dsw

Other Time 1 variables In eligible for paid leaveequation: Unemployment Unemployment rate, countyrale, county population, population, average peraverage per diem, years diem, years open, ICP-MRopen, ICF-IAR stab" INe-in status, Ilve-In workersworkers

Salary and Benefits

Beginning Wage $7.07Current Maximum $9.27

% DSW Full Time 43%

% DSW w/Benefits 59%

% DSW w/Paid Leave 72%

Retention Problems

Annual turnover rate 46% Dercutty caused by turnoverTime 1 2.2

% left wen 6 months 45% Time 2 2.4% left w16-12 months 23%% left after 12 months 32% (1-very little, 4 . very much)

% current staff v./12 months Average tenure*or more tenure** Time 1 19.6 months

Time 1 65% Time 2 26.1 monthsTime 2 87%

p ..05, .` p <.01, ''''p <.001

Changes Influencing Recruitment. Retention or Training

New supervisor

Staff recruitment or retention problemsChange in support needs of residents

Changed hiring or recruitment strategies

More or better training

Changed management practicesWage and benefit changesAgency expansion

4

Strategies to Address StaffingProblems

Encourage team work among staff 80%

Manage fairly/treat workers fairly 65%

Communicate agency philosophy 61%

Establish effective communication 45%

Use clear job roles and responsibilities 34%Provide competitive pay and benefits 30%

Provide realistic information to recruits** 25%

* Lower turnover 11 one of top 5 strategies noted

Higher turnover In homes that valued this

Methods: Selection of New Hires

Up to 3 workers per home eligibleParticipants were new to the home and to theagency

Participants worked regularly scheduled shiftsin the home174 of 333 invited from 110 homes agreed toparticipate (52%)124 of the 174 were included in theseanalyses

Survival Rates for New Hires

."9*itaigierit. WkWONME.Wigis,,,.;,,A0.i*E;i,

124 0 W:Mfi::0

10-12 41 CIO% 1038&i$%

MN Longitudinal Staff Study

Study of Newly Hired Workers

Characteristics of New Hires

81% female

28.8 years old at hire96% white25% married24% with financial dependents1.9 years in field13.6 years of schooling

37% took a course on mental retardation20% currently in vocational/tech. training or college

Recruitment Sources

Inside sourcesReferral sources providing

inside Information nottypically available to personsoutside the companyrehires

referrals

In-house postingsvolunteers

staff friendslama,/ members

Outside sourcesReferral sources providing less

specific Information aboutthe organization as a placeto worknewspaper advertisementsemployment agenciesjob boards and placementoffices in high schools andcolleges

BEST COPY AVAILABLE

Recruitment Source

Source Stayers LeaversMedia Advertisement 31% 51%

Current Employee 43% 21%Friend in Field 12% 6%Other 13% 20%Inside sources 55% 36%F = 3.77, p < .05

Factors Associated withStaying

Significant Predictors Other VariablesOrganizational Current salaryCommitment Job SatisfactionOverall Met Supervisor StructureExpectations Inside recruitmentAlternative Jobs Hours per weekAvailable

Months in field at hireWilk's Lambda zi .737Chi Square = 212, p < .05

Age at hire

74% of original casesintent to leave

correctly identified

r

Advice to New Recruits:Stayers

Positive features Negative features- The job Is imry moaning - Resided betastor sal- TMs is a dry good whirs) maid needs ere

arnironnort challenging

- Tow coworkers an be km - The Mtn my end an beraffia/

Neutral advice - Pm/benefits and chances- Lamm whet dities ere for scharsernert are not

Involved good- Be responsible/met" - Be prepared to range

hese km pasorai core for men and- You wit need patience women- Treat each person vath - Coworkers ere Moult

*PRY and respect

. _

Job Outcomes for NewWorkers

Outcome Stayers LeaversN = 58 N = 46

Salary @ 30 days $7.38 $6.93Likelihood of Promotion** 3.0 1.3Intent to leave** 19.2 25.5Organizational Cornmitmen 83.0 74.9Unmet expectations** 1.6 2.2Job Satisfaction 80.2 77.0Intrinsic Satisfaction 48.5 47.3Extrinsic Satisfaction 23.5 21.4

w5.05, p .01. '"" p < .001

Incidents that Made You Want toLeave: Stayers

I don't plan to leaveProblems with coworkers

Inadequate pay/benefits or incentivesProblems with the supervisorScheduling problemsBehavior of residentsDislike duties

What Could your Employer doto Make your Job Better?

Increase or improve pay, benefits orother incentives

No changes are neededBe more personable and attentive

Do.a better job managing the homeGive me more, better or different hours

r; 7

BEST COPY AVAILABLE

ANTOFFOG

stn s

Ntir.,=.0

(9/92)

U.S. DEPARTMENT OF EDUCATIONOffice of Educational Research and Improvement (OERI)

Educational Resources Information Center (ERIC)

NOTICE

REPRODUCTION BASIS

ec_ 30,09 3

IC

This document is covered by a signed "Reproduction Release(Blanket)" form (on file within the ERIC system), encompassing allor classes of documents from its source organization and, therefore,does not require a "Specific Document" Release form.

This document is Federally-funded, or carries its own permission toreproduce, or is otherwise in the public domain and, therefore, maybe reproduced by ERIC without a signed Reproduction Releaseform (either "Specific Document" or "Blanket").