 

 

 

.3... ,....,.l.
.- . ., ... .3. s. 1 Z. . g . .. . t . . .:....

.- .. 1. .... _ .. . ..,..u ... fa J...»
.

. .o...
. ..

 

 

 

 

 

 

 

 
 
 
 

 
 

 

 

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

THE EFFECTS OF DISABILITY
ON EMPLOYMENT PATTERNS
AMONG OLDER WORKERS

presented by

Ju Hee Hwang

has been accepted towards fulﬁllment
of the requirements for the

Ph. D degree in Rehabilitation Counselor
Education

 

 

_ 54%

 

OW Major Professor’s Signature ‘

7.27.2010

 

 

 

Date

MSU is an Afﬁrmative Action/Equal Opportunity Employer

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE 1 DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:/Proj/Acc&Pres/CIRCIDateDue.indd

THE EFFECTS OF DISABILITY ON EMPLOYMENT PATTERNS
AMONG OLDER WORKERS

By

Ju Hee Hwang

A DISSERTATION

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Rehabilitation Counselor Education

2010

ABSTRACT

THE EFFECTS OF DISABILITY ON EMPLOYMENT PATTERNS
AMONG OLDER WORKERS

By
J u Hee Hwang
Given the substantial number of individuals with disabilities and the sheer rapid
growth of US. aging population, research on labor transition among older people with
disability is vital for improving quality of life for society in general, as well as people
with disabilities. Nonetheless, there has been a lack of research on labor transition and
older individuals with disabilities in the literature. To reduce and further eliminate labor
transition among older people with disabilities, as well as to provide effective
interventions and policies for eliminating, it is imperative to investigate the existence of
labor transition of older workers and elucidate mechanisms in which labor transition
occur
The purpose of this study was to investigate the prevalence of late onset disability
among older workers, the consequences for their acquired disability on employment
patterns, and the determining factors affecting later-life employment transitions. Given a
lack of research employing a national data set for employment and older workers with
disabilities, this study examined older people aged 50 and over and used the object

measures for deﬁne a disability. This study was tested chi-square, ANOVA test, binary

and multinomial logistic regression.
As an ex post facto design, this study employed the Health and Retirement Study

data. Of the total 10,890 age-eligible (age over 50) full time workers who reported

thoroughly free of disability (healthy) at the time of baseline interview, the ﬁnal sampling

consisted of 1,096 older workers reported functional or mental disability two years later.
506 male (46.2%) and 590 female (53.8%) reported that they are experiencing disabilities
It is about 10.1% prevalence of an acquired disability among full time older workers.
Surprisingly, the prevalence of mental disability was 7.3% and the prevalence of

functional disability was 2.7% among older labor force participation.

Findings for descriptive results show that about 62.4% workers still kept their full
time job even though older workers experienced some types of disabilities. Older workers
at older ages were much more likely to change their job to a par-time or quit their job
than younger workers. Older workers with mental disability consistently have a better
chance of being employed and sustaining their full time job than older workers with
functional disability. The ﬁndings also conﬁrm that older workers with high levels of

education, those who are white, those are female and those in white collar occupations

seem to more likely continue their full time job
Chi-square tests, independent t-test, and ANOVA found that signiﬁcant

differences existed in age, the level of education, type of disability, self employed status,
current occupation, industry, and ﬁnancial factors (earnings, household net assets, and
Social Security Retirement Income) between older individuals who kept their job and
who quit their full time job. The binary logistic regression and multinomial logistic
results support that gender, high level of education, mental disability, self employed
status, white collar occupation and white collar related industry, and not depending on
social security retirement pension were independently associated with employed older

workers. Implications for rehabilitation counselor practice, education and future research

are discussed.

© Copyright by
In Hee Hwang 2010
All Rights Reserved.

This dissertation is dedicated to Hee Sun Na.
Without your love, support, patience and understanding,
none of this would have been possible. I love you.

ACKNOWLEDGEMENTS

Anyone who has ever written a dissertation or observed the experience first hand
knows the process can feel exhausting and overwhelming. I would like to sincerely thank
those persons who made this research possible: Dr. John Koscuilek, Dr. Michael Leahy,
Dr. Spyridon Konstantopoulos and Dr. Lawrence Schiamberg. Special recognition goes to
my advisor and dissertation committee chairperson, Dr. John Koscuilek, for both his
direction and guidance throughout this entire process. I would also thank Dr. Michael

Leahy for his knowledge, expertise, and insight into what factors were needed to

strengthen this dissertation.
Similarly, I would like to thank Dr. Spyridon Konstantopoulos, my statistics

committee chair, for his exceptional understanding of statistical issues and his willingness

to share that knowledge. I would also like to thank Dr. Lawrence Schiamberg for offering
valuable insight and quality as dissertation committee members. Their generous time

commitments and personal support throughout this process will always be cherished.

I also have a sincere thank-you to the Project Excellence (PE): Dr. Michael Leahy,
Dr. Virginia Thielsen, Dr. Nancy Crew and my fellow research assistants. Especially, I
would like to express my genuine appreciation to Dr. Virginia Thielsen. Her enthusiasm
in research and caring her research assistants have made me through whole barriers I
have faced during my doctorial program. She often played a role like my mother by

emotionally supporting me and suggesting very clear directions whenever I felt

vulnerable.
I would be remiss if I did not acknowledge my family and friends for their

supports and love. Jae Soo Hwang, my father, always supported me and never gave up

vi

his hope on me through sacrificing his life. Like my father, Hyang Sun Yeoun, my mother,
also gave me her love and never lost her faith on me. My wonderful little brothers,
Woong Hwang and Hyun Hwang, also provided a great deal of supports to me. The most

sincere thank-you is to my loving God who always shows me his sovereignty and

goodness.

vii

TABLE OF CONTENTS

Page

LIST OF TABLES ............................................................................................................. x

LIST OF FIGURES .......................................................................................................... xi
CHAPTER I

INTRODUCTION ............................................................................................................ 1

Theoretical Framework ........................................................................................... 4

Statement of the Problem ......................................................................................... 6

Purpose of the Study ................................................................................................. 9

Research Questions ................................................................................................ 12

Definition of Terms ............................................................................................... 13
CHAPTER 2

LITERATURE REVIEW ................................................................................................. 15

The Value of Work for Older Workers ................................................................... 15

Deﬁnition of Older Workers ............................................................................. 17

The Current Workforce and Its Environment ........................................................ 18

Retirement Pattern and Job Transitions ............................................................ 19

Other Factors Inﬂuencing Job Transitions ...................................................... 21

Disability and Older Workers ................................................................................ 23

The Factors Affecting Employment for Older Workers ................................... 25

The Age Discrimination in Employment Act (ADEA) and American Disability

Act (ADA) ....................................................................................................... 29

Older Workers and Vocational Rehabilitation ........................................................ 31

Aging Population and Rehabilitation Counseling ............................................ 33

Case Management for Older Workers with Disabilities .................................. 34

Employment Nondiscrimination Legislation ................................................... 35

Training Programs for Older Workers with Disabilities ................................... 36
CHAPTER 3

METHODOLOGY .......................................................................................................... 39

Deﬁning and Measuring Disability ....................................................................... 39

Functional Disability and Mental Disability ......................................................... 41

Functional Disability ......................................................................................... 42

Mental Disability .............................................................................................. 43

The Classiﬁcation of Timing ............................................................................ 44

Overview ofthe Health and Retirement Study 45

Research Design ..................................................................................................... 47

Sample Selection .................................................................................................... 48

Independent & Dependent Variables ...................................................................... 51

Dependent Variables .......................................................................................... 51

Independent Variables ....................................................................................... 52

Data Analysis .......................................................................................................... 55

viii

#“i
Descriptive and Bivariate Analysis ............................................................... 55
Binary Logistic Regression ................................................................................... 55
Multinomial Logistic Regression ...................................................................... 57
CHAPTER 4
RESULTS ......................................................................................................................... 58
Descriptive and Bivariate Analysis ....................................................................... 58
Binomial Logistic Regression Analysis ................................................................. 81
Multinomial Logistic Regression Analysis ........................................................... 85
CHAPTER 5
SUMMARY and DISCUSSION ..................................................................................... 93
Summary of the Results ......................................................................................... 93
Limitations of the Study ........................................................................................ 99
Implications of the Study ..................................................................................... 100
Practical and Educational Implication ........................................................... 100
Implications for Further Study ....................................................................... 104
Implications for Policy Makers ...................................................................... 105
REFERENCES .............................................................................................................. 107

ix

Table

10.
ll.
12.
13.
14.
15.

16.

LIST OF TABLES

Page
Variables for the Model .................................................................................... 52
Study Sample ................................................................................................... 6O
Characteristics of the Sample ........................................................................... 61
Sample Characteristics by Disability ................................................................ 64
Sample Characteristics by Gender .................................................................... 67
Sample Characteristics by Disability and Gender ............................................ 70
Employment Status Distribution: Employed vs. Non-Employed ..................... 73
Employment Status Distribution: F ull-Time, Part-Time, vs. Non—Employed .. 75
Employment Status Distribution by Types of Disabilities ................................ 78
Socio-Demographic Characteristics by Labor Status: Disability Type ............ 79
Socio-Demographic Characteristics by Labor Status: Disability Type ............ 79
Employment Status Distribution by Disability Type ........................................ 80
Employment Status Distribution by Gender and Marital Status ....................... 81
Binomial Logistic Regression Models .............................................................. 84

Multinomial Logistic Regression (F ull—Time/Part-Time vs. Non-Employed) .86

Multinomial Logistic Regression (F ull-Time vs. Part-Time) ........................... 87

LIST OF FIGURES

Figure Page
1 . Conceptual Framework .................................................................................... 10
2. Sampling Process ............................................................................................. 50

xi

CHAPTER 1
INTRODUCTION

Today’s aging Americans are expected to make up a large share of the population
of the United States (US) (U. S. Government Accountability Office, 2001). Between the
year of 2000 and 2025, the population who are 55 years of age and older will be
increasing more sharply than any other age group (Toossi, 2004). The causes of this
phenomenon occurring for older Americans are both an increase in life expectancy and a
decrease in birth rates. Contingent with the situation is an aging workforce and a shortage
of labor, thus, the characteristics of labor force participation has changed (Danczyk-
Hawley, McMahon & Flynn, 2002; Brooke, 2003; Critchley, 2004). By 2050, projections
based on 2000 census data indicate that 60.3 percent of workers will be composed of
persons age 55 to 64 while individuals age 65 and older will account for an additional
13.4 percent of the workforce (Toossi, 2004). Dychtwald, Erickson & Morison (2006)
similarly estimated that 80 percent of the native-born workforce came from those ages
over 50 in North America over the next 15 years.

During the past half-century, the rates ofolder men who participate in the labor
force have been the dramatic decline, particularly older men. It was well known by the
term of early retirement trend (Quinn, 1997; Quinn, 2002). Historically, for nearly half a
century, older Americans have been exiting the labor force early. The age leaving from
their labor participation had declined dramatically among American men from age 70 in
1950 to age 65 in 1970, and then to 62 by the mid-19805 (Quinn, 2002; Burtless, & I
Quinn, 2002). It was possible that many have been able to support themselves with Social

Security advantage income from pensions, savings and investments, and earnings from

employment (Burtless, & Quinn, 2002). At that time, the trend leaving the labor force
was mostly the result of increasing prosperity and wealth, some of which was ‘spent’ on
additional leisure late in life (Burtless, & Quinn, 2002).

The earlier trend, however, does not exist today. Millions of skilled and
experienced workers both full time plus part time will keep their employment status since
the current retirement trend has changed (Quinn & Smeeding, 2004, Quinn, 1997;
Burkhauser, and Quinn, 1997; Cahill, Giandrea, & Quinn, 2006). Contemporary ideas
about retirement include employment which is a desirable part of the retirement process
(National Institute on Aging, 2007). Maestas (2004) correspondingly states nearly one-
half of retirees eventually return to work and many of these job changes involve
transitions into new careers. Johnson and Kawachi (2007) also report that 77 percent of
job changes by older workers leaving long-terrn jobs (with 10 or more years of tenure)
involved a move into a different occupation, and 73 percent involved a move into a
different industry.

These concerns, in turn, have had attention to increase employment among older
workers. In addition, many public policies are changing to encourage more work by older
Americans. For example, mandatory retirement has been eliminated, Social Security
work disincentives are being reduced, and the normal retirement age for Social Security

is currently 62 in 2000 and this age will gradually increase to 67 in 2022 (Quinn, 2002).
At the same time, many employer pension plans continue to penalize work, in order to
induce departure at particular late ages.

The ability to work productively and to return to work later in life is an aspiration

for most aging Americans in recent days (Bruyere, Erickson, Wilson, & Somerville,

2004; Finch & Robinson, 2003). In reviewing current trends among older adults, two
dominant trends were found: 1) increased participation in the labor force and 2) a
corresponding prevalence of acquired disabilities (Quinn, 2002; Swett & Bishop, 2003).
Most of aging individuals often choose transition jobs before their retirement or often
immediately return to the labor force in the same, similar, or different occupations after
their retirement (Quinn, 1997; Burkhauser & Quinn, 1997). They bring good skills, good
work habits, and strong work ethics in the form of advantages to the labor force. However,
they also bring the disadvantages of the increased likelihood of acquiring a disability,
which in them least to the needs for rehabilitation services and community programs.
With the age of workers increasing, the older individuals are likely to acquire a

disability in the incidence, severity, and duration of disability due to the aging process
(Kraus & Stoddard, 1991; McNeil, 1997; Bruyere, 2006; Vargo & Grzanowicz, 2002). If
older workers experience acquired disability during their later working lives,
rehabilitation services can serve them to ﬁnd or sustain their employment since obtaining
and maintaining employment for people with disability are signiﬁcant goals and are
considered a major indicator of successful rehabilitation in the rehabilitation counseling
(Michell, Adkins, & Kemp, 2006). Over the last 80 years, “rehabilitation professionals
have been committed to assisting people with disability in finding and retaining
employment” (Gillbride & Stensrud, 2003, p. 407). However, rehabilitation services were
mostly focused on the working age population with disabilities. Now is the opportune
time to attend to the demographic composition of older clientele. This is a key to solve in
the rehabilitation systems of today (Kemp & Mosqueda, 2004). Rehabilitation systems in

both public and private settings can contribute to the retention of aging workers who will

likely need intervention and workplace supports. Nowadays, rehabilitation professionals
take the steps necessary to expand and develop our dated rehabilitation systems and
community programs by promoting greater coordination of services.

Theoretical Framework

The definition of employment includes obtaining economic self-sufficiency,
establishing a route for social identification, and developing a personal network for the
majority of people living in this society (Moore, F eist-Price, & Alston, 2002).
Accordingly, employment is a critical factor in developing independence and community

integration (Brown, 2003). Employment is also strongly related to established variables
such as income, sense of purpose, social relationships, social participation, skill
development, and creativity (Fagin & Little, 1984).

A number of studies regarding QOL report that people may increase their levels of
economic, social, psychological, and physical well-being through employment (Fagin &
Little, 1984; Hess, Mead, Forchheimer, & Tate, 2004; Krause, 1996). Because working or
employment provides profound feelings of well-being, typically the individual makes a
social contribution, which is significantly related to QOL (Fagin & Little, 1984; Hess, et
al., 2004; Krause, 1996; Kinjerski & Skrypnek, 2004). It is about ﬁnding meaning and
fulﬁllment beyond one’s self.

For the aging population, the issues of how to maintain a positive quality of life
have become more important than ever before (Kempt, 2000). To increase a positive

quality of life, aging individuals could be more involved in social and community
activities by maintaining or gaining employment regardless of whether or not they

experience health limitations. A willingness to work has been observed in many studies of

older workers (Reynolds & Horn, 2005; Heidkamp & Horn, 2008). The importance of
older adults undertaking productive activities has been advocated to improve health,
longevity, and quality of life (Lee, Saito, Takahashi, & Kai, 2007).

In terms of QOL issues for older adults, changed retirement patterns have
inﬂuenced older adults to continue an involvement in work related activities.
Traditionally, people have believed that retirement is one big event shifting from work to
leisure. Trends observed in previous studies about an older workforce and in statistics on
labor force participation have documented a dichotomous view of the labor force
transition on whether they remain employed or exit the labor force (Quinn, 1997). The
dichotomous view of the labor force transition means that workers move directly from a
full-time career job to a complete labor force withdrawal as a single event. However,
many Americans do not retire in a dichotomous fashion today. Permanent or onetime
moves directly out of the labor force from full-time work are relatively rare. Scholars
have agreed to view retirement as a process instead of a single event (Quinn, 1997; Ruhm,
1995; Cahill, Giandrea & Quinn, 2006). The concept of retirement is therefore now a
process for many older workers rather than an abrupt transition. Most recent study reports
that about 25 percent of older workers who retired during the 19905 eventually returned
to work (Maestas, 2004), and many of these job changes involved job transitions into part
time work or new careers.

The baby boomer generation is likely to continue to work due to the changed the
pattern of retirement. Contemporary ideas about retirement include employment which is
a desirable part of the retirement process (National Institute on Aging, 2007). Prior to

retirement, the majority of older Americans who have had career jobs are likely to make a

bridge job within the same occupation or across occupations (Quinn, 1997; Burkhauser &
Quinn, 1997; Cahill, Giandrea, & Quinn, 2006). The bridge job means a job that is either
in a different type of occupation (on a new job or self-employment) or a switched job
either to part-time or full-time following a full-time career (Burkhauser & Quinn, 1997).
Quinn and Burkhauser (1994) noted that the importance of part-time work has risen
dramatically with age. According to the Bureau of Labor Statistics (US. Department of
Labor, 1999), 16% of men in 1999, aged 60—64, worked part-time (less than 35 hours per
week), while a full 50 percent of men over 65 did. About 33 percent of women aged 60—
64 worked part-time, and 60 percent of women 65 and over worked part-time. The study
by Purcell (2009), using the Census Bureau statistics conducted between 1990 and 2008,
indicates approximately 70 percent of men and 59 percent of women aged 55 to 64 were
still either working or looking for work compared to the 25 to 54 age group. Ninety one
percent of men and 76 percent of women participated in the labor force in 2008. In
addition, Maestas (2004) correspondingly stated nearly one-half of retirees eventually
return to work and many of these job changes involve transitions into new careers.
Statement of the Problem and Signiﬁcance of the Study

Providing more opportunities for older people to participate in and contribute to
social life, including through employment, has become an important issue regarding the
QOL in our society. Many older people continue to demonstrate a strong desire to
participate or remain in the workforce (Brown, 2003). One reasonable explanation for
this trend is that the desire to participate or remain in the workforce is related to both
financial and social needs in order to maintain or improve their QOL even though the

older workers consider extended employment under more varied conditions.

Another reasonable explanation for this trend is that the majority of older
Americans with a job history have experienced labor transitions prior to retirement
(Quinn, 1997; Burkhauser, and Quinn, 1997; Cahill, Giandrea, & Quinn, 2006). The
Health and Retirement Study (HRS) suggested that the majority of older Americans with
full-time career jobs moved on to jobs of a part-time or short-duration basis that served as
bridges to retirement before exiting the labor force completely at some point following
full-time career employment (Cahill, et al., 2006). Johnson and Kawachi (2007) stated
that 77 percent of job changes by older workers who left their long-term jobs that
involved 10 or more years of tenure, included a move into a different occupation, and 73
percent moved into a different industry.

If older workers experience deteriorated health, they change jobs many times
within several years at the onset of their poor health, since changing jobs (job transition)
is an important way for older workers to adapt to their situation (Bound, Schoenbaum,
Stinebrickner, & Waidmann, 1998). Dwyer and Mitchell (1999) also state that if older
workers experience poor health or disability, they have shown a positive inﬂuence on
labor force participation rather than negative. Because people associated with poor health
may feel in the need for an increases in work to replace lost income, to maintain health
insurance, or to pay for the additional health services consumed.

While researchers have extensively studied the various social and economic
consequences of disability, its impact on labor transition patterns in on older workers
with disabilities in the US. has received little attention. Rehabilitation counseling '
professionals work through in a variety of settings, such as public, private, mental health,

and one-stop centers in the Workforce Investment Act programs may however lack

(Kemp, 1985).

Older clients with disabilities may find or sustain their employment more easily
through rehabilitation services (Michell, Adkins, & Kemp, 2006). A fairly recent study
reveals that the number of clients over the age of 50 has continuously increased every
year (Wadsworth & Kampfe, 2004) because rehabilitation services are designed to (1)
increase the probability of being employed and the duration of periods of employment as
well as to (2) decrease the probability that an employed person would become
unemployed at a later date (Gibbs, 1991). If older workers with disability desire to have
or sustain a job, rehabilitation fields should therefore reﬂect the needs of older workers
with disabilities (Remenyi, 1995; Saxon & Spitznagel, 1992).

Disability system in the US. also needs to continue to provide essential support to
those Who need it, but it must also establish a mechanism that maintains an older
worker’s ability to stay actively employed. To do so, this is essential to understand the
Characteristics of older workers with acquired disabilities and to maintain an awareness of
the labor force transition patterns. Better information about later life career changes

would improve understanding about employment transition patterns and opportunities for

anyone 40 or over (Robert Wood Johnson Foundation, 2006). In proﬁles on older workers
presented by Georgetown University’s Center on an Aging Society, data suggests an older
worker is classiﬁed an age 50 and over. The Health and Retirement Study (HRS) and
other studies also follow this definition. Thus, for the purpose of this current study, older
workers are considered to be 50 years and above, unless otherwise specified.

This current study examines the prevalence of acquired disability on full time
older workers, the consequences of their later-life transition, and the determining factors
affecting the transition pattern of later-life employment as the conceptual framework
illustrated (Figure 1). For determining the factors affecting labor life employment
transition patterns, this research ﬁrst estimated the relationship between the level of
individual, labor market, and financial characteristics and broad concept of two labor

transition outcomes (employment vs. non-employment. Then, three categories, (a)

holding the full time job (b) switching to a part time job, and (c) being non-employed
were examined as outcome variables.

To accomplish its purpose, this study used the longitudinal Health and Retirement
Survey (HRS) to analyze patterns of employment within a group of older workers who
continued to work versus non-employed following a disability. Accordingly, this paper
estimated to assess the prevalence of late-onset disability by using objective measures in
order to minimize measurement errors in most of the health and economic studies.
Objective measures of disability are also crucial for VR ﬁelds since Vocational
Rehabilitation (VR) agencies mostly target VR services to individuals whose disabilities
are signiﬁcant or most signiﬁcant. Eligible VR customers must have a physical or mental
impairment that substantially interferes with the ability to prepare for, enter, engage, or
retain employment. Thus, deﬁning disability using objective measures in this study make

a clear sense for the future VR studies.

10

 

Older workers ’
Jobs I
(Before having
a Disability)
who were:
(1) Age over

50
(2) Healthy
(3) Have
Full-Time

Work

 

 

 

  
  

W

 

 

 

 

 

 

  

2yrs.
Later
Disability

 
   
 

 

 

 

 

 

 

 

 

 

 

 

 

 

Older workers ’ Jobs 11

(After having a Disability)

 

 

 

 

 

 

r Full lime Job

 

 

Ernployed

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Job

 

 

 

 

 

 

 

 

 

 

Part Time

 

 

 

Non-Enrployed

 

 

 

L

Figure 1. Conceptual Framework

Given labor market transitions that are a disability in late life work situation, this

empirical analysis describes comprehensive differences in demographic, labor market,

and financial characteristics among employed (full time or part time work) vs. non-

employed by types ofdisabilities and gender because most studies have not focused on

the gender and types ofditferences

disabilities. In addition, it was important to understanding separately the labor market

in job transitions among older workers with

transition of different types of disability due to the difference 1n the nature of the

disability and to treating “

This study identiﬁed how the demographic factors (age,

11

older workers with disability” as a single group.

 

 

 

 

 

 

 

gender, race, level of

education, marital status, and number of people living in the household), human capital
(earning, household wealth, and Social Security retirement funding), and employment
characteristics (occupation, industry, and self-employed status) affect labor market
transitions after experiencing disability. These factors were used to evaluate where these
characteristics account for the two types of disabilities in full time older workers.
Modeling the effect of disability may offer important new insights into the rehabilitation
ﬁelds that may help to enhance the productivity of older workers and it become
signiﬁcant for those who work in and plan programs of rehabilitation for aging workers.
Research Questions
The followings are research questions:
1. What is the likelihood of older workers who have acquired disabilities in their
later lives?
2. What are the demographics, workforce, or ﬁnancial characteristics of older
workers who have acquired disabilities in their later life?
3. Speciﬁcally, how do employed and non-employed older workers differ in
demographic, workforce, and ﬁnancial characteristics?
4. What factors determine the older worker’s labor transition?
a) What factor affects the choice of a full time job by the disabled older

worker?

b) What factor affects the choice of a part time job by the disabled older

worker

C) What factor affects the choice of a non-employed by the disabled older

worker

12

Deﬁnition of Terms

physical or mental impairment that substantially limits one or more of major life
activities (McNeil, 2001).

Functional Disability: Deﬁnes as a multidimensional concept encompassing
mobility, large muscle functioning, ﬁne motor skills, gross motor skills, and the ability to

perform activities of daily living (ADLs) such as eating, dressing, bathing, toileting,

grocery shopping, and being mobile outside the home, managing money, and using a
telephone).

Mental Disability (Depression): Defines as a depression or depressive symptom
measured by the Center for Epidemiologic Studies Depression (CES-D) scale (Radloff,
1997). The CES-D is a self-reported inventory of depressive symptoms (“was depressed,”
“everything was an effort,” “sleep was restless,” “was happy,” “felt lonely,” “enjoyed
life,” “felt sad,” and “could not get going”) that occurred in the week prior to the
respondent’s interview date.

Late Life Disability: It is “those that occur in the later years and are often
associated with the aging process” (Corthell and Fleming, 1990, p.17).

Older Workers: Deﬁne an “older worker” as beginning at 51 years old.

Labor F orce Participation: Labor force participation includes all individuals who

are currently employed regardless of their previous history of retirement.

F all-Time Job: Deﬁned as employment that consists of at least 35 hours per week

13

or 36 weeks per year (HRS codebook, p. 921).

Part-lime Job: Deﬁned as employment that consists of less than 35 hours per
week or 36 weeks per year including partly retired individuals (a person reports working
part time and mentions retirement before) (HRS codebook, p. 921).

Non—employment: Refers to the individual who is not working at the time of the
interviews and has the following work status: unemployed, retired, applied disability
beneﬁts, or exit the labor force.

Bridge Job: Refers to a job that follows a full time career either in a different type
of occupation (on a new job or self-employment), that is, a switched job either part-time
or full-time following a full-time career (Quinn & Burkhauser, 1994; Auinn, Burkhauser,

Cahill, & Weathers, 1998; Ruhm, 1995).

14

CHAPTER 2
LITERATURE REVIEW
The Value of Work for Older Workers

The deﬁnition of employment includes obtaining economic self-sufﬁciency,
establishing a route for social identiﬁcation, and developing a personal network for the
majority of people living in this society (Moore, Feist-Price, & Alston, 2002).
Accordingly, employment is a critical factor in developing independence and community
integration (American Association of Retired Persons, 2001). In addition, employment is
strongly related to established variables such as income, sense of purpose, social
relationships, social participation, skill development, and creativity (Fagin & Little 1984).
For aging individuals, the value of work can also apply as a source of personal identity, a
reward, and a strong component of QOL (Szymanski, Ryan, Merz, Trevion, & Jonhston-
Rodriguez, 1996). Beneﬁts associated with employment include income and health care.
It is possible for the worker to participate in productive activities, make a social
contribution, and have psychological beneﬁts. Such psychosocial beneﬁts often comprise
a sense of well-being and self value (Fagin & Little 1984; Hess, Mead, Forchheimer, &
Tate, 2004; Krause, 1996; Turner & Turner, 2004). Many of these beneﬁts have a deﬁnite
effect on speciﬁc or global indicators of the QOL for people with disabilities.

A number of studies regarding QOL have reported that people increase their
levels of economic, social, psychological, and physical well-being through employment
because it is about ﬁnding meaning and fulﬁllment beyond one’s self (Fagin & Little,
1984; Hess, Mead, Forchheimer, & Tate, 2004; Krause, 1996; Turner & Turner, 2004).

Working or employment provides profound feelings of well-being and typically one’s

15

work makes a social contribution, which is signiﬁcantly related to QOL (Fagin & Little
1984; Hess et al., 2004; Kinjerski & Skrypnek, 2004). Researchers also suggested that for
older workers employment results in less stress, and more satisfaction with their personal
lives. They were less health care, take fewer sick days, and staying longer with their
company than those of their counterparts who do not have a job (Gallup Organization,
2003; Harter, Schmidt, & Hayes, 2002).

For the aging population, the issues of how to maintain a positive quality of life
have become more important than ever before (Kempt, 2000). To increase the
individual’s positive quality of life, elders could be more involved in social and
community activities by maintaining or gaining employment regardless of whether or not
they experience health limitations. Hively (2004) indicated that older workers who
voluntarily continue to participate in the labor force have the desire to add meaning and
purpose to life, to use special skills and education, and to preserve an identity tied to
work.

Studies have observed the willingness to work among older workers (Reynolds,
Ridley, & Horn, 2005; Heidkamp & Horn, 2008). It is because “for older persons, work
provides a means to stay active and independent. It is rewarding, promotes good health,
and contributes to a sense of usefulness and purpose in life” (Myers, 1992. p. 283). The
importance of older adults undertaking productive activities has been advocated to
improve health, longevity, and quality of life (Lee et al., 2007). Other studies have
supported the notions that the ability to work productively and continue to participate in
work activities in later life is an aspiration for the quality of life among most aging

Americans (Alwin, 1998; Bruyere et al., 2004).

16

Menec (2003) examined the longitudinal relationship of activity among the aging
population. He found that many older people continue to complete a substantial amount
of productive work, regardless of whether they are paid or unpaid. This means that social
and productive activities were positively related to happiness, function, and mortality.
The MacArthur Foundation Study of Aging in America (1999) conducted by Seeman,
Unger, McAvay, & Mendes de Leon also suggested that more than 40 percent of the
elderly report at least 1,500 hours of productive activity per year including volunteering.
Other studies revealed similar ﬁndings. A study from the Georgetown University’s Center
on an Aging Society (2000) found this effect of employment: based on self-reported
health status young retirees age 51 to 59 who have no job are much less healthy than
older workers (age 60 and older) and their working counterparts (young retirees who
have a job). Furthermore, workers aged 60 and older are almost two times more likely to
report that they are in fair to poor health than their nonworking counterparts.

Definition of Older Workers

Myers (1992) indicated that the criteria for deﬁning an individual as an older
person remain elusive. Generally, studies have disagreed on the deﬁnition of older
workers. In the United States, the deﬁnition of “old age” is different in legislative and
administrative policies: “the Older American Act designates age 60 as the base age for
eligibility for a variety of social and support services” (Cottrell & Fleming, 1990, p. 11),
and “the Social Security Act deﬁnes age 65 as a beginning point of entitlement for the
collection of full old age and retirement beneﬁts” (Cottrell, & Fleming, 1990, p. 11).

Most deﬁnitions are based on chronological age, and some are associated with

role status. Age 55 to 64 is widely regarded as the age of older workers for statistical

l7

purposes but “the legal deﬁnition tallies with a pervasive attitude among many corporate
recruiters who consider 40 and up unacceptably old” (Robert Wood Johnson Foundation
2006, p 6). In the Age Discrimination in Employment Act of 1967, an older worker was
deﬁned as an individual who is 40 years or older from employment discrimination (U.S.
Equal Employment Opportunity Commission, 1997). Researchers frequently deﬁne older
workers based on the their status, such as career changers, retirees returning to the labor
force full or part time, displaced/dislocated workers, or displaced homemakers (Moon &
Hushbeck 1989). Inconsistencies in deﬁning older workers have resulted in complexity
when comparing and contrasting demographic predictions, research ﬁndings, and
analyzing policies (Lockwood, 2003).

There are various age eligibility requirements for membership in organizations
that service the older adult. For example, American Association of Retired Persons
(AARP) requires that individuals must be at least 50 years of age to apply for
membership (Rix, 2004). In the proﬁles on older workers presented by Georgetown
University’s Center on an Aging Society, the data suggest that being an older worker
starts above age 50; studies generally consider an employee aged 50 or over to be an
older worker as the Health and Retirement Study does. Thus, for the purpose of this paper
older workers are considered to be above 50 years. Unless otherwise specified.

The Current Workforce and Environment

The current workforce trend in the United States is turning gray and shrinking
because young adults entering the workforce comprise a relatively small percentage of
the population (Giorgianni, 2005; Pitt-Catsouphes & Matz-Costa, 2008). Thus, concerns

about “whether employers was able to meet future workforce needs, and whether Social

18

Security and Medicare trust funds remain solvent” (Abraham & Houseman, 2004, p 1)
have been raised. These concerns, in turn, are recognized by federal policy because “the
smaller share of the population that is of major working age make it more difficult for
employers to satisfy their demand for labor” (Abraham & Houseman, 2004, p 1), and the
number of workers with disabilities covered by Social Security beneﬁts has increased
over time” (Ho, 2002, p 1). Thus, an increase in labor force participation among older
Americans regardless of the existence of disability could be the solution for these
problems.

Retirement Pattern and Job Transitions

Traditionally people have thought that retirement is one big event (the
dichotomous view) shifting from full-time employment to complete withdrawal from the
labor force (Quinn, 1997). However, not all older Americans retire in a dichotomous
fashion today. Thus, many scholars have agreed to view retirement as a process instead of
a single event (Quinn, 1997; Ruhm, 1995; Cahill, Giandrea, & Quinn, 2006). Often older
workers leave the labor market in a gradual fashion. Thus, contemporary ideas about
retirement include labor force participation as a desirable part of their retirement process
(Cahill, et al., 2006).

Contemporary ideas about retirement include employment as a desirable part of
the retirement process. Conducted by an Brown’s study (2003), a nationwide telephone
survey designed of 2,001 individuals between the ages of 50 and 70 years old was to
explore the current deﬁnition of retirement. This report states that some older workers
deﬁne retirement as leisurely pursuits, fun, and time with family and friends, but “over
half (53%) state that their deﬁnition of retirement includes working for enjoyment, not

money, and 42 percent report that their retirement deﬁnition includes having to do some

19

kind of work to help pay the bills” (p 3). Maestas (2004) correspondingly states nearly
one-half of retirees eventually return to work, and many of these job changes involve
transitions into new careers. Johnson and Kawachi (2007) report that 77 percent of job
changes by older workers leaving long-term jobs (with 10 or more years of tenure)
involved a move into a different occupation, and 73 percent involved a move into a
different industry.

Other retirees use job transitions and second careers, sometimes called bridge job.
After retirement, older workers re-entered the work force in a number of different ways.
Some transition is from one to another similar job. Some of them have part time work as
a second career. It often includes part-time work and sometimes self-employment (Cahill,
et al., 2006). According to Bergman (2007), older workers with job experiences are
ﬁnding part-time jobs in school bus transportation, real estate sales and leasing, retail
stores, health care, and accounting and tax preparation services.

The self employed option offers ﬂexible work schedules and expects to ﬁnd a
growing number of interested and available older workers because older workers are
expressing concerns about the ﬁnancial burdens of being out of work for extended
periods during retirement. Evidence from the Census Bureau Statistics Survey conﬁrm
that between 1990 and 2008 approximately 70 percent of men and 59 percent of women
aged 55 to 64 were still either working or looking for work compared to the aged 25 to 54
group whereby 91 percent of men and 76 percent of women participated in the labor
force in 2008 (Pucell, 2009). He also reports that the employment rate of men aged 55 to

61 was 73 percent in 2008.

20

Other Factors Inﬂuencing Job Transitions

Federal policies have altered to encourage older Americans to work longer. The
recent policy initiatives have tended to focus on ﬁnancial incentives and dis-incentives.
For example, the policies for Social Security beneﬁts, Social Security ﬁnancial incentives,
and employer pensions have changed. First, Social Security beneﬁts are expected to
replace only about 36 percent, instead of 40 percent, of pre-retirement income for a 65-
year-old worker who retires in 2025 (Social Security Administration, 2006). In addition,
strong Social Security ﬁnancial incentives to claim beneﬁts at age 65 have been
eliminated for the average worker because of a continuing increase in the delayed
retirement credit (Giandrea, Cahill, & Quinn, 2008).

Dramatic changes in employer pensions, the key source of retirement income
provided by employers, also affect younger retirees. Munnell, Sunde’n, and Lidstone
(2002) and Munnell and Perun (2006) found that deﬁned-beneﬁt (DB) pension plans that
pay a lifetime annuity at retirement have declined substantially in recent decades as
companies and workers shift to deﬁned—contribution (DC) pension plans in which
retirement beneﬁts depend on workers’ contributions. In contrast to those under DB
pensions, individuals in DC plans who manage their own accounts face ﬁnancial market
risk and work disincentives. Thus, “labor force participation among people 55 and older
might increase due to the trend away from deﬁned-beneﬁt pension plans that pay a
guaranteed beneﬁt for life toward deﬁned contribution plans, which often pay out a single
lump sum at retirement” (Purcell, 2009, p. 5).

Several studies have identiﬁed factors that inﬂuence employment among older

workers, such as general economic conditions, eligibility for Social Security beneﬁts, the

21

availability of health insurance, and the prevalence and design of employer sponsored
pensions (Purcell, 2009; Heidkamp, Carl, & Horn, 2008; Reynolds, Ridley, & Horn,
2005). According to the Heldrich Center survey, more than one in three (35%) workers
has insufficient retirement savings such and thereby they have to rely fully on Social
Security beneﬁts when they do retire (Reynolds, et al., 2005). That is why 39 percent of
male recipients and 33 percent of female recipients remain employed even though older
individuals receive employer pensions and retirement savings plan (Heidkamp, et al.,
2008; Purcell, 2009). Another primary reason to return to work is insufficient personal
savings or Social Security pensions. In addition, the insufficient retirement savings and
personal savings plan are likely driven by their need to obtain affordable employment-
based health insurance and the need to continue to accumulate savings in employment.
However, the reasons given for a return to work or to postpone retirement are varied. The
older workers consider extended employment under more ﬂexible conditions. Financial
need and access to health insurance are important for older workers, but so is interest in
keeping active, challenged, and engaged (Brown, 2003).

Other factors additionally affect employment opportunities for older workers
(McNaught & Barth 1992). The nature of work in our society has changed from
manufacturing-based jobs to service— and knowledge-based employment. As a result,
most American jobs no longer involve heavy physical demands and are therefore suitable
for older workers. Advances in medicine and technology have allowed individuals to
function well in work settings and have reduced the severity of some medical conditions.
Correspondingly, “as the workplace adapts to the needs of a more diverse workforce,

accommodating more women and ethnic employees, managers and corporate policy may

22

also become more ﬂexible with respect to the needs of older workers, providing
opportunities for part-time or ﬂex-time positions. In addition, the pressures of a
globalizing economy place increased pressure on employers to retain skilled workers and
older” (Hansson, DeKoekkoek, Neece, & Patterson, 1997, p. 203)
Disability and Older Workers

Older individuals are likely to acquire a disability due to the aging process in
combination with an increase in life expectancy (Kraus & Stoddard, 1991; McNeil, 1997;
Bruyere, 2006) and further states “the incidence and prevalence of disability increases
with age” (He et al., 2005. p. 59). In 1995, about 16 million of an estimated 31 million
seniors age 65 and over reported some level of disability (McNeil, 1997). If this trend
continues to 2030, half the older population will have some disability. Numbers will be
much larger because increased life expectancy is accompanied by an increased
prevalence of chronic diseases and conditions that are positively associated with aging
(National Center for Health Statistics, 2006). According to the National Institute on
Aging (2007), “persons who had some form of health event (acute, chronic, or decline in
functional ability) between 1992 and 1994 were about twice as likely to be out of the
labor force in 1994 and 1996 compared with persons who had not experienced a
signiﬁcant health event” (p 33). Platt (2001) suggested that by the age of 50, a person’s
ﬁrst serious medical problem might occur with a 25 percent chance that it was a life-long
condition.

As mentioned earlier, the decision to exit the labor force (retire) is affected by a
number of factors, including the availability of health insurance, Social Security

eligibility, and ﬁnancial resources. However, several studies have also pointed to a

23

deteriorated health status (disability status) as a signiﬁcant determinant. Generally, it is
believed that deteriorated health would lower productivity and thus reduce the probability
of being employed based on the concept of the economic rationale (Doshi, Cen, &

Polsky, 2007). In a study Maestas (2004) also supported that “the individuals acquiring
new health conditions were somewhat less likely to go back to work than were those who
saw no change in health (16.1 percent versus 18.4 percent)” (p. 21). Santiago &
Muschkin (1996) similarly report that disability greatly reduced labor force participation
and the return to work. They state that “labor force participation rates were approximately
50 to 75% lower for persons with disabilities than their non-disabled counterparts”
(p.304).

Being older and having a disability makes it twice as hard for older workers to

ﬁnd or sustain a job. Recent evidence reveals that in the US. older workers over age 50
experienced a disproportionate share of involuntary job losses due to chronic disease
and/or a declining health status (Gallo, Bradley, Dubin, Jones, Falba, Teng, & Kasl, 2006;
Chan & Stevens 2001; Hipple 1999). The older workers’ share of involuntary job losses
in the United States has grown consistently in recent decades (Gallo, et al., 2000).
Researchers also have found that as workers age they become more vulnerable to injury
(Minter, 2002). A study using state data on workers’ compensation claims linked to
longitudinal earning records notes that older workers are more likely to have permanent
disabilities because of work-related injuries (Biddle, Boden, & Reville, 2003), even
though older workers reportedly have fewer work-place accidents (Biddle et al., 2003). A
similar pattern was observed in the studies by UnumProvident Company (2005) and the

Bureau of Labor Statistics (2007): the chance of becoming either long- or short-term

24

disabled increases with age. “Workers ages 50 to 59 make up approximately 19 percent of
the workforce and 19 percent of short-term disability claims, but 34 percent of long-term
disability claims” (UnumProvident Company, 2005, p. 7).

Bound, et al., (1998), however, report if older workers experience deteriorated
health, they change their jobs many times within several years at the onset of their poor
health, since changing jobs (job transition) is an important way for older workers to adapt
to their situation. Dwyer and Mitchell (1999) also supported the notion that if older
workers experience poor health or disability, they have shown a positive inﬂuence on
labor force participation rather than negative. It is because people associated with poor
health may induce an increase in the need for work to replace lost income, to maintain
health insurance, or to pay for the additional health services they have used. Even though
there is a question how the severity of disabilities or injuries among older workers affect
keeping a job in later life, essential questions arise ﬁrst: how many older workers
employed in full time jobs will experience an acquired disability which limits their work
activities? What are the consequences of disability on labor transition patterns among
older workers? Are there any differences by gender and types of disabilities? Are there
any different characteristics between employed workers (continued workers) and non-
employed workers (discontinued workers)?

F actors Affecting Employment for Older Workers
In recent years, researchers have acknowledged a negative association between
employment status and disability due to the nature of the disability. Platt (2001)
suggested that the nature of disability which is a person’s ﬁrst serious medical problem

among aging population might occur with a high percent chance that it was a life-long

25

condition. As an older worker have a permanent disability, the chance to get or sustain a
job would be more difficult. Choi (2000) examined the determinant factors of work status
among low income older workers and reported that older workers and those with
permanent disabilities were less likely to work regardless of gender. Although white
males represent the majority of the older working—age population, the proportions of
women and ethnic minorities have grown steadily (Barresi & Stull, 1993). In studies of
disability in men and women workers aged between 45 and 64 years old, some
researchers report that labor participation are differently shaped by gender. (Wray, 1996;
Reimers & Honing, 1996). Wray (200 ‘1) stated that women are concentrated in industries

and occupations that tend to receive lower Social Security and pension beneﬁts than men.
Many prior studies also emphasize the role of race (Wray, 1995; Flippen &
Tienda, 2000; Hayward, Hardy, & Chen, 1996) in the relationship between employment
and disabilities. The recent study on disability and work among pre-retirement-age
Americans by Wray (1995) ﬁnd disproportionate results between White and non-White:
“African Americans and Latinos had disproportionately more potentially disabling health
conditions, greater difﬁculties with function, and more severe symptoms” (p. 294). Given
the links between work status and pre- and postretirement economic security, “African
Americans and Latinos tended to report lower job status and wages as well as fewer job
beneﬁts than did Anglo/others” (Wray , 1995, p. 294). Generally research reports black
workers at middle and older ages reports are more likely than white to experience
disability and to experience restriction in a major activity, including work (Hayward,
Friedman, & Chen, 1998). Flippen & Tienda (2000) also demonstrated that mature blacks

and Hispanics, who are more vulnerable to lay—off, disability. This population frequently

26

reports that they have few retirement beneﬁts. Thus, the role of race and gender
differences among older workers has signiﬁcant implications for inequity in economic
well-being among aging population.

Among the older workers, the effects of marital status on employment differ for
men and women. For men, being married is positively associated with the likelihood of
working, whereas for women, being married is negatively correlated with the likelihood
of working. Loprest, et al., (1995) reported that work status among older workers with
disabilities is positively associated with having no child for women. They found that
single women are more similar to men on labor force participation pattern than to married
women. In the study of Reitzes, Mutran, & Fernandez (1998) researching general older
women regardless of existence of disabilities, they similarly found that women at any age
are more likely than men to exit the labor market due to the response of family care-
giving obligations. This phenomenon is casing of women’s unstable work histories and
their concentration in lower skill occupations. It could make women more vulnerable to
involuntary job loss as they approach retirement age (Flippen & Tienda, 2000).

Most previous studies have shown that education was positively related to the
labor participation (Holtmann, Ullmann, Fronstin, & Longino, 1994; Hayward & Hardy,
1985). Education also plays a signiﬁcant role in the different types of labor force
participation in which people with disabilities are engaged (Schechter, 1999). Older
workers with higher levels of educational attainment are found to be in white-collar
employment sector. The factor of education attainment level might reﬂect economic and
may lead to more desirable jobs and working condition. For example, older black

workers are “more likely to have lower levels of education and fewer economic resources

27

and tend to work in more unstable industries than white workers” (Flippen & Tienda,
2000, p6). It is because blacks are disproportionally concentrated on more physical-
demanding jobs such as mining, transportation, construction, and manufacturing.

Baldwin (1999) demonstrated the effects of different types of impairments on
employment and wages among low income older workers. Her ﬁndings show that
persons with mental conditions such as mental disorders, emotional problems, mental
retardation, alcohol/ drug problems, or learning disabilities have the poorest labor market
outcomes (low employment rates and wages) compared with those experiencing physical
disorders. If older workers rarely have other sources of incomes such as pensions or
savings for retirement, the role of Social Security beneﬁts is the most important.
According to Burtica (2008), Social Security provides at least three-quarters of total
income for low-income and poor elderly.

Further, several studies have demonstrated that job characteristics affect the
ability of older workers to remain in the workforce. For example, a number of studies
show older workers reaching in retirement ages are lower for individuals who work in
physically demanding occupations (Hayward &Grady, 1990; Gustman & Steinmeier,
1986). Job characteristics are different by gender; women are concentrated in industries
and occupations that tend to receive lower Social Security and pension beneﬁts than men
(Wray, 2001). Related to the race on job characteristics, blacks and other minority older
workers typically are less likely to work in professions or jobs with high beneﬁts. They
have less education, fewer skilled jobs, lower salaries, and long periods of unemployment
(Bacon, 1995).

Many prior studies determined the various factors affecting retirement decisions

28

among older workers (Bryant, 1990; Hatcher, 2002; DeVaney & Chien, 2001; DeVaney
& Su, 1997; Hatcher, 2002; Montalto, Yuh, & Hanna, 2000; Holtmann, Ullmann,

F ronstin, & Longino, 1994; Honig, 1996). First, ﬁnancial resources are important factors
affecting retirement decision. Older workers with more ﬁnancial resources will be more
likely to exit the labor force early based on the previous studies. Bryant (1990) and
Hatcher (2002) found that there is a positive relationship between higher level of assets
and a lower level of debt and the demand for leisure time which means to decrease the
number of years of work. Second, household assets are undoubtedly hypothesized that
these have an inﬂuence on the decision to fully retire (F lippen & Tienda, 2000). Third,
income from a career position has been recognized as having a negative effect on the
expectation of retirement (Flippen & Tienda, 2000). Previous studies revealed that a high
wage rate or more proﬁt from work induces the worker to remain in their career longer
(Bryant, 1990; Holtmann, Ullmann, Fronstin, & Longino, 1994; Honig, 1996). Income is
indirectly related to the decision of retirement since income is related to the accumulation
of household assets and more household assets itself are likely to have an encouraging
retirement (Bryant, 1990). Income itself is also a direct relationship with continuing to
work and to delay retirement (Bryant, 1990; Honig, 1996). For example, retirement
income is based on Social Security beneﬁts, pensions, and earning from ﬁnancial assets.
If older workers have some forms of retirement income, they are more likely to feel
conﬁdence about their retirement (DeVaney & Chien, 2001; DeVaney & Su, 1997;
Fronstin, 1999; Quinn et al., 1998).

The Age Discrimination in Employment Act (ADEA) and American Disability Act (ADA)

“Overriding all other issues is ageism (that is, age bias), which is the most

29

fastest growing category of discrimination cases ﬁled with the US. Equal Employment
Opportunity Commission (EEOC). They also make up the largest percentage of all
charges ﬁled (Morton, Foster, & Sedlar, 2005). The prohibition against discrimination in
discharging employees also implies increased employment of older workers due to
reduced involuntary terminations (Stock & Beegle, 2004). There have been numerous
studies on the empirical impact of age discrimination laws on labor market outcomes. For
example, Neumark and Stock (1999), using US. Census data, reported that age
discrimination prohibitions increase employment of workers age 60 and over.

Similarly, Adams (2000), using the Current Population Survey, found that age
discrimination prohibitions have negative effects on the probability that a worker age 65
or older would have a new job. He attributed this to the reduced incident of retirement for
this age group because of the ADEA. Other studies have been conducted to examine
consequent problems with the ADEA. Bendick, Brown, and Wall (1999) and Lahey
(2005) suggested that employers discriminate against older workers in the job application
process. After they are on the job, age discrimination is common and impacts job‘security,
deployment, retention, and promotion of older workers (Goldman, Gutek, Stein, & Lewis,

2006). In a study that focused on the ADEA’s mandate to eliminate mandatory retirement,

30

Ashenfelter and Card (2000) found that this provision for professors was associated with
reductions in retirement. Consequently, ineffectively managing the aging workforce and a
non-accommodating environment may contribute to claims of discrimination (Bjelland,
Bruyere, Schrader, Houtenville, Ruiz-Quintanilla, & Webber, 2009).

Compared to age discrimination, fewer studies have been conducted regarding
disability employment discrimination among older workers (Goldman, et al., 2006). The
Americans with Disabilities Act (ADA) is also applied when older workers experience
any type of disabilities. The ADA enacted in 1990 mainly to increase the employment
rate of people with disabilities. The act made it illegal to discriminate against an
individual with a disability. The employment provisions of the Americans with
Disabilities Act of 1990 (ADA) came into effect for employers of 15 or more employees
and require to make “reasonable” accommodations to allow qualiﬁed workers with
disabilities to participate in the workforce (Blanck, 1996). The act has also been helpful
in providing access to jobs, especially in the small business sector, which has created
two-thirds of all new jobs since the early 19705 (Bruyere, 2000). According to Bjelland et
al., (2009), when examining charges made by those protected under both the ADA and
ADEA, results from a logit model indicated that compared to charges ﬁled under the
ADA alone, jointly-ﬁled ADA/ADEA charges were more likely to be ﬁled by older
individuals, by those who perceive discrimination in hiring and termination, and to
originate from within the smallest ﬁrms.

Older Workers and Vocational Rehabilitation Professionals
Szymanski (1985) deﬁned rehabilitation counseling as “a profession that assists

persons with disabilities in adapting to the environment, assists environments in

31

accommodating the needs of the individual, and works toward full participation of
persons with disabilities in all aspects of society, especially work” (p.3). The aim of
vocational rehabilitation is to retain and reintegrate individuals with reduced work
capacity as a result of physical, mental and/or social circumstances on the labor market
(Gupta, Nabanita, & Larsen, 2007). Another role of rehabilitation is that “the negative
human consequences of work injuries and subsequent unemployment compel
rehabilitation counseling professionals to learn how to assist employers in the adoption of
company practices that can prevent injury and maintain employment when disability
occurs” (Habeck, Leahy, Hunt, Chan, & Welch, 1991, p. 213).

Older workers are likely to acquire a disability due to the aging process (Kraus &
Stoddard, 1991; McNeil, 1997; Bruyere, 2006). If older workers with disability choose to
remain in the workforce, the need for career planning will increase for this population
because they also need to adjust their disability as well. Without doubt, there is a current
and continued need to address the employment concerns of older workers with disability
as they continue to participate in the workforce. If older workers experience deteriorated
health or disability, they change their jobs many times over several years from the onset
of their poor health since changing jobs (job transition) is an important way for older
workers to adapt to their situation (Bound, et al., 1998).

Dwyer and Mitchell (1999) also supported the notion that if older workers
experience poor health or disability, they have shown a positive inﬂuence on labor force
participation rather than negative withdrawal. People’s lower earnings, associated with
poor health, may induce an increase in the need for work to replace lost income, to

maintain health insurance, or to pay for the additional health services consumed (Dwyer

32

& Mitchell, 1999). Gupta, Nabanita and Larsen (2007) also stateed that “vocational
rehabilitation might be a pathway to retirement in those cases where there is a return to
the labor market” (p 211). Rehabilitation counselors should effectively facilitate older
workers with disabilities in need of employment.

Aging Population in Rehabilitation Counseling

Over the past two decades, relatively few articles have been published in
rehabilitation journals aimed at the issues of the aging and their disabilities. The focus of
the studies for a member of rehabilitation practitioners has been speciﬁed to the need of
older individuals with disabilities (Dixon, Richard, & Rollins, 2003). The subjects are on
(1) identifying the future research, training, and service directions for geriatric
rehabilitation (Kemp, Brummel-Smith, & Plowman, 1989); (2) aging bias and ageism and
rehabilitation counseling (Bearden & Head, 1986; Benedict & Ganikos, 1981; Reiner,
1974; Saxton & Spitznagel, 1992), (3) creating an integrated agenda for the aging and
disability (Zola, 1989), and (4) the factor of aging among persons with disabilities (Lewis,
1989).

At the same time there were large volumes of studies under disability
management in the rehabilitation counseling journals (Habeck & Kirchner, 1999; Shaw &
Betters, 2004; Rosenthal, Hursh, Lui, lsom, & Sasson, 2007). The concern was for the
“increasing cost of medical care and workers’ compensation insurance, and the rise of
private (insurance) rehabilitation in the mid-1970’s (Strensrud & Gilbride, 2004, p. 222).
During the 1980’s, employers, insurers, policy makers, and service providers began to
recognize the negative impact of employees’ health care costs, the aging workforce, and

the increasing occurrence of disability in the workplace (Galvin, 1991). Thus, private

33

sector rehabilitation counseling quickly initiated the concept of disability management
and emphasized cost-effectiveness (Shaw & Betters, 2004). As a result, these initial
disability management programs began to show a positive return on investment and
resulted in ﬁnancial savings and positive outcomes for injured workers who returned to
work (Habeck & Kirchner, 1999).

More recently, the issues on older consumers in rehabilitation studies have
developed. This is wider areas than before shown in writings on (1) the phenomenal
growth of the aging population and their late onset disability (Finch, & Robinson, 2003);
(2) the collaborated approaches for aging population with multiple services needs (Harley,
Donnel, & Rainey, 2003); (3) identifying a number of challenges and concerns facing
rehabilitation counselors in terms of ageism, legislation, and employment (Dixon et al.,
2003) and (4) the characteristics and outcome of vocational rehabilitation services for
older adults (Wadsworth & Kampfe, 2004; Drebing, Losardo, Ormer, Krebs, Penk,
Nasser, Ray et al. 2002).

Case Management in Rehabilitation Counseling

Disability management is defined by Shrey & Lacerte (1995), who state that
disability management is “an active process of minimizing the impact of an impairment
(resulting from injury, illness, or disease) on the individual’s capacity to participate
competitively in the work environment (p. 5)” regardless of age. Historically, the concept
of disability management has been employed by various parties (e.g., medical providers,
claims managers, insurance carriers, and rehabilitation professionals) with their oWn
purposes or interests. Thus, the term disability management has been broadly deﬁned and

sometimes treated as the most blurred terms in health care (Shrey & Lacerte, 1995). In

34

the rehabilitation counseling ﬁelds, the importance of disability management for the
needs of older workers was reiterated by Bruyere (2006). Bruyere (2006) stated that “the
primary emphasis of disability management in the employment setting is preventing or
minimizing the impact of the disability on the employer and employee and assisting in
job retention (p. 155).” There is a signiﬁcant need for employers to ﬁnd proven strategies
to return employees to work after disability (Vargo & Grzanowicz, 2002) due to the
rising medical costs associated with disability that challenges employers. As a result,
more employers are seeking strategies to return employees to work and reduce their costs
related to disability (Gottlieb, Vandergoot, & Lutsky, 1991). Disability management
strategy provided by rehabilitation counselors can be the one fulﬁlling the employers’
needs. Especially in an aging society, this strategy is able to assist in assuring the
productivity of workers during all phases of life (National Institute of Disability
Management and Research, 2005). If society or business ﬁelds are struggling with an
aging workforce and their disability, professions like rehabilitation counseling directly
become as a disability management professions for them (National Institute of Disability
Management and Research, 2005).

Employment Nondiscrimination Legislation

The culture of the workplace itself is a signiﬁcant factor inﬂuencing the retention
or rejection of older workers. Age-based stereotyping perpetuates discriminatory
practices and discourages older workers from remaining in or returning to the workplace.
Thus, rehabilitation counseling professionals should be knowledgeable about
employment nondiscrimination legislation (Finch & Robinson, 2003) because the nature

of the workforce and of older workers with disabilities are too often seen as a problem to

35

be dealt with, rather than a workforce opportunity (Morton, et al., 2005). In order to
create more opportunities with these groups, it is crucial that rehabilitation professions
consider how to both minimize discrimination toward and maximize participation of
older workers with disabilities.

Rehabilitation professionals also need to be aware of the possible attendant
mental health considerations in the aging process (Swett & Bishop, 2003), and to keep in
view the aging of people with existing disabilities (Larkin, Alston, Middleton, & Wilson
2003). Increasing retraining requires knowledge of speciﬁc usable concerns and
adaptations to technology to ensure effective functioning.

Training Programs for Older Workers

For older workers training is an important component of retention (The
Commonwealth Fund, 1993) or of a return to work. Given the current nature of work,
which involves more brain than physical strength, the need for continuous skill updating
in light of rapid technological changes make it possible for older workers to continue on
the job (Poulos & Nightingale, 1997).

Today, older workers may receive training through both private companies
(employers) and publicly funded programs. However, most private companies or
employers seem to have little attention given to retraining for older workers because the
U. S. General Accounting Ofﬁce (2001) report that they “simply [have] not considered”
(p 23) doing so. Employers have shown a lack of willingness to train their employees.
According to a survey of training in establishments of 50 or more employees conducted
by the US. Department of Labor in December 1996, older workers reported substantially

fewer hours of training than younger workers aged between 25 and 44. This report also

36

mentioned that generally workers in larger ﬁrms receive more training than those in
smaller ﬁrms.

The public training programs for older workers are also limited. The key of public
job training programs in the United States is the Workforce Investment Act (WIA) of
1998 (United States Government Accountability Office, 2001). Poulos and Nightingale
(1997) mentioned that “WIA has eliminated speciﬁc targeting for certain populations,
including older and disadvantaged workers, and instead encourages state and local
Workforce Investment Boards to develop integrated systems and one-stop services that
best meet the needs of their areas” (p 1). This new law “also represents a shift in
emphasis to incumbent worker training, including retention, skill upgrading, and

concurrent education” (Poulos & Nightingale, 1997, p l).

The US. General Accounting Office (2003) reported that “about 12 percent of the
1.3 million older pe0ple who were not working and wanted a job were enrolled in these
programs between July 2000 and June 2001” (p3). More than two thirds (156,000) ofthe
participants receive some form of aid from the Senior Community Service Employment
Program (SCSEP) which is developed for low income persons 55 and older. These low
income participants were typically placed in subsidized minimum-wage community
service jobs (Rix, 2004). Carthell and Flemming (1990) stated that this program “places
older workers in part-time public sector jobs at senior citizen centers, in schools,
hospitals, libraries, social service projects, and other community projects” (p. 13).

Rix (2004) however argued that this program “does not seem suited to meeting
the training needs of large numbers of better educated boomers” (p 29) since “this small

but very popular program, which serves about 100,000 persons a year, was designed to

37

assist economically disadvantaged persons aged 55 and older” (p 29). Poulos and
Nightingale (1997) reported that “older workers with high education or technical skills
would probably continue to be in demand as long as they maintain or upgrade their skills
as appropriate” (p 31). Still, there are no public services for older workers with high
education or technical skills. Rix (2004) also concludes that “[public training] resources
have never been sufficient to reach more than a small proportion of potentially eligible
[older] workers” (p 29).

If the participants have a disability, SCSEP is working with vocational
rehabilitation to enroll and place persons with a disability (Michigan Office of Services to
the Aging (2004). However, speciﬁc strategies, interventions, and outcomes for older
workers with disabilities are not addressed by these reports. The SCSEP program may
have much potential to serve elderly people with disabilities by expanding the service
areas and developing some specialties. In order to better serve the aging population with
disabilities, collaborative strategies including SCSEP, vocational rehabilitation, and the
JTPA Adult Training should be promoted among service systems. Potential vocational
planning with practical job goals can become a major task for those who desire to get a
job because they may be unfamiliar with the current labor market and contemporary job

seeking skills (Finch & Robinson, 2003). Thus, alternative programming should be

needed in the future.

38

CHAPTER 3
METHODOLOGY

The purposes of this study were: 1) to estimate the proportion of late onset
disability among the older workers with full time job; 2) to identify the characteristics of
the older worker with disability; 3) to examine the changes in demographic, household,
and labor market characteristics due to the late onset of disability within employed and
non-employed workers; and 4) to determine the effects of disability on employment
transitions of those who (a) continuing to work full time (b) change to work part time,
and (c) change to non-employed.

Keeping in mind the purpose of the current research, the following discussion
provides two issues of deﬁning and measuring disability, and the overview of the Health
and Retirement Study (HRS). Then, the study’s overall design, sampling, dependent and
independent variables, procedures and data analyses will be addressed.

Deﬁning and Measuring Disability

Deﬁning of disability is imperative for this study because of the variation of
measuring the multi-dimensional natures of disability and health limitation (Loprest, et
al., 1995). Historically, deﬁnitions of disabilities vary. There have been numerous ways
to deﬁne a sample of disabilities for the purpose of the study (Oi & Andrews, 1992;
Loprest & Maag, 2001). Unquestionably, each deﬁnition includes a different group of
individuals as a sample (Loprest & Maag, 2001). For example, medical professionals

believe that the problem of disability is directly caused by disease or health conditions.

The sets of health conditions or impairments may be key indicators for disability in

39

medical studies. If some individual reports that he or she suffers from a circulatory
condition, then researchers in medical ﬁelds consider that he or she has a disability.

Studies developed from the point of view of industry on based on economic
concerns deﬁne disability from a different angle. Researchers refer disabilities to
recipients of disability beneﬁts rather than look at individuals’ health conditions or
impairments since they believe that “disability has large economic and social costs”
(Dunlop, Manheim, Yelin, Song, & Chang, 2003, p. 34). The simplest and most
commonly used question to establish the presence of disabilities in large surveys and
economic studies is the level of difﬁculty (or, on one scale, the level of ability) to
perform their job demands, or whether an impairment or health problem limits the kind or
amount of paid work for the respondent. This is a measure of work disability or
limitations. These kinds of questions may have more bias than questions that identify
disability, since there is no standard measurement on limitations of work. For example, in
the question of whether an individual has a disability that limits his/ her work, individuals
who all have the same level of physical limitation may respond differently because they
consider work differently depending on what tasks they are considering as work. The
relationship between this work disability measure and the labor force is a poor predictor
of the actual effect of the limitation on work (Loprest , et al., 1995).

Most criteria that current researchers use for selecting individuals with disability
have their own problems. The tendency of any current national survey is to record
respondents’ subjective self-assessment of their own physical capacity. If researChers
employ the deﬁnition of disability based on the medical model that uses sets of

conditions or impairments, serious problem of bias could result. In the study, neither the

40

presence of medical conditions was conﬁrmed, nor was any physical exams undertaken
to measure health conditions. Generally, the majority of studies, however, focus on
individuals with disability utilize the entire criteria of factors that include health
conditions, recipients of disability beneﬁts, and individuals who reported limitations on
work and limitations of daily activities. Researchers often choose some of these criteria
depending on the study purposes or the characteristics of the data set they used. It is
therefore imperative to have a suitable deﬁnition that “was relevant to ability to work
with suitable measurements, but could be implemented the different characteristics
among those who are working and not working” (Loprest & Maag, 2001, p. 3).

There are several objective health measures that may be more closely tied to the
functional capacity for work, such as indices of activities of daily living (ADL) and
instrumental activities of daily living (IADL), rather than using health characteristics,
such as health conditions or chronic illness or subjective question about work disability
for identifying a disability. Researchers in the disability ﬁeld often utilize such measures
to identify work disability among the disabled (Nagi, 1966).

Functional Disability and Mental Disability

Any empirical studies of disability must not ignore the possibility of a reverse
causal relationship between depression and functional disability. It is not only true that
“chronicity of depression has a large impact on physical decline over time” (Penninx,
Deeg, Eijk, Beekman, & Guralnik, 2000. p. 1), but also the loss of physical ability to
perform everyday self and home care tasks increases depressive symptoms and the risk of
clinical depression (Bruce, Seeman, Merrill, & Blazer 1994). Due to the reverse causal

relationship between depression and functional disability and its effect on labor

4]

participation, it is therefore reasonable to include older workers with depression as well
as ﬁrnctional disability for older workers with disability in emphasized studies.
Functional Disability
There remains the question of how to construct an index of disability
measurements. For the functional disability, this study deﬁnes functional disability to
some level of difﬁculty in performing at least one of a speciﬁc set of activities or the
inability to perform at least one of a set of functions among the ever worked full time
older workers who age over 50. This deﬁnition is useful since it includes those
individuals whose condition or impairment has manifested itself as a limitation in one of
a wide range of daily activities that likely impacts employment, and it involves asking
people about the inability or the lack of physical ability to perform various tasks.

It measures by assessing the respondent’s difﬁculty in performing activities of
daily living (ADLs) and the household management level of activities (Instrumental
Activities of Daily Living, IADLs). Basic Activities of Daily Living (ADL) range from
the ability for dressing, eating, bathing to the ability to move around indoors (e.g., “Does
anyone ever help you bathe or shower?”; yes/no). Instrumental Activities of Daily Living
(IADL) include shopping, cooking a meal, balancing a checkbook/paying bills, and
driving) (e.g., “Are you able to sh0p for groceries without help?” : yes/no). Thus, in the
current investigation, the standard measures of both (ADLs) and (IADLs) were used
since these are the most commonly used measures in the literatures (Zwerling, Whitten,
Sprince, Davis, Wallace, Blanck, & Herringa 2002; Ho, 2002). Furthermore, Fried,
Herdman, Kuhn, Rubin, & Turano (1991) suggest that ADL and IADL limitations

represent a fairly advanced stage of functional decline.

42

Mental Disability

The most common and debilitating form of mental illness among the aging
population is depression or depressive disorder (McKenna, Michaud, Murray, & Marks,
2005). Due to its prevalence and its association with physical disabilities, depression in
the older population is a major concern for the disability studies. For this current study,
the Center for Epidemiologic Studies Depression (CES-D) scale is useful For the
measuring depression (Radloff, 1997). The abbreviated eight-item CES-D version in the

HRS is a self-reported inventory of depressive symptoms, and it includes test items such

3’ £6 ,3 66

as “was depressed,” “everything was an effort, sleep was restless, was happy,” “felt

’9 ‘6

lonely, enjoyed life,” “felt sad,” and “could not get going” and their symptoms that
occurred in the week prior to the respondent’s interview date. This test has shown a high
level of reliability and validity as well as high internal consistency comparable to that of
the widely used and validated twenty-item CES-D version (Lyness, Noel, Cox, King,
Conwell, & Caine, 1997; Turvey, Wallace, & Herzog, 1999; Stefﬁck, 2000; Wallace,
Herzog, Ofstedal, Stefﬁck, Fonda, & Langa, 2000).

Identiﬁcation of individuals who are suffering from a depressive episode is
predicated on the reliability and validity of CES-D cutoff scores. In order to best identify
individuals using this score, this researcher relied on an extensive literature on the
usefulness of the CES-D in identifying episodes of depression (Lyness et al., 1997;
Kohout, Berkman, Evans, & Comoni-Huntley, 1993). Several methods have been
advanced to convert a cutoff score consistent with probable depression in the full survey

(Unutzer, Patrick, Simon, Grembowski, Walker, & Katon, 1997; Turvey, Wallace, &

Herzog, 1999; Emptage, Sturm, & Robinson, 2005; Doshi, et a1, 2007). Some studies

43

employed a cut-off point of 3 or more for deﬁning the present of active depression since
these studies has been found to have a sensitivity of 71 percent and speciﬁcity of 79
percent for major depression (Turvey, Wallace, & Herzog, 1999; Emptage, Sturm, &
Robinson, 2005; Doshi, et al, 2007). However, it is still not clear that a cut-off point of 3
or more is severe enough to be classiﬁed as major depression.

A study by Lyness et al., (1997) investigated the validity of the full CES-D for
recognizing depression in older community dwelling populations, aged 55 and older.
They report that the optimal cut-off score for identifying major depression in an older
adult population using the full CES-D to be 20-21 out of a possible total score of 60.
Because a score of 20—21 on the full CES- D survey is equivalent to a cut-off score of 4 in
the shortened version, this study, therefore, employs the low cutoff score (24). It is set to
achieve a higher rate of detection of depression in older workers.

The Classiﬁcation of Timing

Like the relationship between functional disability and depression, studies in
disability and labor supply also mention that there is the possibility of a reverse causal
relationship. In other words, health not only affects labor supply, but labor supply may
affect health. Generally, deterioration in health or having a disability would lower
productivity and thus reduce the probability of being employed based on the economic
rationale (Doshi, Cen, & Polsky, 2008). Conversely, labor activities may have a direct
impact on the health of the individual. For example, labor activities may have a negative

impact on health through the stress or hazards associated with certain working
environments, but the lack of activity during non-labor participation may result in the

deterioration of health/ disability (Sickles & Taubman 1986; Stern 1989). This latter

44

possibility may be especially relevant for retired workers. For example, non-labor
participation itself may result in disability (Szinovacz & Davey 2004).

Like the relationship between labor activities and health deterioration, the
classiﬁcation of timing (before and after disability) can therefore be an essential aspect of
an empirical study of the causal relationship between independent variables (socio
demographic, labor market, and ﬁnancial factors) and labor participation patterns due to
disability. The classiﬁcation of timing was chosen for this study to emphasize captured
acquired disability status for the study purposes. Therefore, it can be ensured that
disability event captured by timing may explain the changed job transition considering
that the experience of disability may explain how disability may change for older
workers’ participation patterns in the labor force.

In addition, this empirical model this study presented was based on the extensive
labor economic literatures to control for economic factors such as Social Security
Retirement income that may potentially bias the relationship between labor participation
and disability (Gruber & Wise, 2004; Rogowski & Karoly, 2000; Doshi et al., 2007).

Overview of HRS Data
The data used in this study was obtained from the Health and Retirement Study
(HRS) that provides the most nationally comprehensive representative longitudinal
survey on health and disability for aging individuals over the age of 50 and their families.
It is based on core biennial interviews of nearly 20,000 individuals representing the US.
population since 1998. In general, the goal of the HRS is to investigate the impact of
health transitions on employment outcomes and personal ﬁnancial transfer of income and

assets among those over 50 years old.

45

Briefly, the HRS is a multistage probability cohort sample of US. households.
The HRS is taken from community-dwelling respondents who were 51 years or older,
and also contains spouses/ partners who are in various ages. The survey population
therefore includes individuals who are not themselves in the age eligible, age over 50.
More complete details on the HRS design and methods were published (Heeringa &
Connor, 1996). The current version of HRS (RAND data) is representative of all birth
cohorts born in 1953 or earlier and is collected at two-year intervals. The beneﬁts of
using the RAND HRS data is that it is an extraordinarily user friendly ﬁle. It contains
several important components: (1) a wide-ranging set of derived variables from the ﬁrst
to the most recent data; (2) consistent depression and disease reports over the survey
period; (3) imputations of wealth, income, and medical expenditures, and (4) an intuitive
and consistent naming scheme and values labels that have been assigned throughout in
order to compare variables across survey waves (RAND, 2006).

The HRS is managed through a cooperative agreement (N IA U01AG009740)
betweenthe National Institute on Aging (N IA) that provides its primary funding, and the
Institute for Social Research (ISR), which administrate and conduct the survey. The HRS
has shown several strengths compared to the other national surveys. The Current
P0pulation survey (CPS) provides detailed data on work disabilities in a long series of
cross sections (Loprest, et al., 1995) and The National Health Interview Survey (NHIS)
contains detailed information on disability but NHIS data source did not pay speciﬁc
attention to retirement age (L0prest, et al., 1995). However, the HRS has comprehensive
information is available on various dimensions of disabilities (functional limitations,

mental limitations and health impairments) and explicitly pay attention on aging

46

population. In addition, comprehensive information of labor market variables that include
details in employment status and industrial data is available in the HRS.

In addition, the HRS provides comprehensive information about assets other than
pensions and social security (Juster, 1998). It is known for the high quality measurement
of income and wealth matter (J uster, Smith, & Stafford, 1999). For example, a very
comprehensive and detailed set of questions was used to measure household wealth.
Questions were included in each core interview on housing assets and were separated into
the following eleven categories, (a) other real estate, (b) vehicles, (c) business equity, (d)
IRA or Keogh, (e) stocks or mutual funds, (1) checking savings or money market funds,
(g) CD'S, (h) government savings bonds or treasury bills, (i) other bonds, (j) other assets,
and (k) other debt.

The HRS uniquely deals with the typical high rates of non-response or refusal to
answer the questions about the value of stock holdings, IRA balances or amounts found
in other narrow asset categories. The special approach for dealing with non-response or
refusal in ﬁnancial questions is to use a series of bracketed questions whenever an open
ended question about the value of an asset is not answered. This approach greatly
increases reporting wealth and reduces item non-response or refusal (Moon & Juster,
1995)

Research Design

The present study employed a descriptive, ex post facto research design.
According to Leedy and Ormarod (2001), the main uses of descriptive quantitative

research have either obtaining a general overview of the characteristics of an observed

phenomenon, or exploring possible correlation among two or more phenomena. Thus,

47

descriptive research design is a valid method for this study.

The ex post facto design means that researcher starts with the observation of the
independent variable(s) and a dependent varibe(s) in which have already occurred. With
no manipulation and lack of random assignment, this type of research also uses the same
types of data as do descriptive studies, but it is the different in that it seeks to generate or
explore hypotheses about causes (Hadly & Mitchell, 1997; Heppner, Kivlinghan, &
Wampold, 2008). Even though the ex post facto design is much less able to determine
causality than true experiments, this method is important for either ethical reasons, or an

interest in organismic variables. In this study, all independent variables were not able to
be manipulated as they had already occurred. Moreover, this study explored possible
factors which can account for the labor transition of the dependent variables. Therefore,
ex post facto design was suitable for the present study.
Sample Selection
The HRS data allowed this study to examine the effect of disability on the older
respondent’s job transitions from 1998 through 2006. All analyses for this current study
used person-weights and sampling error codes for the 2006 HRS data developed at the
University of Michigan to provide valid inferences to account for the sample design in
the HRS (Heeringa & Connor, 1995, RAND, 2006). Respondents with sampling weights
of zero indicated they were institutionalized, non-age eligible or deceased. Thus, these
cases were excluded from this study. Then, only those persons who met the criteria in
Phase I and II were repeatedly pooled: between 1998 survey and 2000 survey, 2000
survey and 2002 survey, 2002 survey and 2004 survey or 2004 survey and 2006 survey to

achieve a sufﬁcient number of cases for the analyses in this investigation. As mentioned,

48

the time when the respondent was a free of disability referred as Time I. The time when
the respondent has disability after the previous survey referred. as Time II for this current
study.

As Figure 2 illustrates, there are steps in following rules applied in order to be
eligible for inclusion in the sample: in the Time I (before disability) as baseline, (1)
individuals were age eligible (age over 50) at the time of the HRS survey interview (2)
the age eligible individuals were currently employed at least 35 hours per week or 36
weeks per year (full time job) and (3) the age eligible full time workers should have a
good healthy condition at the baseline survey (year of 1998, 2000, 2002, and 2004) since
this study wanted to look at the labor transition patterns due to the experience of
disability. Respondents who reported that they were non-employed or retired at the time
of baseline (Time I) were excluded because this population may be underemployed or
already in post retirement (Szinovacz & Davery, 2005). In this study, full time workers in
Time I was the single value in the employment status. In the Time II, the samples for this
study were additionally restricted: individuals who met criteria in Time I were self-

deﬁned as ‘functional’ or ‘mental’ disabilities at the subsequent survey (2000, 2002, 2004

and 2006- Time II).

49

 

 

 

 

J l l l g
l I I I
1998 2000 2002 2004 2006
6
Healthy ------------------ Disabled
(Job 1) (Job 2)
I v
Healthy -------------------- Disabled
(Job 1) (Job 2)
V
Healthy Disabled
(Job 1) (Job 2)
‘7
Healthy -------------- Disabled
(Job 1) (Job 2)

 

 

 

 

 

 

Figure 2. Sampling Process

* Baseline (Jobl): The time (Time I) when individuals who are age over 50 and
currently working reported that they were a free of disability.
* Job 2: The time (Time II) when individuals in Time I reported the experience of an

acquired disability in the subsequent survey.

The ﬁnal samples for this study were the cases who met criteria in both Time I

and II. The current working status (labor force status) was therefore the main variable to

determine employment status (full time, part time, or non-employee). Respondents who

reported non-employed, applied for the disability beneﬁts, retired, and exited the labor

force at the time of experiencing disability (Time II), were referred to as non-employed.

50

Dependent and Independent and Variables

Dependent variable

As shown in Table 1, two categories of employment status
(employed/unemployed) were used as dependent variables in the bivariate descriptive and
binomial logistic regression analyses in this study. To examine the effect of disability on
detailed employment transitions among older workers, this researcher examined three
alternative transitions between two surveys: whether they are employed full time, part
time, or are not employed. These categories are deﬁned to be mutually exclusive.

Unlike the other labor studies, current working status (all individuals who are
currently employed regardless of their previous retirement history) was therefore the
main variable to determine employment or non-employment status. This current study
does not require older workers to move out of long-term jobs or into long—term jobs for
the transition to qualify as a career change. Many older workers who change occupations,
in fact, were not in long-term career jobs before they moved to a different occupation
(Johnson & Kawachi, 2007). The probability of employment status was based on the
response to this question in the HRS “Are you currently working?” and “What is your
current labor force status?”. In cases where respondents classiﬁed themselves as retired,
but are still working for pay at the time of survey were classiﬁed as employed. The cases
where respondents are currently working for pay regardless of whether they have
continued to work or have returned to work status were also classiﬁed as employed.

The dependent variable, labor force status, consist of seven mutually exclusive
categories: (1) working full-time (2) working part-time (3) unemployed (4) partly retired

(5) retired (6) disabled (7) not in the labor force. In this study, labor status was grouped in

51

the following three categories: ( 1) working full time, (2) working part-time, and (3) non-
employed. Since he/she is working part-time and mentions retirement before, the HRS set

this case to partly retired. Thus, working part-time includes the cases of partial retirement.

Table 1. Variables for the Model

 

Criterion and Predictor Variables HRS Variables

Continued be Employment

Criterion Variables Full Time Work

Part Time Work

Non- Employed

Age

Demographic Variables Gender

Race/ Ethnicity

Level of Education

Marital Status

Types of Disability

Predictor Slumber of People living in the
. ousehold

Variables

Labor Market Variables Self-Employment. Status

Types of Occupation

Types of Industry

Individual Earnings from Work

Financial Variables Amount of Household Asset

Social Security Retirement Income
(at Time II)

 

 

 

 

 

Independent Variables

The independent variables were chosen based on the theoretical model and
previous studies. A set of independent variables is used to predict changes in adults’ labor
force status. Independent variables are grouped as three potential predictors of the
dependent variables. The first category considered is demographic variables, such as age,
gender, educational attainment, race, and marital status as essential variables. Age is

measured in years up to 90 in the HRS. Gender was a dichotomous variable. Regarding

52

education level, the HRS collects information from respondents the highest grade or year
of school they completed. In this study, education level was re-coded into three
categories: 1) less than high school (up to grade 11); 2) high school graduates (grade 12
or GED); and 3) more than high school (some college or technical school or college 4
years or more). The HRS asked respondents whether the respondent is Hispanic (p. 111).
The HRS measured the respondents’ multiple races by allowing them to choosing all
categories that apply to them and also asks what race the respondents think they belong to
through this question: “Do you consider yourself primarily [race responses separated by
‘or’]?” (p. 109). Respondents were provided the following categories for the last two
questions: White/Caucasian, Black/African American and Other. Race in this study was
grouped in the following four categories: 1) non-Hispanic white; 2) non-Hispanic African
American; 3) Hispanic; 4) non-Hispanic other race or multiracial. Marital status was a
categorical variable consisting of three groups that reﬂects overall status (i.e., married or
widowed, separated or divorced, and never married). The marital status was collapsed
into two categories: I) married and 2) unmarried including divorced, widowed, separated
and never married.

The second set of variables were more likely to have a strong probability of
association with workforce participation consisting of the number of people living in the
household (including number of children) and the types of disability based on other
studies. The third set of variables was related to the labor market: self employment, types
of occupation (blue collar vs. white collar), and types of industry (agriculture/ mining!
construction/manufacturing/ transportation, wholesale/ retail/ services, vs. ﬁnancial/

insurance/ real estate/ professional related services/ public administration) on the current

53

job. Self employment status was a dichotomous variable. Blue collar occupations
included farming, forestry, and ﬁshing, production, operations, and the military. White
collar occupations include managerial and professional, sales, clerical, administrative,
and service occupations. The current types of industry were collapsed into tree categories:
1) agriculture/ mining/ construction/manufacturing transportation, 2) wholesale/ retail/
services, and 3) ﬁnancial/ insurance/ real estate/ professional related services/ public
administration.

Financial variables were three for this study: (1) individual earnings, (2) amount
of household net asset, and (3) Social Security income. Labor income was a continuous
variable and it included the respondent’s annual salary, bonuses, overtime pay,
commissions, tips, second job or military reserve earnings, professional practice or trade
income and income from business, if there was a business.

Net worth includes the value of all real and ﬁnancial assets, including equity in
primary and secondary homes, but it excludes the value of future Social Security and
Disability Beneﬁt pension rights. Social Security Retirement income which was a
continuous variable was the respondent's income from Social Security retirement, Spouse
or widow beneﬁts. The variables in ﬁnancial assets analyzed in this study relied on
imputed data since the HRS used a series of unfolding bracket follow-up questions for

non-response to wealth questions. The imputed ﬁnancial data produced signiﬁcantly
lower levels of non-response and more reliably imputed values (Smith, 1995).
The ﬁrst three categories including basic demographic, work related variables,
and labor market variables (self employed, occupation, and industry) were included in the

empirical model as preference shifters. All of the predictor variables are from baseline

54

interviews, except for the social security retirement income variable.
Data Analysis

The present study was designed for four purposes: 1) to estimate the proportion of
late onset disability among the older workers with full time job; 2) to identify the
characteristics of older workers with a disability; 3) to examine the changes of
demographic, household, and labor market characteristics due to a late onset of disability
within two groups: the employed and non-employed; and 4) to determine the effects of
disability on employment transitions: (a) continuing to work full time job (b) changing to
work part time job; and (c) changing to non-employed. There are three types of analyses
this study used: descriptive, bivariate analyses, binomial logistic regression, and
multinomial logistic regression.

Descriptive and Bivariate Analyses

For research questions 1 “What is the likelihood of older workers who have
acquired disabilities (a health change/health shock) in their later lives?” and question 2
“What are the demographics or workforce characteristics of older workers who have
acquired disabilities in later life?” descriptive statistics, and frequencies. For research
question 3 “How do employed and non-employed older workers differ in demographic,
workforce, and ﬁnancial characteristics?” bivariate analysis and binary logistic regression
were run for an overview of the older workers with disability.

Binary Logistic Regression

For this study, binary logistic regression and multinomial logistic regression were
employed. For the question 3 “How do employed and non—employed older workers differ

in demographic, workforce, and ﬁnancial characteristics?” binary logistic regression

55

analysis was also appropriate since binary logistic regression analysis has been
acknowledged as the data analytic tool of choice when criterion variables are
dichotomous and predictor variables are continuous or categorical (Agresti & Finlay,
1997; Tabachnick & F idell, 2007).

Similar to multiple regressions, logistic regression investigates the association
between one criterion variable and multiple predictor variables. In logistic regression,
however, the relationship between one categorical criterion variable and a set of predictor
variables is examined. The categorical criterion variable can be dichotomous,
polychotomous, or ordinal. In interpreting results, binary logistic regression emphasizes
the probability of a particular outcome for each case by one or more independent
variables. Thus, in this study, binary logistic regression was employed to predict the
probability of job transition pattern among older workers with disability. Binary logistic
regression model this study presented was used to assess the strength and magnitude of
associations between a dichotomous employment outcome (employed or non-employed)
and a set of independent variables.

The overall goodness of fit of the models was assessed on the basis of the
Hosmer-Lemeshow Chi-square test (Hosmer & Lemeshow, 2000). The Likelihood-ratio
test was used to determine how the ﬁt of the models change as additional variables are
introduced to equations. The effect of the independent variables on employment
outcomes was evaluated on the basis of the corresponding odd ratios and the signiﬁcance
level of the estimated parameters provided by the Wald statistic. The results of Wald

statistic was expressed by p-value in this study. The logistic models are estimated using

SPSS 18.0.

56

Multinomial logistic regression analysis

For the questions question 4 “What factors determine the older worker’s labor
transition?,” multinomial logistic regression was used because the dependent variable has
more than two nominal, unordered categories. If observations are independent, the
multinomial or polychotomous logistic regression model can be used to assess the
inﬂuence of explanatory variables on the nominal response variable (Agresti, 1990).

An advantage of this technique is that it is suitable to evaluate the determinants of
various demographic, job characteristics, and ﬁnancial factors on the likelihood of three
types of employment outcomes (full-time, part-time, vs. non-employed outcome) for the
older workers with disability. The goodness of ﬁt of the models was evaluated on the
basis of Pearson’s Chi-square and of the Deviance tests (Chan, 2005). In multinomial
logistic regression models, the logistic coefficients (log odds) and the exponential form of
the coefﬁcients (odds ratio) was estimated using Maximum Likelihood Estimation (MLE)
and presented for comparison. The likelihood ratio test was used to check the model’s
goodness of ﬁt. The Multinomial logistic regression models were estimated using SPSS

18.0. All explanatory variables in this model are measured at the time of the baseline

interviews except for the Social Security Retirement income variable.

57

CAHPTER 4
RESULTS

This chapter presents the ﬁndings of the data analyses in order to investigate the
effect of disability on employment patterns among older workers. All results were based
on the following purposes of this study: 1) to estimate the likelihood of late onset
disability among the older workers with full time job; 2) to identify the characteristics of
older workers with a disability; 3) to examine the relationship between demographic,
household, and labor market characteristics due to a late onset of disability and labor
force behaviors among full-time older workers; and 4) to determine the factors affecting

employment transitions due to disability including (a) keeping a full-time job, (b)
changing to a part-time job, and (c) changing to be non-employed.

For the ﬁrst and second purposes, descriptive statistics for all variables were
employed, followed by the bivariate analyses, binominal logistic regression and multi-
nominal logistic regression. Bivariate analyses were used to investigate how the
characteristics of employed and non-employed older workers differ by conducting a
series of Chi—square tests, independent t-tests, and binary logistic regression. Multi-
nominal logistic regression is the most frequently used model for multinomial outcomes
(Long, 1997). For this study, a multinomial logit model was designed to examine which

individual level variables were accounted for the following criterion variables:
employment status (full time job/part time job/non-employed).
Descriptive and Bivariate Analysis
Sample Characteristics
Research Question 1: What is the Likelihood of Older Workers who Have

Acquired Disabilities in Their Later Lives?

58

The present study was restricted to the individuals who had a full-time career job
at age 50 or older and had a healthy condition at the baseline interviews. These sample
selection criteria were used since this study was interested in the labor transition
behaviors after older worker experience some level of disability. As shown in Table 2, a
total 15,049 age-eligible (over age 50) full-time workers reported being free of disability
(healthy) at the time of interview (baseline interviews) in 1998, 2000, 2002, and 2004. At
time I (baseline interview) and time 11 (after disability), there were 4,159 missing data in
ADL, IADL, and CESD variables. Thus, this study obtained a sample of 10,890 age-
eligible older workers who worked full-time and were reportedly healthy at the baseline.
Of these 1,096 respondents, 506 male and 590 female reported disabilities in the
subsequent survey (2000, 2002, 2004 and 2006) with following deﬁnitions: inability to
perform a set of physical ability to do various tasks (1) measured by respondents’
difficulty in performing activities of daily living (ADLs), (2) measured by respondents’
household management level of activities (instrumental activities of daily living, IADLs),
or (3) episodes of depression measured by the CSD-S. Per Table 2, this current study
begins to investigate the potential prevalence of late onset disability among the older
workers with full-time jobs. Among the total 10,890 age-eligible (age over 50) full-time
workers reported being free of disability (healthy) at the time of baseline interview, then,

1,096 older workers (10.1%) reported functional or mental disability two years later.

59

Table 2. Study Sample

 

Variable 1998 2000 2002 2004 Total
F ull-Time Workers Free of 3,118 2,694 2,042
Disability excluding Missing
Data at the Baseline Surveys

3,036 10,890

1998- 2000- 2004004 2004- Total
2000 2002 2006

Disability at the Subsequent 321 278 182 315
Surveys

* Excluding all missing data in ADL, IADL, and CESD variables at the time 1 (n=1 ,564)
and time 11 (n=2,595).

1,096

Research Question 2: What are the Demographics, Workforce, or Financial
Characteristics of Older Workers who have Acquired Disabilities in Their Later
Life?

Table 3 shows that socio-demographic characteristics of the 1,096 participants
differed to varying degrees. Most respondents’ age was 59.1 years old (SD= 6.3), on
average, and had mental disability (72.9%). Slightly over half (53.8%) were women. The
sample was largely non-Hispanic White (69 %); almost 17.4% were non-Hispanic
Blacks. The vast majority (63.0%) was currently married, but about 37% lived alone due
to the separated, divorced, widowed, or never married. The majority (73.5%) had at least
a high school education. Like in marital status, 67.2% of older workers were lived in their
household with two or by themselves. A majority of the study sample had white collar
occupations including managerial and professional specialty Operation, sales,
administrative support, and various services (69.9%) and their industry areas (66.5%)

such as wholesale, retail, services, real estate, professional related services, public

60

administration also presented similar pattern with white collar occupations. A total of

15.5% of the sample was self-employed.

Table 3. Characteristics of the Sample

 

 

Variable at the Baseline Frequency % Valid %
Gender
Male 506 46.2 46.2
Female 590 53.8 53.8
Total 1,096 100.0 100.0
Race
Non-Hispanic White 756 69,0 69.1
Non-Hispanic Black 191 17,4 17.5
Hispanic 86 7.8 7.9
Other 61 5.6 5.6
Total 1,094 99.8 100.0
Marital Status
Married 690 63.0 63.0
Not-Married 405 37.0 37.0
Total 1,095 99 .9 100.0
Level of Education
< High School 290 26.5 26.5
High School Only or GED 346 31.6 31.6
< College and More 459 41.9 41.9
Total 1,095 99.9 100.0
Type of Disability
Functional Disability 297 27.1 27 .1
Mental Disability 799 72.9 72.9
Total 1 ,096 100.0 100.0
Number of People in Household
One or two 737 67.2 67.2
>=Three 359 32.8 32.8
Total 1 ,096 100.0 100.0
Current Job Occupation
White Collar 756 69.0 69.9
Blue Collar 326 29.7 30.1
Total 1,082 98.7 100.0

 

61

Table 3. (cont’d)

 

 

 

Current Job Industry
Agri/ Mining/
Construction/Manufacturing/ 361 32.9 33.6
Transportation
Wholesale/ Retail/ Services 304 27.7 28.3
F inancial/ Insurance/ Real Estate/
Prof related services/ Public 411 37.5 38.2
Administration
Total 1,076 98.2 100.0
Self-Employed Status
Not self-employed 924 84.3 84.5
Self-employed 170 15.5 15.5
Total 1,094 99.8 100.0
Continuous Variable Frequency Mean SD
Age 1,096 59.1 6.3_
Financial Savings
Individual Earnings at the baseline 1,096 $ 29,227 $ 29,340
Household Net Asset at the 1’09 6 $ 243,165 $ 603,790
baseline
Socral Security Retirement Income 1’09 6 $ 2,419 $ 4,829

after disability

 

Note: Some N’s do not sum to 1,096 due to missing data.

The greatest discrepancies in participants’ socio-demographic characteristics were

in disability type (Table 4), gender (Table 5) and both disability type and gender (Table 6).

Chi-square tests, independent t—tests, and ANOVA (The Analysis Of Variance) were

conducted to investigate the existence of differences between older workers with

functional disability and mental disability, differences between males and females and

both disability type and gender. The differences were considered statistically signiﬁcant

when p-values were significant at the significance level of .001 or .05. The significant

differences between the two groups by types of disability and by gender are indicated.

62

As compared to older workers with functional disability, older workers with
mental disability were more likely to be younger, female, and to have white-collar
occupations. Per the pattern of occupation, older workers with mental disability were
more likely to work at the following industries: Wholesale, Retail, Services, Financial,
Insurance, Real Estate, Prof related services and Public Administration. However, older
workers with functional disability were more likely to work at the industries of

agriculture, mining, construction, manufacturing or transportation.

Even though race was not statistically different between functional disability and
mental disability, minority including non-Hispanic black, Hispanic and others were more
likely to have functional disability rather than to have mental disability. In addition, older
workers with mental disability were more likely to have a higher education level than

older workers with functional disability, but there is no significant difference reported in

education level (p value = 0.058).

There is no significant difference in earnings from work between older workers
with functional disability and mental disability. Since the average age for older workers

with functional disability was older (mean: 60.7 years old), they reported a higher

average social security retirement income of about $ 1,693 than older workers with

mental disability. This finding means a larger prOportion of older workers with functional

disability were more likely to quit their full-time job if they were old enough to get social

security retirement.

63

Table 4. Sample Characteristics by Disability

 

 

 

Variable Functional Mental p-value
Disability Disability
(N=253) (N=799)
Age (Mean) 60.7 58.5 H
(7.3) (5.7)
Gender
Male 63.6 39.7 * *
Female 36.4 60.3
Race
Non-Hispanic White 63.2 71 . 3
Non-Hispanic Black 20. 3 16.4
Hispanic 9.5 7.3
Other 7.1 5.0
Marital Status
Married 64.0 62.7
Not-Married 36.0 37, 3
Level of Education
< High School 31.3 24.7
High School Only or GED 27.6 33.1
< College and More 41.1 42.2
Number of People in Household
One or two 69.7 66.3
>=Three 30.3 33.7
Current Job Occupation
White collar 62.5 72.6 H
Blue collar 37.5 27.4
Current Job Industry
Agri/ Mining/
Construction/Manufacturing/ 41 .8 30.6
Transportation * *
Wholesale/ Retail/ Services 23.3 30.1
F inancial/ Insurance/ Real Estate/
Prof related services/ Public 34.9 39.4
Administration
Self-Employed Status
Not self-employed 83.5 84.8
Self-employed 16.5 15.2
Continuous Variable
Age 60.7 58.5 H
_ (7.3) (5 .7)

 

64

Table 4. (cont’d)

 

Financial Savings

Individual Earnings at the baseline $ 27,566 $ 29,845
($ 27,111) (S 30,120)
Household Net Asset at the $ 230,727 $ 247,788
baseline ($ 503,365) (8 637,324)
Social Security Retirement Income $ 3,654 $ 1,960
after disability ($ 5,864) ($ 4,298)

 

t-tests and ,‘(2 were signiﬁcant at the ** p< 0.001 and * p< 0.05 levels.

Table 5 displays the greatest discrepancy in participants’ socio demographic and
ﬁnancial characteristics by gender. There are obvious importance of overall gender
differences in race, marital status, types of disabilities, occupation, industries, and
earnings and total household net value. Based on the deﬁnition or the measure of
disability in this investigation, the rate of acquired disability is substantially different for
men and women. It is different ﬁndings from other studies. Other studies from the SIPP
(Doyle, Burwell, Good, Kalish, Martini, et al, 1989), the NHIS (LaPlante, 1996), and
CPS (McNeil, 1993) indicated that prevalence of work disabilities is not considerably
different by gender. The reason of these different ﬁndings may be resulted of two
differences: (1) other studies used the deﬁnitions of work disability rather than objective
measures of disabilities (2) other studies were focused on the aging p0pulation rather than
focused on full time older workers.

Even though F Iippen and Tienda (2000) state that “women are over represented in

white collar jobs which tend to offer greater stability than service and blue collar

occupations,” (p. 8) the results from this investigation conformed that female older

65

workers were more likely to white collar occupation (86.2%), and to work at white collar
occupation related industries (80.1%), not physically demanding industries. Although
changes in the structure of employment generally favor women and rising stocks of
human capital enhance their employment prospects, related to men, women older workers

are less likely to earn income from work, and less amount of total household worth than
men. Gender seems a central role negatively, especially for female older workers. Most of

female old workers experienced less favorable labor market condition than younger

cohorts (Hardy, 1991).

66

Table 5. Sample Characteristics by Gender

 

 

Variable Male Female p-value
=506) (N=590)
Age (Mean) 59.9 58.4
(6.6) (5.9)
Race
Non-Hispanic White 704 68,0
Non-Hispanic Black 13.] 21.2 **
Hispanic 10.5 5.6
Other 6.0 5.3
Marital Status
Married 77.3 50.8 * *
Not-Married 22.7 49.2
Level of Education
< High School 30.2 23.3
High School Only or GED 29,2 33.6 *
< College and More 40.5 43.1
Types of Disability
Functional Disability 37,4 18.3 H,
Mental Disability 62.6 81.7
Number of People in Household
One or two 64.2 69.8 *
>=Three 35.8 30.2
Current Job Occupation
White collar 50.9 86.2 H
Blue collar 49.1 13.8
Current Job Industry
Agri/ Mining/ 49.6 19.9
Construction/Manufacturing/
Transportation
Wholesale/ Retail/ Services 26. 5 29.7 *4:
Financial/ Insurance/ Real Estate/ 2 39 50.4

Prof related services/ Public
Administration

67

Table 5. (cont’d)

Self-Employed Status

Not self—employed 80.6 87.8
Self—employed 1 9,4 12,2

Continuous Variable

Financial Savings

Individual Earnings at the baseline $ 33,681 $ 25,408 *4:
($33,439) ($24,699)

Household Net Asset at the $ 263,324 $ 225,875 *

baseline ($583,616) ($620,541)

Social Security Retirement Income $ 2,933 $ 1,979

after disability ($5,533) ($4.08 5) *

 

Table 6 also displays the differences between older workers with functional
disability and mental disability by gender. In general, females were younger and
contained a greater proportion of not-married status than males including widowed,
divorced, separated, or never married. The level of educational attainment by gender
showed a signiﬁcant difference. Females were more likely to have a higher level of
education attainment than males. Females with functional disability (77.8%) showed the
highest proportion of high school diploma or more education than any other groups:
63.5% of male with functional disability, 76.5% of female with mental disability, and
73.5% of male with mental disability. Speciﬁcally, non-Hispanic black females and
Hispanic females were more vulnerable to have functional disability than their
counterparts. Black female workers were more likely to have disabilities than any other

racial grOUp.

Generally research on older black workers at middle and older ages reports are

68

more likely than white to experience disability and to experience restriction in a major
activity, including work (Hayward, Friedman & Chen, 1998). It is because blacks and
other minority older workers typically have less education, fewer skilled jobs, lower
salaries, and long periods of unemployment and are less likely to work in professions or
jobs with high beneﬁts (Bacon, 1995). Older black workers are “more likely to have
lower levels of education and fewer economic resources and tend to work in more
unstable industries than white workers” (F lippen & Tienda, 2000, p.6). The results of this
current study conﬁrmed that black female workers were the most vulnerable to have a
disability among other races. Hispanic female were more valuable to have functional
disability than their counterparts (Hispanic male).

The labor market characteristics, such as type of occupation, industry, status of
self-employed and ﬁnancial status, were also different by gender and types of disabilities.
With the ﬁndings on educational attainment in female, mostly female workers with
disabilities worked with white collar occupations. The proportion of white collar
occupation in female group was higher than male group. Specifically, male workers with
functional disability are more likely to have a self-employed position (20.6%) and blue
collar occupation (54.8%) than females does (self-employed= 9.3%, blue collar
occupation=7.5%). Compared to the male workers, female workers’ household total net
value is less than male group and females’ average income is less than males even though
female’s education level is higher than male group. The social security retirement income

was also less in female group than male. It can be resulted of inconsistence oflabor

activities (instability) among female.

69

Table 6. Sample Characteristics by Disability and Gender

 

 

Variable Functional Disability Mental Disability
(N=253) (N=799)
Men Women Sig. Men Women Sig.
(N=l89) (N=108) (N=3l7) (N=482)
Age (Mean) 61.7 59.0 ** 58.7 58.3 **
Race
Non-Hispanic White 75.0 68.5 84.5 74.7
Non-Hispanic Black 18.6 23.1 10.1 21.4 **
Hispanic 6.4 8.3 5.4 3.9
Other
Marital Status at Program
Entry
Married 76.7 41.7 *4: 77.6 52.8 H
Not-Married 23.3 5 8.3 22.4 47.2
Education at Program
Entry
< High School 36.5 22.2 26.5 23.5
High School Only or 29.1 25.0 * 29.3 35.6
GED
< College and More 34.4 52.8 44.2 41.0
Number of People in
Household
One or Two 69.3 70.4 * 61.2 69.7 *
>=Three 30.7 29.6 38.8 30.3
Work Status
Full Time 50.0 50.6 * 72.2 64.5
Part Time 15.5 12.9 7.6 13.3 *
Non-Employed 34.5 36.5 20.2 22.2
Self-Employed Status
Not self-employed 79.4 90.7 * 81.4 87.1 *
Self-employed 20.6 9.3 18.6 12.9
Current Job Occupation
White collar 45.2 92.5 *4: 54.3 84.8 M
Blue collar 54.8 7.5 45.7 15.2
Current Job Industry
Agri/ Mining/ 57.5 14.2 44.9 21.1
Construction/Manufact .
uring/ Transportation * * **
Wholesale/ Retail/ 22.0 25.5 29.2 30.7
Services

 

70

Table 6. (cont’d)

 

Financial/ Insurance/ 20.4 60.4 26.0 48.2
Real Estate/ Prof
related services/ Public
Administration
Financial Savings Mean Mean Mean Mean
(amount in dollars) (80) (SD) (SD) (SD)
Individual Earnings $ 28.757 $25481 $36,617 $25,391
($29,304) ($22,751) ($35,396) ($25,137)
Household Net Asset $261403 $177,045 $264,470 $23,6817
($593,002) ($280,188) ($578,889) (567.3359)
SOCial Security $4 467 $2 231 $2 018 $1 922
Retirement Income after ($6536) (“:6”) n (54.3/75) ($3.958) u

Disability
t-test and 12 are signiﬁcant at the ** p< 0.01 and * p< 0.05 levels.

 

Research Question 3: Speciﬁcally, How do Employed and Non-employed Older Workers
differ in Demographic, Workforce, and Financial Characteristics?

Employment Status

Table 7 and 8 illustrate the distribution of employment status after acquiring a
disability among older workers who had full—time jobs. Compared to non-employed older
workers, employed older workers were younger, females (54.3%), had mental disability
(77.0%), more than college graduates, had white collar (71.9%) occupation, worked at
white collar related industries, and had higher earnings from work. The data also showed
that non-employed older workers were more likely to receive Social Security Retirement
income (mean, $4,705), compared to employed older workers received (mean, $1,851).

Employed younger older workers were much more likely to sustain their
employment status. While the mean age of respondents’ who continued in a full-time job
was 57.79 (SD: 5.39), respondents who quit their full-time job after acquiring a disability

were 3.53 years older than the respondents who keep their full-time job. Nonetheless, it is

71

 

remarkable that 25.8% of workers ages 60 and above and 10.8% of workers ages 65 and
above continue to work full-time and 50.4% of workers ages 60 and above and 24.4 % of
workers ages 65 and above continue to work part-time even after they experience
disability (Table 8).

In addition, older workers who switched their full-time job to part-time were more
likely to be female, had a low level of educational attainment, had functional disability,
and be self-employed (Table 8). More highly educated workers were more likely to
sustain their job either full-time or part-time. Older workers with mental disability were
more likely to be employed. Females seemed to be more likely to sustain their
employment than males even though there was no found a statistically signiﬁcant
difference in gender.

Employed workers with part-time jobs were more likely to have white-collar
occupations, but their earning from work (mean, $22,139) was almost 8 3,500 in average
less than those of non-employed older workers (8 25,700). The total house net value for
older workers with a part-time job was the highest among three groups. It means that the

total house net value could be an important factor for older workers with disability to

choose a part-time job.

72

Table 7. Employment Status Distribution: Employed vs. Non-Employed

 

 

Variable Employed Non- Employed .
Total (N=817) =201) 5'3-
Age (Mean) 58.3 61.32 *
Gender
Male 45.7 47.5
Female 54.3 52.5
Total 100.0 100.0
Race
Non-Hispanic White 69.2 68.8
Non-Hispanic Black 18.0 15.8
Hispanic 6.7 11.1
Other 6.0 4.3
Total 100.0 100.0
Marital Status at Program Entry
Married 63.1 62.7
Not-Married 3 1.9 37.3
Total 100.0 100.0
Education at Program Entry **
< High School 23.8 34.3
High School Only or GED 30.3 35.4
<= College or more 45.9 30.4
Total 100.0 100.0
Type of Disability **
Functional disability 23.0 38.9
Mental Disability 77.0 61 . 1
Total 100.0 100.0
Number of People in Household
One or Two 67.0 67.9
<=Three 33.0 32.1
Total 100.0 100.0
Self-Employed Status
Not self-employed 83.8 86.3
Self-employed 16.2 13.7
Total 100.0 100.0
Current Job Occupations *
White collar 71.9 63-9
Blue collar 28-1 36-1
Total 100.0 l00.0
Current Job Industry *
Agri/ Mining/ 31.0 41.1
Construction/Manufacturing/
Transportation
Wholesale/ Retail/ Services 27.7 29.8
Financial/ Insurance/ Real Estate/ Prof 41.3 29-1
related services/ Public Administration
Total 100.0 100.0

 

73

 

Table 7. (cont’d)

 

Financial Savings

Individual Earnings $30,438 8 25,700 ,,
($ 31,230) (5 22,643)
Household Net Asset $ 257,263 $ 202,076 ,
(S 670,463) ($ 339,735)
Social Security Retirement Income $ 1,851 $ 4,075 "
after Disability ($ 4.384) ($ 5,630)

 

t-test and 12 are signiﬁcant at the ** p< 0.01 and * p< 0.05 levels.

74

Table 8. Employment Status Distribution: F ull-Time, Part-Time, vs. Non-Employed

 

 

Variable Full Time Part Time N0"-
Employed Sig.
(N=685) (N=l3l) (N=201)
Age (Mean) 57.79 60.92 61.32 **
Gender
Male 46.6 41.2 47.5
Female 53.4 58.8 52.5
Total 100.0 100.0 100.0
Race
Non-Hispanic White 69.2 69.5 68.8
Non-Hispanic Black 17.5 20.6 15.3
HiSpaniC 6.7 6.9 1 1.1
Other 6.6 3.1 4.3
Total 100.0 100.0 100.0
Marital Status at Program Entry
Married 62.0 68.7 62.7
Not-Married 38.0 3 1.3 37.3
Total 100.0 100.0 100.0
Education at Program Entry **
< High School 22.7 29.8 34.3
High School Only or GED 30.7 28.2 35.4
<= College or more 46.6 42.0 30.4
Total 100.0 100.0 100.0
Type of Disability * *
Functional disability 21.2 32.8 38.9
Mental Disability 78.8 67.2 61.1
Total 100.0 100.0 100.0
Number of People in Household
One or Two 66.4 70.2 67.9
<=Three 33.6 29.8 32.1
Total 100.0 100.0 100.0
Self-Employed Status **
Not self-employed 86.4 70.2 86.3
Self-employed 13.6 29.8 13.7
Total 100.0 100.0 100.0
Current Job Occupations *
White collar 71.0 75.6 63.9
Blue collar 29.0 22.9 36.1
Total 100.0 100.0 100.0
Current Job Industry **
Agri/ Mining/ 32.5 23.0 41.1
Construction/Manufacturing/
Transportation
25.6 38.9 29.8

Wholesale/ Retail/ Services

 

75

 

Table 8. (cont’d)

 

 

 

Financial/ Insurance/ Real Estate/ 41.9 38.1 29.1
Prof related services/ Public
Administration
Total 100.0 100.0 100.0
Mean
(SDI
Financial Savings
Individual Earnings $ 32,025 $ 22,139 $ 25,700 H
(5 31,478) ($ 28,606) ($ 22,643)
Household Net Asset $ 216,886 $ 468,399 $ 202,076 H
(S 443,695) (1,314,846) (S 339,735)
Social Security Retirement 8 1,380 $ 4,311 $ 4,075 **
Income after Disability (8 3,763) ($ 6,223) ($ 5,630)

 

ANOVA and 12 are signiﬁcant at the ** p< 0.01 and * p< 0.05 levels.

Table 9 through Table 11 display more detail information regarding employment
status (employed vs. non-employed/ full time, part time, vs. non-employed) by disability
type. Chi-square tests and independent t-tests were conducted to determine if there were
signiﬁcant differences between those who sustain their employed status or switched to be
non-employed.

Differences by group were shown for the amount of total assets and social
security retirement beneﬁts, regardless of disability type. More highly educated workers
were more likely to sustain their job either full-time or part-time, regardless of disability
type. Interestingly, a higher proportion of employed workers with mental disability had

significant differences in occupation and industry in comparison to individuals with

functional disability.

Table 10 and Table 12 illustrate the results of chi-square tests and ANOVA tests.

Several ﬁndings regarding switched employment status from full-time are worth

mentioning. First, employed older workers who sustained their full-time job were

76

younger, regardless of disability type, compared to employed older workers who changed
their full-time job to a part- time job. It appears that as workers age, they have a tendency
to choose either part-time job or be non-employed. Second, transitions from full-time to
part-time jobs were much more likely among self-employed older workers, regardless of
disability type. However, self-employed older workers with functional disability were
about 11% more likely to choose a part-time job than self-employed older workers with a
mental disability. Economic characteristics also appeared to inﬂuence transitions into
part- time job or non-employed status (Table 10). Regardless of disability type, social
security retirement beneﬁts and total household net value appeared to be important
factors influencing the transition from employed to non-employed status (Table 11).

If older workers have a more stable ﬁnancial status, they are more likely to switch
their full time job to part-time or non-employed (Table 12). Further, the interface of
marital status and disability type also impact employment transitions, as non-married
older workers who have a mental disability were more likely to be non-employed
compared to the non-married older workers with functional disability (Table 13). Of
1,096 respondents who had a full-time job, about 74.7 percent of married men and 70.4
percent of unmarried men were continued to choose employed, compared to 74.6 percent
of married female and 75.9 percent of unmarried female (Table 13). This ﬁnding means
that married men and unmarried women were more likely to continue their job even after
they experience disability. This pattern, however, was observed to be different in the non-

employed group as 29.6% of unmarried males were non-employed from the full-time job

after acquiring a disability, compared to 24.1% married women.

77

 

 

 

 

.m_o>o_ mod Va ... 28 Ed Va 4... on. .m 58...?me 2.... id can an.-.
. . . . 5:86.553 0:95 \ 803.8 3.22 we... l
w om o av o mm a an 3.3mm Rom $052.55 >285?—
m. . m mom mum _. _ m moomtom \zﬂom Emma—0:3
. . . . =o_.mtoam..aﬁ.\w:_..:.o£::m2
n am v mm a E o ow Eo_.o=bm..oo\w:_£_2 new...
bums—E— ach. E9550
.éQaho 93%.: 03.25 0.3%.? 3:8 2.5 2336 32>»
2.323530 new 39.2.0
922%.. an :3... ”.33.. 2:3. .w 86159.3 66539.6... .3.
3.5m Baa...e.m.._om
o. God... 3&93 mam: .8 wand». K 03% :2. 202 \ 03... .o 2.0
22.3.5: 5 «icon no .8383].
E: 3.. 2... S... 26:. .o 68:8 uv
Ndm v. _ m Yam mew QmO .o ch .oozom 5:.—
E... .2 3... EN 323 aw... v
55353—
vdmédo ”.333 cam: .3 mHQwNo 3:32-82 %o_te.2
2.3% .31.“:
3. 0m w... m... .050
.3 E a: a... 6.5%.:
mm. 3. 3. SN V.85 6.5%.z-=oz
m. E a: .36 NS 2...; 2:82:82
34%
0.8%.? 5.32.2. ﬁondo NGQQS 29:6“. $.22
315$
3... 2.. 08 6% .582. 814.1
5.12. asuz. .2..qu 3.216
tome—nan.— -.SZ .83. come—aﬁm tome—aEH .52 .86... 3.33:5— 03.3.3,
.2.qu 9.33.: .552 San... 9.5.2.: .2265...

 

 

 

mon=Ewm5 .«o mom»... .3 coasntﬁa 3.9m Evie—aﬁm .o 03m...

78

8.26. 8... X. ._. Ba .8... X. t e. 232.6% 2.... N... .68 <>oz<
€585

 

 

 

 

 

 

 

 

 

a... Maﬁa maﬁmm 23% a... 8.3% wmhdm mmngm 8...... 2.80.:— 2.2.3.qu
435.com Eoom
.1. 09.2.; o. :03 Zodgm Sofa www.mnmm mvmdomm .33.. .62 2058301
8.4.on 91.me v.a_o.mmm ... wmm.vmm 30.on 39mm m mmﬁgm Esﬂﬁnﬁ
Ewe—.6 mung—«m. 3652.3
3:13 .316 safe 5:16 $.16 W24...
.39. tome—QEH :52 0.5—. 2.3 25% ==...— ﬁrm. coho—9.5— -..oz 2:; ﬁsh 2:: ==m
8205 banana .555 2.3"... 5:33.: .3385..—
oEC. 5:535 ”maxi 83.3 .3 855.0820 oEamEoEoQémoom .2 036...
.232 36 Va ... 9.... 50.0 va .1. on. .a .cmommawﬁ Be a». can .8...
. . . . 3585 .9...
a... mom ma omm .m .2. coo aw mow mm 2.805 .coEobﬁm 435.com .ﬁoom
... 35.2.; www.comm www.cvmm madame .39... .62 29.8.52
... nmvdmm mnhdmm * wmm.vmm o g mdmm meEmm 32.36.:
Ago—.6 ww5>am .5235..—
.se 3...... ea ﬁe
.ua. come—aEH 1.82 come—aﬁm .MG Esme—aﬁm incl come—a5“.—
Aaabuev 3:33.: .352 Aroma... banana €552.5—

 

 

oncmr 3:385 $355 .083 .3 85:28.820 oEaﬁmoEonéﬁom .o. 223.

79

20>“.— mod vu * E... Sod Va .1. on. .m .cwoamcwmm 0... NR 9... <>OZ<

 

 

 

 

 

co_.a..m_:_:6< 0.3.5 \ 30.3% 3.50.
”.3 N... 3.. SN 93 o... .2. .25... .8. .8552. 32.2.”.
m. . N «N. 3N .NN N. . N Z: 30.2% .53. 3.322....
cougoamaﬂF \wctsﬁﬂsﬁwz
EN .3: a2 3... N. . N «NV E23328 @552 EN...
* .3346..— acn. 2.2.50
.333 3:2... 3N5... $3.8 o. . N53 Nag... 3:8 2.... 22.8 2......
* 23.5....qu :3. 2.9.30
0.2;... ..oN\o.NN a. z. .2. w... :3. NNQwNo «NS... 8.2.80.3 ﬁéaeﬁsm .oz
.. 3. 2.5m 3.2.5....«m
o. 3% Emma. 93...... 3.»... .8 oNN...NN 3N... .N o}. E... 222 \ 03.. s 25
22.353. 5 2.30.. .3 25.5.2
3.... 2. m3 2.. N. . m NS 28. .o em9:5 av
N... a. . m .V. . N §N .2 SN am... 8 .30 .ooﬁm Pa...
3.. 2.. a. .N 2... SN EN 323 E: v
* * =c_.ao=_.m
.338 932: 335.8 0.3. .8 35. .8 3... .No 352.82 @252
mzwaum 3:55:
3 N M... c... N... 3 .25
to .3 3 a... S o... 025%.:
N... EN 3. 3. EN wNN .85 0.5%...-82
m... Won v. E wéc 3b :5 32>? o...a%_:-:oz
3nd
@3va NNSNN 93...? 338 N63... 3. 2 .No 225... .22.
* .5259
t. 98 a... E. t. 3.. N... N... £32. a...
3..."... 5:"...
-8... .9 t .. h .. m .52 .9 t .. F ._ ,. o_.ata>
Nara... 9.5%... .552 .SNu... 9.3%... 3.3.3.5...

 

0.5 9.585 N. 8.3.5... .33 .58935 .N. 2%.

80

Table 13. Employment Status Distribution by Gender and Marital Status

 

 

Variable Married Unmarried

Male (%) Female (%) Male (%) Female (%)
Employed ** 74.7 74.6 70.4 75.9
Non-Employed ** 25.3 25.4 29.6 24.1
Total 100.0 100.0 100.0 100.0

 

12 is signiﬁcant at the ** p< 0.001 and * p< 0.05 levels.

Binominal Logistic Regression

A binomial logistic regression model was calculated for the total sample. The
logistic regression models were calculated where odds ratios of intended non-employed
were estimated in three consecutive models, using the total sample of 1,071 participants
with complete data and no signiﬁcant interaction effect was found. The Hosmer and
Lemeshow statistic, which evaluates the model’s goodness-of-ﬁt, provides a measure of
overall model ﬁt by comparing the observed and predicted values (Hosmer & Lemeshow,
2000). The odds ratios or the exponentiated value of the raw regression coefﬁcient are
used to interpret the change in the odds ratio associated with a l-unit increase in each
predictor variable (Tabachnick & Fidell, 2007). The Variance inflation Factor (VIF)
measures the impact of collinearity among the variables in a regression model. In general,
values of VIP that exceed 10 are often regarded as indicating the presence of strong
multicollinearity. For this study, there was no found the multicollinearity among predictor
variables.

As illustrated in Table 14, results of the binary logistic regression analysis
suggested that the models indicated that some predictors reliably distinguished between

older employed workers and non-employed workers. The ﬁrst model was restricted to the

81

demographic characteristics such as age, gender, level of education, race, marital status,
and disability type.

Similar to the results of the descriptive analyses, younger age and having mental
disability rather than functional disability were individually strongly associated with be
his or her sustained employment in model 1. The odds ratio for mental disability relative
to functional disability in model I was 2.04. This means that the odds of deciding to
continue to work are 2.04 times higher for older workers with mental disability than they
are for older workers with functional disability. In addition, low level of education
attainment was negatively associated with sustaining his or her employment. The odds
ratio of keeping employment were decreased by these two factor of 0.56 and 0.57
respectively when the older workers have more than high school level of education
compared to less than high school attainment and equal level of high school attainment.

In the second model, the three labor market variables were entered simultaneously.
Unlike the results of Model 1, gender showed statistical signiﬁcant. Women were
signiﬁcantly less likely to be employed than men. It was conformed many prior studies.
There have reported evidence of lower labor force participation rates for women across a
wide age range. Self employed older workers were individually strongly associated with
choosing employed. The odds ratio for self employed older workers relative to non-self-
employed older workers was 1.68. This means that the odds of deciding to continue to
work are 1.68 times higher for self employed older workers than they are for non-self-
employed older workers. Industries with physically demanding jobs such as mining,

construction, and transportation were also negatively related to be employed compared to

82

 

the white collar related ' ‘
Industries such as ﬁnancial, insurance 1
, rea estat
e, PFOfes '
810113]

related services and public administration

' Cla .

and II, age did ' ' ' m m T u
not statistically linked with be e ployed in odel III he res lt f
Hosmer and Lem w w — c = 0w h
esho test as not statistically signiﬁcant ( {2 — 5 92 p 0 657) h'
conﬁrmed that th mod w v . mo, d . w w
e '
el as reliable. In addition, the relati ely best e1 fit as

observed ( 2
z = 111.1 -
3, p< .001) in model 111. It indicated that these predictors

-employed workers.

Signiﬁcant] ' ' '
Y dlStmgUished between employed older workers and non
s as

The last m
OdCI Was

indicat d b able to Correctly ClaSSIfy 763% of the formally processed case
e y the pred'

lcted and Observed classiﬁcation statistic. This is considered

acceptable in th '
e soc1al seiences (SPSS Inc., 2006; Tabachnick & F idell, 2007)-
61‘ age,

In sum
maf)’, result ' ' '
s indicate that higher level of education, female, young
ocial security

self-em
ployed stat '
US, white collar related job industry, and not receiving S

retiremen
t have all b nk w
een ' - w
11 ed to sustained employment among older orkers ith a
non-

disabilit
y. A Substam' .
emPIOyed Wh 13' PYOportion of workers switched their employment status to
en they be . .
came eligible for early and ﬂ!“ Social Security beneﬁts at age 62.

83

 

2 ._ :

3.8:
Avadtmwdv “Iwwd
3o._-em.8 3o
82-3.8 3.:

33.88 33
3883.8 33

2240.8 ”20
33-8.: :2
3n2¥8 2:
53-3.: £3
Geo-$8 Led
33.38 $5

2%. 78.8 23
33.3.38 33
3_ . 738 86

8.00
3.”: 2

33-38 o;
Geo-£8 .32

33.788 mg
393-8.: .83
82,38 8..
thdtovgv tame-N
33-3.8 ...ee.o
33-8.8 $8

2.2-3.8 85
20.3-8.8 a...

 

 

 

 

 

5.;
09%:

as: 78.8 :._
83-80. 8 :33
3.88.8 1&2
5.33.8 .83

32.98 of
5.3-$8 R;

 

 

 

Abzﬁwmmm 5th 0:50:— EoEoEom eat-Boom 300m

\wctﬁoﬂscmzxcocoeamcoo

$-20
t. we;
0 5
comm/Q 002. Eon—823.: :4
AmeEmm 13230:: :4
352% 36:25
moﬂtom \zﬂom Esme—0:3
:oﬁmtOQmEﬁ-H
use: Ew<
32080855010. 23.5 853.5% 0838 005
\Bﬁmm .wom 305:3:— >£o=mci “mobbmsvﬁ no.- E250
€2.00 35 "mo-o $533380 no.- 0:250
Awake—nEm-toméoz H005 magnum coho—aEm tom
A03..- 55 222 "-6: 29.3.5: E Bacon mo BLE: Z
3:535 aeoeeei "020 £535 0o 25.
9.0 5 8:0 323 as:
.023 .3: v
coozom :3: :2: 202 £08 cacao—0.0m
8252-8282838 .352

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.550
5.73.8 $6 38738 $8 e53:
83.3.8 82 3.03.98 a? 33.88 8: 0.85
8.23%: 83.
Geo-$8 :3 23.88 ewes 33-3.8 Rd 332822850
Hos-38 36 32.88 :23 God-88:86 9.?
3.0 $30 2.0 $30 2.0 8.8
l 33. 2:5 222 2:5 2.5.3.5 l e355»
m .252 3 ace: 0 .252
GoonEméoZ Hoococomomv

 

H:8; “20 =e .8 2825 858

=8 e\emm use 8.58 £008 £0022 :owmmocwom unawmou 3802mm .3 oBm-n

84

Research Question 4: What Factors Determine the Older Worker’s Labor
Transition?
Multivariate Analysis

Multinomial logistic regression was used to identify the factors inﬂuencing older
workers’ labor transition patterns. This section examines the various factors that
inﬂuence the patterns of older workers’ transitions after they experience disability. The
current study estimated the multinomial model based on the HRS respondents who had a
full-time job at the baseline interviews and for whom later their disability was identiﬁed.
The speciﬁcation examines three types of employment transitions as follows: (I)
remained in a full-time job, (2) moved to a part-time job, and (3) exited the labor force
directly from a full time job. The multinomial logit model assumes independence among
the choice outcomes, and the probability of any one alternative can be calculated using
the outcome of a comparison between the changes in the odds of all three alternatives.

Since the current model involves three different types of labor force participation
and no clear hierarchy exists among the alternatives, the multinomial logit model is
appropriate (Maddala, 1983). The estimates from multinomial logit regression analysis
are presented in Table 15 and Table 16. In order to investigate the odds of being
employed either full-time or part-time among older workers with disabilities verses non-
employed, the reference category for the dependent variable was non-employed (Table
15). For the odds of being full-time workers versus part-time workers, the reference

category for the dependent variable was full-time work (Table 16).

85

 

 

 

 

 

 

 

86

85 v: .. :5 Sod va .1.
”mm—m: vooﬁﬁxzmoq
onﬁo ... 52 5.8200
38.7858 so: :5: 88.8358 8:: :mm 3- 3:535 .25 2:85 505:5: 558m 588 5
38.7338 58.: 58.: 38.7338 352 ”5.:- cam< 52 228.85 5
38.7538 89.: 58.:- 327338 85: 985 @555 5555:: 5
mw:_>am 35505-5
35725.8 35.: 3: .o- 38.7378 29o :88- 3228 :53. 5585:?
33.88.38 323 .5320- szd-ammov 23 58.8 Szaeoamcg 55585525252580 has: :53
AcozwbmmEEv< 0:5?— \moo_>._om .0822 .005
3:85 .moM 50:8sz 23052:". ”00:595sz no.- 5250
35.7278 3;: 83.8 35. 785.8 E: 585 2:50 2:5 ”5.0 225280 52 5050
8:35-32. : :nm 18.: 33.3-20.8 :2 5.8520 803555-5852 H50 355 8358503
35.7858 8;: 5 .o 63752.8 ME: ES 33 .55 202 ”50 20:85: 5 058: 5 55:52
88.7258 533.. 335 32.3%.: 83.3 $33.: 3:585 58:85 ”5: E5355 25.
33758.8 83 $3- 38.3038 25.: :83- 550 5 5:0 523 :5:
83.7338 53 $3- 38,338 was .33.? 523 :5: v
Coozom :wi 52: 822 ”1005 525053
3.53-53.8 :5: :83 63736.8 85.: $3.:- GoEazaozucavmaam .252
283-338 820 39:. 283538 2 n. :30 5:5
33.7538 3.: 83- : 2.32.8 $5.: :30- 055.2:
38.3-35.8 25.: £3 $83-$58 52.. 835 :85
Siam-58 85.
35.7308 3a.: 83- :35538 89o .588- 552%: 5:50
38.7358 22 5.8.8 88738.8 520 5.8.:- a?
2.0 58 m 2.0 :58 m
2.5— 250 2.5— 3.00 «Si-:5
i 25:- 55 2:: =5. l
Gaza—559.52 ”8:05-35 £59m EoEzoEEm :23 35553 8:5:on

 

nomnmﬁwoﬁoaéﬁom
”commmocwom omammmoq 3:552:26: 03mg.

 

 

 

 

 

8.: v: _.. Ba 8:: v: .1.
_N_. G: vooszoxzaoq
*NowN- EﬁmcoU
Amwm. 708. 3 37: 1.026 00:5me Ban 0600:. EoEoEoM binoom .3009 :4
38. 33.8 «8.: :86 383 52 223.65 :1.
35.7.2.3 wmad pod- AmeEwm 33036:: :4
meZwm 328:5
Ame—$58.8 52 :86 822% :53: 5335:?
3?. 7:208 8:5 8:5. 8:55:25 55.385525302380 @552 83
€0585E§u<
2.3::— Ruotcom @232 00.5 $235 :03— Bocﬁsmﬁ 280555 "00800320:— no_. 2.8.50
5:. 7538 SS «Ed. 350 2:5 ”:20 2208300 :3 52.50
28.338 $3 85: 8085555882 ”.5: 355 8855: :5
300. 7.2.0.8 «8.. 036 33:. :9: 222 n55 20:88: E 2:on :0 35:52
83.838 82 .332- 3:535 55:85 “:20 3:535 5 25
ANNE 7.30.8 mg.— 256 QmO 5 3:0 Bop—om gm?—
33.~-:w.8 :3 33 529: :5: v
coozom :w_: 52: 0.52 ”we: E06323
GmQNém _. 3 ME: .3310 AvﬂtaiéoZHwomeSm 3:82
8%. 7828 58o 80¢. 350
ﬁmm.m-mov.8 m: I 8 7o 25ng
Amvoddoﬁov gm; 3 _ .0 x85
GEE/"02v 8am
53.38. 8 85.: LE»: 55512: 5:50
38. 78.0.8 :8: wood a?
3.0 $3 m
3.5— .03.0 05323
«Ea. tam
0.83 05513 n ”8:88.05 maﬁm “coEonEm 58> 853:3 0_Eocoom dEmﬁonoQémoom

”commmohwom ozmmwoq _a_Eo:£:E.o_ 2an

87

F ull-T ime Work versus Non—Employed among Older Workers with Disabilities

Table 15 summarized the odds of being employed either full-time or part-time
among older workers with disabilities verses non-employed. The main drivers of post-full
time work transition were male, attained a higher education, mental disability, self-
employed status, white collar related job industry, and Social Security Retirement beneﬁt
(see Table 15). One of the notable differences was that older workers with mental
disability were more likely to continue their full-time jobs compared to the older workers
with a functional disability, when controlling for other variables. The signiﬁcantly higher
odds ratios for a mental disability (2.23, p value=0.001) sustaining their full time were
found. This means that the odds of being employed full-time for older workers with a
mental disability were 2.23 times higher than the odds for older worker with functional
disability.

Female workers relative to male workers decreased the likelihood of employed
full time work versus being non-employed. The odds of being full-time rather than non-
employed was deceased by a factor of about 0.62 by being female rather than male,
controlling for other variables in the model. As hypothesized, male workers were more
likely than female workers to be employed full time rather than to be non-employed. It
conﬁrmed that most of female old workers experienced less favorable labor market
condition than male (Hardy, 1991).

Higher educational attainment increased the likelihood of full-time work versus
the likelihood of non-employed. A higher level of education may impact the worker’s
ability to have more options in regard to their work (more desirable jobs and better

working conditions) (Kim & Devaney, 2005). If an older worker has higher education

88

attainment, he or she might be able to sustain his or her job. It is expected that the level of
education will be positively related to be employed either ﬁill-time or part-time.

Married older workers relative to older workers who lived alone decreased the
likelihood of full-time work versus the likelihood of non-employed. The odds of being
full—time rather than non-employed was deceased by a factor of about 0.86 by living with
spouse rather than living alone, controlling for other variables in the model. This means
that couples can be an incentive for leisure in an individual’s later life and sometimes,
they have substantially more ﬁnancial resources due to the extra earning potential. It is
hypothesized that those who live alone will be more likely to have full-time rather than
the married. Hispanic older workers had a decreased likelihood of full-time work even
though it was not statistically signiﬁcant. Such a ﬁnding is consistent with the statement
that “elderly Hispanics are twice as likely and elderly blacks three times as likely to be
poor as are elderly whites (Quinn & Kozy, 1996 p. 364).

The likelihood of full-time work versus non-employed increased when older
worker were self-employed. Self-employed workers were more likely to sustain their full-
time work compared to non-self-employed older workers. The choice of employed full-
time versus non-employed in self-employed is logical because the self-employed have
real advantages such as working for one-self and ﬂexible work hours. The odds of being
full-time employed rather than non-employed for self-employed older workers were 1.72
times higher than the odds for non-self-employed older worker.

If older workers were employed in the industries of ﬁnance, insurance, real estate,
professional related services, and public administration, they would be more likely to

keep their full-time job compared to older workers worked at labor related industries. Job

89

industries such as mining, transportation, construction, and manufacturing or wholesale,
retail, and services were signiﬁcantly negatively related to the likelihood of full-time
work versus the likelihood of non-employment.

Receiving Social Security Retirement income was negatively related to the
likelihood of employed full-time work. As hypothesized, older workers who received
Social Security Retirement income was less likely to be employed full time rather than to
be non-employed.

Part Time Work versus Non-Employed among Older Workers with Disabilities

Logistic regression for the part-time sample (see table 15) versus non-employed
revealed fewer statistically signiﬁcant coefﬁcients because small sample sizes for older
workers who switched their full-time job into part-time appeared to have limited the
extent to which statistical signiﬁcance could be found. However, three ﬁndings from
multinomial logistic regression remained.

First, older workers with self-employment were positively related to the
likelihood of part-time work. The choice of employed part-time versus non-employed in
self-employed is also logical like full-time work versus non-employed since the self-
employed have the ﬂexibility to establish their hours. The odds of being part-time
employed rather than non-employed for self-employed older workers were 2.51 times
higher than the odds for non-self-employed older worker. As hypothesized, self-
employed older workers were more likely to have a part—time job compared to the older
workers who are not in labor participation.

Second, work industries requiring physical demands such as mining, agriculture,

construction, transportation, and manufacturing versus white collar related industries

90

were negatively related to the likelihood of part-time work. The physical demands of
work industries were likely to make it more difﬁcult for older workers to continue to
work. Lastly, gender, types of disability and ﬁnancial factors had no effect on choosing
part-time work compared to choose non-employed. It may possible that deciding to
change their full-timejob into part-time among older workers may be more complicated
than deciding to sustain their full-time job.

Full Time Work versus Part Time Work among Older Workers with Disabilities
Logistic regression for the part-time sample (see table 16) versus full-time
revealed some signiﬁcant ﬁndings. Older workers with mental disability relative to older

workers with functional disability were less likely to choose part-timejobs compared to
the older workers who choose full-time work, when controlling for other variables. The
odds of choosing part-time rather than full-time employment was deceased by a factor of
about 0.55 by having mental disability rather than functional disability, controlling for
other variables in the model. This means that the likelihood of choosing part-time work
versus full-time work increased for older workers with functional disability relative to
older workers with mental disability.

Female workers relative to male workers increased the likelihood of choosing
part-time versus full-time. The odds of being part-time rather than full-time was
increased by a factor of about 1.61 by being female rather than male, controlling for other
variables in the model. As hypothesized, female workers were more likely than male
workers to choose part-time work rather than to choose full-time.

Receiving Social Security Retirement income was positively related to the

likelihood of employed part-time work. As hypothesized, older workers who received

91

Social Security Retirement income was more likely to be employed part—time rather than
to be full-time. Income and household net assets had no effect on the likelihood of part-

time versus full—time employment.

92

CHAPTER 5
SUMMARY and CONCLUSIONS

The current investigation focused on the prevalence of acquired disability among
full-time older workers, the consequences of their acquired disability on employment
patterns, and the factors inﬂuencing later-life employment transitions. The descriptive
and analytical results have important implications for rehabilitation practitioners,
researchers and policy makers. This discussion chapter focuses on the implications for the
development of practice, research, and policy, with an emphasis on the implications for
older workers with disabilities and their employment pattems. The summary of ﬁndings
is ﬁrst presented.

Summary of the Results

Any empirical studies including both depression and functional disability must
not ignore a reverse causal relationship. That is, among older people, not only does
depression affect functional disability (McKenna, et al., 2005), but also functional
disability affects depression (Bruce, et al., 1994; Gurland, Wilder, & Berkman 1988). For
example, one estimate using HRS data (Dunlop et al. 2005) indicates that depressed
adults between the ages of 54 and 65 have more than four times the risk of ADL disability
than their non-depressed counterparts. Limitations in physical and emotional health or
disability and the presence of a work disability all signiﬁcantly affect the employment
status of older workers (Santiago & Muschkin, 1996). It is therefore necessary to
consider both depression and functional disability as aspects of disability in any research
analysis.

Of a total 10,890 age-eligible (age over 50) full-time workers who reported to be

93

thoroughly free of disability (healthy) at the time of baseline interview, the ﬁnal sample
in the current study consisted of 1,096 older workers who reported a functional or mental
disability two years later. Of these individuals, 506 male (46.2%) and 590 female
(53.8%) reported that they are experiencing disabilities. In the HRS population data, a
10.1% prevalence of acquired disability exists among full-time older workers.
Speciﬁcally, the prevalence of mental disability was 7.4% and the prevalence of
functional disability was 2.7% among older labor force participation.

The sample used in this study was restricted by the full-time worker status and
free of disability characteristic at the baseline interviews because the intent of this
investigation was to examine the relationship between disability and job transitions. Thus,
the characteristics of the sample (older workers who have an acquired disability during
their work life) are unique in some ways compared with other aging studies.

In the sample, a majority of older workers had mental disability (72.9%). The
mean age of the sample in this study was 59.1 years (SD= 6.3). Slightly over half
(53.8%) were women. This pattern is similar to the general p0pulation of older workers
without disability. A study using Bureau of Labor Statistics (BLS) data that surveyed
older workers without disability noted that “the older labor force has become increasingly
female” (Rix, 2004. p. 4).

The sample was largely non-Hispanic White (69%) and 17.4% of the sample was
non-Hispanic Blacks. The majority (63.0%) was currently married and 37% lived alone
due to the separated, divorced, widowed, or never married. The majority (73.5%) had
high school educations or better. Majority participants of this study had white-collar

occupations including managerial and professional specialty operation, sales,

94

administrative support, and various services (69.9%). In addition, the sample
employment industry areas (66.5%) such as wholesale, retail, services, real estate,
professional related services, and public administration also presented similar pattern
with white-collar occupations. A majority (84.5%) of the sample was not-self employed,
that is, they were wage and salary workers. An 84.5% non self-employed rate is
comparable to the results of Rix’s study (2004) using 2000 Bureau of Labor Statistics
(BLS) data. This researcher reported that the large majority of older workers were wage
and salary workers (87%).

Descriptive ﬁndings indicate that 62.4% of older workers kept their ﬁill-time job
even though they experienced some type of disability. Older workers at older ages were
much more likely to change their job to a part—time or quit their job than younger workers.
According to the data, older workers with mental disability consistently have a better
chance of being employed and sustaining their full-time job than older workers with
functional disability. One possible explanation of these ﬁndings is that the characteristics
of mental disability may not severely affect current job demands as compared to the
distinctiveness of functional disability. In other words, older workers with mental
limitations may be overcoming or accommodating their work-related challenges better
than older workers with functional disability.

The small chance of being employed among the older workers with functional
disability may be the result of the fact that their disability is too severe to sustain their
full-time job or that the functional disability is more likely to interfere with work if the
requirements of the job are more physically demanding (Loprest, et al., 1995). That is

why physically demanding industries such as mining, transportation, construction, and

95

manufacturing or wholesale, retail and services consisted of a high (41.1%) proportion of
the non-employed group (see Table 8). Therefore, older workers with functional
disability may have special needs in getting or keeping a job and beneﬁt from ongoing
support and services provided by vocational rehabilitation counselors. In addition to
ongoing VR support and services, work accommodations that promote return to work and
work retention may be used to help older workers with disabilities.

Findings from the results of binary analyses indicated that as younger older
workers were much more likely to change their employment status. The ﬁndings also
suggest that older workers with a higher education, those who are younger, who are
female, and those in white-collar occupations are more likely continue their employment
(see Table 8). According to Loprest, et al., (1995), on average, male and female older
workers perform different types of jobs within the overall labor market. These authors
also report that “among the older workers between age 51 and 61, only 48 percent of men
but 65 percent of women are in three occupational categories: managerial, professional,
sales, and clerical” (p. 302). This phenomenon was also similarly observed in the present
study (see Table 8). According to Table 9, the total net house value and Social Security
Retirement income among older workers with a part—time job was the highest among
three groups. It can be assumed that if older workers have ﬁnancial stability (saved more
money), they were more likely to choose a part-time job when they experienced disability.

The results of binary logistic regression supported that male, high level of
education, mental disability, self-employed, white collar related industry, and Social
Security Retirement income were all related to being an employed older worker. While

virtually all workers may be covered by Social Security, in general, there is no further

96

decline in the age of eligibility for Social Security beneﬁts (Rix, 2004). However, based
on the results of this investigation, it is true that if an older worker experiences disability,
he or she is more likely to receive Social Security beneﬁts.

Male workers were more likely than female workers to keep their full-time job or
change to a part-time job rather than be non—employed. It was conformed many prior
studies. There have reported evidence of lower labor force participation rates for women
across a wide age range. Education increased the likelihood of being employed versus the
likelihood of being non-employed. The attainment of more education impacts a worker’s
ability to have more options in regard to their work. Older workers with a college degree
or more education were more likely to sustain their job than older workers who had less
than a high school education. If older workers were employed in the industries of ﬁnance,
insurance, real estate, professional-related services, and public administration, they were
more likely to keep their full-timejob. Industries such as mining, transportation,
construction, manufacturing or wholesale, retail, and services were signiﬁcantly
negatively related to the likelihood of being employed either full—time work or part-time
work versus the likelihood of non-employed status. Receiving social security retirement
beneﬁts also signiﬁcantly decreased the likelihood of employed status versus the
likelihood of non-employed status. If older workers received social security beneﬁts, they
were more likely to be non-employed.

The striking result from the multinomial logistic regression is that the self-
employed older workers were more likely to keep their full-time job or part-time job
compared with the non-employed older workers. This may be because the self-employed

have ﬂexibility in work hours. The work industries requiring physical demands such as

97

mining, agriculture, construction, transportation, and manufacturing versus white collar
related industries were negatively related to the likelihood of full-time or part-time work
versus non-employed. Logistic regression for the part-time versus full-time (see table 16)
revealed some signiﬁcant ﬁndings. Older workers with functional disability and female
workers were more likely to choose part-time jobs compared to the older workers who
choose full-time work, when controlling for other variables. This means that the
likelihood of choosing part—time work increased for female older workers and older
workers with functional disability. As hypothesized, there is the small chance of being
employed among the older workers with functional disability because sometimes their
disability is too severe to sustain their full-time job or the functional disability is more
likely to interfere with work if the requirements of the job are more physically demanding
(Loprest, et al., 1995). Female workers were more likely than male to choose part-time
work rather than to choose full-time. This ﬁnding may reﬂect the fact that men are often
more likely to have a full-time position rather than part-time due to their stable work
histories. However, many women may have weakerjob histories rather than man due to
the career interruption for family matters. Or this ﬁnding may reﬂect the fact that women
who are still working full-time in their late middle age may tend to remain in their full-
time job because those women may have strong job attachment compared to most women
who experience more career interruption for family matters (Kim & DeVaney, 2005).
These women may be more likely to have part-time position rather than non-employed.
Thus, it is hypothesized that compared to men, women will be more likely to have a part-

time job.

98

Limitations of the Study
Multiple limitations exist in the current study that must be acknowledged and
considered in any data interpretations and conclusions. First, like all ex post facto designs,
the present study examined the relationship between the predictor and criterion variables
in relation to the experience of disability. The results of this study only capture the effect
of disability (disability effect) in a one time, cross-sectional view. Even though this study
used the classiﬁcation of timing (before and after disability) in order to examine the
relationship between predictor variables and labor participation patterns and attempted to
control for economic factors such as Social Security Retirement income that may
potentially bias the relationship (Gruber & Wise, 2004; Rogowsk & Karoly, 2000),
conclusions regarding causality may not be drawn solely from the results (Hadley &
Mitchell, 1997; Heppner, et al., 1999). For example, it is possible that for some older
workers, the transition from a full-time job to non-employed status may be the result of
job related stress, a layoff, or involuntary job loss. The long-term effects of disability on
older workers’ employment pattern need to be addressed more fully in a longitudinal
manner (Santiago & Muschkin, 1996).

The HRS study was based exclusively on self-report data. Use of self-report data
is susceptible to several confounds resulting from participant distortions, social
desirability motivations, attribution errors, and/or the participant’s relative self-awareness
(Katz, Rodin, & Devins, 1995; Schwarz, 1998). Criterion validity is also an issue when
self-report data is used since there were no employment records to conﬁrm self-reported

work status. Some participants may deliberately present a fake-good or fake-bad image,

or may unintentionally provide incorrect information. Further, older workers may have

99

tendency to not accept of some level of disability or limitation. Older individuals may
also provide inaccurate information if they view questionnaires as an invasion of their
privacy. Consequently, uncorroborated self-report data can result in erroneous or
inconclusive ﬁndings.

In addition, limitations related to the research sample impact the study’s external
validity. The samples utilized are based on small selected samples: full-time worker status
and no disability at the baseline surveys, and then the existence of disability measured by
validated measures. This situation raises the question about the study’s external validity
to other populations. Thus, the results may not be generalized to other older workers with
disabilities. In the same manner, there may be the unintentional potential selection bias
for the female samples. As mentioned earlier, over half of the sample used for this current
study was women. In general, women who are still working full-time in their late middle
age may be very different in their extent and type of job attachment than other female
workers. Older working women tend to remain in their full-time or employed status,
regardless of whether they need or want (Kim & DeVaney, 2005).

The issues raised by the study’s ﬁndings summarized above have a number of
important implications for the development of effective and responsive polices,

rehabilitation counseling practice, and research efforts. The implications in each of these

areas are discussed below.

Implications of the Study

Practical and Educational Implication

As descriptive ﬁndings indicated, 62.4% of older workers kept their full-time job

even though they experienced some type of disability. It means that older workers with

100

disabilities will become valuable members of the job pool and rehabilitation counselors
can contribute to the retention of aging workers who will likely need disability
management intervention and workplace supports (Bruyere, 2006). Even though
rehabilitation professionals may have been trained in broad areas of knowledge in
disability, vocational rehabilitation services in both the public and private sectors are
often not well-equipped to deal with the speciﬁc educational and occupational needs of
older workers with disabilities (Bishop, 2000; May & Vieceli, 1983). Rehabilitation
professionals in a variety of settings such as public, private, mental health, and one-stop
centers in the Workforce Investment Act programs may lack knowledge of characteristics
of older workers. It may true that they may approach all aged persons with disabilities
uniformly, regardless of different characteristics of their consumers. Studies of the service
utilization of older workers with disability display that older persons have traditionally
been largely neglected in formal rehabilitation programs (Kemp, 1985) and few older
persons apply for services from the Rehabilitation Services Administration (RSA;
Kampfe, 1994; Walls, Misra, & Majumder, 2002).

Now is the opportune time to attend to the demographic composition of older
clientele. Kelly (2003) emphasized that rehabilitation professionals should expand their
role and function as well as specialized knowledge domain when working with older
individuals. In order to provide effective and efﬁcient services, to prepare well for this
population and to obtain successful outcomes, rehabilitation professionals should have
speciﬁc knowledge about customers’ disability severity, type, prognosis, and work-
related skills as well as the expectation and needs of employers in relation to older

workers. Kennedy (2000) reported that improved accommodations in the workplace and

101

changes in the nature of work for workers with disabilities have allowed many such
individuals to rejoin the workforce. Schechter (1999) also found that special equipment
and modiﬁcation of work schedules (e.g., ﬂexible work schedule) were important factors
in job retention for disability insurance beneﬁciaries. Previous research also indicates that
some employers have expressed positive attitudes toward workers with disabilities who
are placed by vocational employment or supported employment programs (Hernandez,
2000; Petty & F ussell, 1997). In particular, employers who received supported
employment services (workers obtain training and support from a job coach to adequately
perform their jobs) are most satisﬁed with workers with mental or emotional disability.
Therefore, it is possible to reduce disparities in employment by providing vocational or
supported employment programs for older workers with disabilities.

Creating better estimations of trends in health and disability for older consumers
is much more than an academic exercise. Much is predicated on trends in the health of
the today’s middle—aged adults. As rehabilitation professionals gauge current and future
needs for retention services, “clarity about trends in speciﬁc measures of disability and
independence, based on knowledge about what is driving trends, is essential” (Trends
Research Brief, 2008, p. 6). If rehabilitation professionals realize the speciﬁc skills in
labor transition for older workers they may then minimize any adjustment difﬁculties that
may be experienced by some older workers. In this manner, rehabilitation programs could
prevent some workers from involuntarily exiting the labor force due to declining health
status or potential job limitations and encourage the aging population of individuals with
disabilities to continue to participate in or return to work.

Traditional rehabilitation settings as well as private case management systems

102

need to train their staff about heterogeneous characteristics of older workers such as
health care, job training, and retirement planning programs. The rehabilitation agencies
mainly dealing with employment as well as community—based programs need to train
their staff on the fact that older workers have unique concerns and values about work and
retirement. Moreover, rehabilitation agencies may improve their services for older
workers by cooperatively working with other agencies that have already been working
with the older population. For example, collaborative strategies including SCSEP,
vocational rehabilitation, and Welfare-to-Work should be promoted among service
systems. Needs assessment surveys and focus group meetings are effective tools for
initiating such collaborations.

As mentioned earlier, physically demanding industries such as mining,
transportation, construction, and manufacturing consisted of a high (41.1%) pr0portion of
the non-employed group (see Table 7). If older workers have functional disability, it is
more difﬁcult to sustain their work when the requirements of the job are more physically
demanding (Loprest, et al., 1995). Therefore, older workers with functional disability
may have special needs in getting or keeping a job and beneﬁt from ongoing support and
services provided by vocational rehabilitation counselors.

Rehabilitation counseling students need to be well-informed on the current life
course perspectives, the issues between older workers and their disability, and political,
economic and social forces that bring the older individual into the rehabilitation service
system. Students also need to be made aware of diverse characteristics of older workers,
their coping skills, resources, and problems facing due to the disability. Improving

curriculum to reﬂect the changing demographics of the current work force is an important

103

task for rehabilitation counseling education. If students gain knowledge about older
workers with disabilities, as they enter the workforce themselves they can provide agency
administrators with information about demographic trends and their implications for an
aging society as well as help evaluate and develop services that affect older workers with
disability. Additionally, hiring faculty members who are able to teach or conduct research
on topics related to the older workers with disabilities, recruiting students who are
interested in this population, and placing students in agencies serving this population are
important actions as well.
Implication for Research

This study is the ﬁrst investigation to empirically attempt to show the prevalence
of acquired disability and its effects on employment patterns among late middle aged
workers in the United States. Hopefully, this effort will spur a new line of research on
older workers regarding the need for expanding rehabilitation services for older workers
with disabilities and the education of rehabilitation counselors. While the analysis and
results presented here constitutes a signiﬁcant step in understanding the determinants of
older workers’ job transition patterns after they experience a disability, many important
unanswered questions remain. Perhaps, most importantly, the present study does not
answer the question of why older workers with disabilities enter non-employed status.

Important issues need to be considered for further research on job transitions
related to the experience of disability during later life. First, the importance of social
context inﬂuencing retirement planning needs to be examined. For example, the role of
race, ethnicity, and social class in the differential rates of wealth accumulation, saving

behavior and attitudes towards retirement planning merits further research. Second,

104

interaction between gender and ethnicity/race or between gender and disability types
needs to be explored. There has been relatively little research investigating the interaction
issue in disability researches. Third, the further differences between the impact of aging
on individuals with disabilities who acquired their disabilities at older age and those who
are living with long-term disabilities, acquired at birth, in childhood or young-adulthood
needs to be examined.

Longitudinal research on older workers with disabilities is also needed. Most of
such studies completed to date are cross-sectional in nature. Research also should be
conducted regarding the effectiveness of services for older workers with disabilities.
Increased research attention on older workers with disabilities may provide rehabilitation
practitioners with valuable data for providing effective services. In addition, further
research must be conducted to determine the most effective methods for assisting older
workers to adapt to a rapidly changing world of work.

Implication for Policy Makers

A number of studies have emphasized that rehabilitation counselors should
become more involved in providing services that meet the needs of older workers (Brown,
1997; Brown, & Rees, 1996; Fowles & Dunker, 1996). Previous research suggests that
traditional retirement preparation programs have not been sensitive to minority elders
(F owles & Dunker, 1996), particularly those who have disability. As previously noted,
current private and public funded vocational programs are very limited for older workers
with disabilities. Studies show that older workers are more likely to become unemployed
and to experience greater difﬁculties in obtaining new jobs (Mor-Barak & Tynan, 1993)

‘than other workers. Only 3 percent of 8.4 million low income workers age 45 to 70

105

participated in federally sponsored employment programs such as the JTPA Adult
Training, On-Stop-Career Center, and SCSEP (Poulos & Nightingale, 1995). Clearly,
there is a need to reexamine current programs and to develop more programs for this
speciﬁc population to attract older workers who need employment-related services.

The implications of these ﬁndings for policy are addressed follow. First, before
work can be supported, people need to ﬁnd jobs. Programs helping with the job search or
even preparation for the job search may alleviate the difﬁculties of some older adults with
disabilities. Programs can provide information about where jobs are or serve as an
intermediary between employers and older people with disabilities seeking jobs. Further,
programs could be targeted toward those older individuals who are experiencing the most
difﬁculty looking for jobs, those who have less education, female, or those who are
lacking skilled work experience. Clarifying distinctions among trends in disease,
functional limitations, and disability and their interrelationships will help policymakers

and program ofﬁcials anticipate the relative demand for rehabilitation services.

106

 

 

References

Abraham, K. G, & Houseman, S. N. (2004). Work and retirement plans among older
Americans. Employment Research, 11(4), 1-4.

Adams, S. J. (2004). Age discrimination legislation and the employment of older workers.
Labour Economics, 11(2), 219-241.

Agresti, A. (1990). Categorical Data Analysis. New York: Wiley.

Agresti, A. & Finlay, B. (1997). Statistical methods for the social sciences (3rd). Upper
Saddle River, New Jersey: Prentice Hall.

Alwin, DP. (1998). Linking aging and social change in the study of political identities. In
Hardy, M.A. (Ed.), Conceptual and methodological issues in the study of aging
and social change. Beverly Hills, CA: Sage.

Ashenfelter, O. C. &Card, D. (2002). Did the elimination of mandatory retirement affect
faculty retirement ﬂows?. American Economic Review, 92 (4), 957-980.

Bacon, C. (1995). A portrait of older minorities, Washington DC: Research Statistic
Center, Research Group, AARP.

Bearden, L. J ., & Head, D. W. (1985). Attitudes of rehabilitation professionals toward
aging and older persons. Journal of Applied Rehabilitation Counseling, 17(1), 17-
19.

Bendick, M., Brown, L.E., & Wall, K. (1999). No foot in the door: An experimental
study of employment discrimination against older workers. Journal of Aging and
Social Policy, 10(4), 5-23.

Bergman, M. (2007). Census Bureau Launches Older Worker Profiles for 31 States, US.
Census Bureau News.

Biddle, J. E., Leslie I. B., & Reville. R. T. (2001). Permanent partial disability from
occupational injuries: earnings losses and replacement in three states. In P. P.
Budetti, R. V. Burkhauser, J. M. Gregory, and H. A. Hunt (Eds), Ensuring Health
and Income Security for an Aging Workforce (pp. 263—290), Kalamazoo, MI: W.E.
Upjohn Institute for Employment Research.

Bishop, M. (2000). Multiple sclerosis and epilepsy: Vocational aspect of best
rehabilitation practices. Journal of Rehabilitation, 66(2), 50-55.

Bjelland, M.J., Bruyere, S.M., Schrader, S., Houtenville, A.J., Ruiz-Quintanilla, A., &

107

Webber, DA. (2009). Age and disability employment discrimination:
Occupational rehabilitation implications. Journal of Occupational Rehabilitation,
10(3),l81-193.

Blanck, P. (1996). Transcending Title I of the Americans with Disabilities Act: A case
report on Sears, Roebuck and Co. Mental and Physical Law Reporter, 20(2), 278-
286.

Bound, J., Michael, 8., Todd, S. R., & Timothy, W. (1999). The dynamic effects of health
on the labor force transitions of older workers, Labour Economics, 6(2), 179-202.

Brooke, L. (2003). Human resource costs and beneﬁts of maintaining a mature-age
workforce. International Journal of Manpower, 24(3), 260—283.

Brown, R.I., & Rees, T. (1996). The evaluator’s role in developing quality services. In R.
McConkey (Ed.) Innovations in evaluating services for people with intellectual
disabilities (pp. 149-166). Lancashire: Lisieux Hall.

Brown, RI. (1997). Quality of life for people with disabilities: Models, research and
practice. Cheltenham: Stanley Thomes.

Brown, SK. (2003). Staying ahead of the curve 2003: The AARP Working in Retirement
Study (Executive Summary). Washington, DC: AARP Knowledge Management.
Retrieved August 30, 2009, from
httmﬂassets.aargorglrgcenter/econ/multiworki2003rlgldf.

Bruce, M. L., Seeman, T. E., Merrill, S. S., & Blazer, D. C. (1994). The impact of
depressive symptomatology on physical disability: MacArthur studies on
successful aging. American Journal of Public Health, 84(11), 1796-1799.

Bruyére, S. (2000). Disability employment policies and practices in private and federal
sector organizations. Ithaca, NY: Cornell University, School of Industrial and
Labor Relations Extension Division, Program on Employment and Disability.

Bruyere, SM. (2006). Disability management: Key concepts and techniques for an aging
workforce. International Journal of Disability Management Research, 1(1), 149-
158.

Bruyere, S., Erickson, W., Wilson, D., & Somerville, N. (2004). The therapist’s role in
maintaining employment. In B. Kemp and L. Masqueda (Eds), Aging with a
disability (pp. 129—149). Baltimore: Johns Hopkins University Press.

Bryant, W. K. (1990). The economic organization of the household. Cambridge:
Cambridge University Press.

Burkhauser, R., & Quinn, J. F. (1997) Implementing Pro- Work Policies for Older

108

Americans in the Twenty-First Century. US. Special Committee on Aging (Serial
Number 105-7), Retrieved April 15, 2005, from http://fmwwwbcedu/ec-

p/ﬂ378.pdf.

Cahill, K. E., Giandrea, M. D., & Quinn, J. F. (2006). Retirement Patterns from Career
Employment. The Gerontologist, 46(4), 514-523.

Chan, S. & Stevens, A. F. (2001). Job loss and employment patterns of older workers,
Journal of Labor Economics, 19(2), 484-521.

Chan, Y.H. (2005). Biostatistics 305. Multinomial logistic regression. Singapore Medical
Journal, 46(6), 259-269.

Choi, N. G. (2000). Potential consequences of raising the Social Security eligibility age
on low-income older workers, Journal of Aging and Social Policy, 11(4), 15-39.

Corthell, D.W., Corthell, D.W., & McAlees, D. (1991). Aging America: Implications and
impact on vocational rehabilitation. American Rehabilitation, 17(1), 27-33.

Cottrell, G. W. & Fleming, M. K. (1990). Face recognition using unsupervised feature
extraction. In Proceedings of the International Neural Network Conference (pp.
325-325).

Critchley, R.K. (2004). Doing nothing is not an option. Mason: South-Westem.

Danczyk-Hawley, C., McMahon, B., & Flynn, B. (2002). Progression of disability
beneﬁts as a measure of disability management program effectiveness:
Implications for ﬁJture research. In Wehman (Ed), An Evaluation of the
Progression of Disability Benefits Among Workers in American Industry: Impact,
Outcomes and Implication (pp. 71-84). Virginia Commonwealth University.

DeVaney, S. A., & Chien, Y. (2001). A model of savings behavior and the amount saved
in retirement accounts. Journal of Financial Service Professionals, 55(2), 72-80.

DeVaney, S. A., & Su, Y. (1997). Predictors of the most important source of retirement
income. Compensation and Working Conditions, 2(Fall), 25—3 1.

Disney R. & Hawkes D. (2000). Boosting employment of older workers in Britain:
Prospects and Policies. Employment Policy Institute.

Dixon, CG, Richard, M., & Rollins, CW. (2003). Contemporary issues facing aging
Americans: Implications for rehabilitation and mental health counseling. Journal
of Rehabilitation, 69 (2), 5—12.

Doyle, P., Burwell, B., Good, T., Kalish, S., Martini, A., Miller, E., & Thronton, C. (1989).
Task 1: Population proﬁle of disability. Washington, DC: Mathematic Policy

109

Research.

Doshi, J ., Cen, L., & Polsky, D. (2007). Depression and retirement in late middle aged
US. workers. Health Services Research. 43 (2). 693-713.

Drebing, C. E., Losardo, M., Omer, A. V., Krebs, C., P, & Nasser, E. (2002). Vocational
rehabilitation and older adults: Patterns in participation and outcome. Journal of
Rehabilitation, 68(1), 24-32.

Dunlop, D. D., Manheim, L. M., Yelin, E. H., Song, J., & Chang, R. W. (2003). The costs
of arthritis. Arthritis Rheum, 49(6), 101-113.

Dunlop, D.D., Song, J ., Daviglus, M.L., Manheim, L.M., & Chang, R.W. (2007).
Racial/ethnic differences in the deve10pment of disability among older adults.
American Journal of Public Health. 97(12), 209-2215.

DunIOp, D.D., Manheim, L. M., Song, J., Lyons, J .S., & Chang, R.W. (2005). Incidence
of disability among pre-retirement adults: The impact of depression. American
Journal of Public Health, 95 (11), 2008008.

Dwyer, D., & Mitchell, 0. S. (1999). Health problems as determinants of retirement: Are
self-rated measures endogenous? Journal of Health Economics, 18(2), 17793.

Dychtwald, K., Erickson, T.J., & Morison, R. (2006). Worlg’orce Crisis. Boston: Harvard
Business School Press.

Emptage, N. P., Sturm, R., & Robinson, R. L. (2005). Depression and comorbid pain as
predictors of disability, employment, insurance status, and health care costs.
Psychiatric Services, 56 (4), 468-474.

Fagin, L. & Little, M. (1984) The forsaken families. Harmondsworth: Penguin.

Finch, J ., & Robinson, M. (2003). Aging and late-onset disability: Addressing workplace
accommodation. Journal of Rehabilitation, 69(2), 38—42.

Flippen, C. A., & Tienda, M. (2000). Pathways to retirement: Patterns of late age labor
force participation and labor market exit by race, Hispanic origin and sex. Journal
of Gerontology, 553(1), 14-28.

Fowles, D. G, & Dunker, A. P. (1996). A proﬁle of older Americans. Washington, DC:
American Association of Retired Persons and the Administration on Aging.

Fried, L.P., Herdman, S.J., Kuhn, K., Rubin, G, and Turano, K. (1991). Preclinical
disability: Hypotheses about the bottom of the iceberg. Journal of Aging and
Health. 3(2), 285—300.

Fronstin, P. (1999). Retirement patterns and employment beneﬁts: Do beneﬁts matter?

110

The Gerontologist, 39(1), 37—47.

Gallo, W. T., Bradley, E. H., Dubin, J. A., Jones, R. N., Falba, T. A., Teng, H. M., & Kasl,
S. V. (2006). The persistence of depressive symptoms in older workers who
experience involuntary job loss: Results from the health and retirement survey.
Journal of Gerontology: Social Sciences, 61(4), 221-228.

Gallup Organization (2003). Bringing work problems home. Gallup Management Journal.
Retrieved August 30, 2009, from hLLQ://g,mj.galltm.com/.

Gallup Organization (2006). Gallup study: Engaged employees inspire company
innovation. Gallup Management Journal. Retrieved September 30, 2009, from
http://gnljgallmgcomk

Galvin, DE. (1991). The disability management research review. Washington DC:
Institute for Rehabilitation and Disability Management, Washington Business
Group on Health.

Giandrea, M. D., Cahill, K. E., & Quinn, J. F. (2007). Bridge jobs: A comparison across
cohorts, Boston College Working Papers in Economics ((Publication Number
670), Boston College Department of Economics.

Gibbs, J. C. (1991). Socio-moral developmental delay and cognitive distortions:
Implications for the treatment of antisocial youth. In Kurtines, W. M. & Gewirtz, J.
L., Handbook of moral behavior and development (Vol 111). Lawrence Erlbaum
Associates: Hillsdale, NJ.

Gilbide, D, & Stensrud, R. (2010). Job placement and employer consulting: Services and
strategies. In E. Szymanski & R. Parker (Eds). Work and Disability (2rd ed.) (pp.
407-43 9). Austin TX: ProEd.

Giorgianni, S. (2005). Global summit on the aging workforce. The Pfizer Journal, 9(3),
4-38.

Goldman, B., Gutek, B., S, J., & Lewis, K. (2006). Antecedents and consequences of
employment discrimination in organizations. Journal of Management, 32(6), 786-
830.

Gottlieb, A., Vandergoot, D., & Lutsky, L. (1991). The role of the rehabilitation
professional in corporate disability management. Journal of Rehabilitation, 57(2),

288-301.

Gruber, J .. & Wise, D. A. (1999). Social Security and retirement around the world.
Chicago, University of Chicago Press.

Gruber, J. & Wise, D. A. (2004). Social Security programs and retirement around the

111

world: Micro-Estimation. NBER Working Paper (Publication Number 9407).
Retrieved May 22, 2009, from Qty/WWW.nber.org/_p_a_pgrs/w9407.

Gupta, D. Nabanita & Larsen, M. (2007). Health Shocks and Retirement: The Role of
Welfare State Institutions, MPRA Paper (publication Number 15497), University
Library of Munich, Germany.

Gurland, B.J., Wilder, D.E., & Berkman, C. (1988). Depression and disability in the
elderly: Reciprocal relations and changes with age. International Journal of
Geriatric Psychiatry, 3(3), 167-179.

Gustman, A. & Steinmeier, W. (1986). A Disaggregated structural analysis of retirement

by race, difﬁculty of work and health. Review of Economics and Statistics, 68(3),
509-513.

Habeck, R. V., & Kirchner, K. (1999). Case-management issues within employer-based
disability management. In F. Chan & M. J. Leahy (Eds), Health care and
disability case management (pp. 239-264). Lake Zurich, IL: Vocational
Consultants Press.

Habeck, R. V., Leahy, M. J., Hunt, H. A., Chan, F. &Welch, E. M. (1991). Employer
factors related to workers’ compensation claims and disability management.
Rehabilitation Counseling Bulletion, 34(3), 210-226.

Hadley, R., & Mitchell, L. (1997). Counseling Research and Program Evaluation,
Brooks/ Cole.

Hansson, R., P. DeKoekkoek, W. N., & Patterson, D. (1997). Successful aging at work,
1993996: The older worker and transitions to retirement. Journal of Vocational
Behavior. 5 1(2). 202—233.

Hardy, M. (1991). Employment after retirement: Who gets back in? Research on Aging,
13(3), 267-288.

Harley, D.A., Donnel, C., & Rainey, J .A. (2003) Interagency collaboration: Reinforcing
professional bridges to serve ageing populations with multiple service needs,
Journal of Rehabilitation, 69(2), 357.

Harter, J. K., Schmidt, F. L. & Hayes, T. L. (2002). Business-unit-level relationship
between employee satisfaction, employee engagement, and business outcomes: A

meta-analysis. Journal of Applied Psychology, 87(2), 268—279.

Hatcher, C. B. (2002). Wealth, reservation, wealth, and the decision to retire. Journal of
Family and Economic Issues, 23(2), 167-187.

Hayward, M. & Grady, W. (1990). Work and retirement among a cohort of older men in

112

the United States: 1966-1983. Demography, 27(3), 337-356.

Hayward, M. D., & Hardy, M. A. (1985). Early retirement processes among older men:
Occupational differences. Research on Aging, 7(4), 491—515.

Hayward, M. D., Friedman, S., & Chen, S. (1998). Race inequities in men’s retirement,
Journal of Gerontology: Social Sciences, 51B(1), 1-10.

He, W., Sengupta, M., Velkoff, V. A., DeBarros, K. A. (2005). 65 in the United States:
2005. US. Department of the Census: current population reports. Washington DC:
US. Government Printing Office.

Heeringa, S.Gr & Connor, J. (1995). Technical Description of the Health and Retirement
Study Sample Design, HRS/AHEAD Documentation Report (Publication Number
DR-002), Ann Arbor, MI: University of Michigan, Institute for Social Research.

Heeringa, S.G, & Connor, J .H. (1995). Technical description of the health and retirement
survey sample design. Ann Arbor, MI: University of Michigan, Institute for Social
Research.

Heidkamp, M,. & Horn, C. (2008). Older and Out of Work: Trends in Older Worker
Displacement. Sloan Center on Aging and Work, Issue Brief 16, Boston, Mass:
Boston College.

Heppner, P., Kivlinghan, D., & Wampold, B. (2008). Research Design in Counseling
(3rd). Belmont, CA: Thomson Brooks/Cole.

Hernandez, B. (2000). Employer attitude toward employers with disabilities and their
ADA employer right: a literature review. Journal of Rehabilitation, 66(4), 4-16.

Hess, D., Meade, M., Forchheimer, M., & Tate, D. (2004). Psychological well-being and
intensity of Employment following Spinal Cord Injury. Topics in Spinal Cord
Injury, 9(4), 1-10.

Hipple, S. (1999). Worker Displacement in the Mid-19903. Monthly Labor Review,
122(7), 15-32.

Hively, J. M. (2004). Fact sheet about older workers. Retrieved October 24, 2009, from
mp://www.van.umn.edu/options/2b5 factsheetasp.

Ho, P. (2002). Disability and employment status among older workers. Washington, DC:
National Rehabilitation Hospital, Center for Health and Disability Research.

Holtmann, A. G, Ullmann, S. G., Fronstin, P., & Longino, C. F. (1994). The early

retirement plans of women and men: An empirical application. Applied
Economics, 26(6), 591—601.

113

Honig, M. (1996). Retirement expectations: Differences by race, ethnicity, and gender.
The Gerontologist, 36(3), 373-3 82.

Hosmer, D.W., & Lemeshow, S. (2000). Applied logistic regression (2”). New York:
Wiley.

Johnson, R. W., & Kawachi.J. (2007). Job changes at older ages: eﬂects on wages,
benefits, and other job attributes. Washington, DC: Urban Institute. Retrieved

March 13, 2010, from
http://www.urban.org/retirement_policy/url.cfm?lD=—'3 1 1435.

J uster, F. T. (1998). Commentary: Career trajectories and the older worker. In K. W.
Schaie & C. Schooler (Eds), Impact of work on older adults (pp. 246-251). New

York: Springer Publishing Company.

Juster, F. T., Smith J. P., & Stafford, F. (1999).The measurement and structure of
household wealth, Labour Economics, 6 (2), 25876.

Juster, F., T. & Smith, J ., P. (1997). Improving the quality of economic data: Lessons
from HRS and AHEAD. Journal of the American Statistical Association, 92 (440),
1268-1278.

Juster, E, Thomas, J. L., James, P. S. & Stafford, F. P. (1999). The Measurement and
structure of household wealth, Labour Economcs, 6(2), 258-275.

Kampfe, C. M. (1994). Vocational rehabilitation and the older population. The American
journal of occupational therapy, 39(5), 325-326.

Katz, M. R., Rodin, G., & Devins, G M. (1995). Self-esteem and cancer; theory and
research. Canadian Journal of Psychiatry, 40(10), 608—615.

Kelly, S. D. M. (2003). Prevalent mental health disorders in the aging population: issues
of comorbidity and functional disability, Journal of Rehabilitation, 69(2), 19-25.

Kelly, T, (2003). Working with older people in groups: An exploration of universal
emotional reactions. 25th Annual Symposium on Social Work with Groups.
Boston, Massachusetts.

Kemp , B. Brummel-Smith, K., & Plowman, V. (1989). Geriatric rehab program focuses
on research, training and service. Journal of Rehabilitation, 55, 9-11.

Kemp, B. (1985). Rehabilitation and the older adult. In E. Birren & K. W. Schaie (Eds),
Handbook of the psychology of aging (2nd ed.). New York: Van Nostrand
Reinhold.

114

Kemp, B. (2000). Quality of life while aging with a disability. Retrieved July 25, 2002,
from http://www.jik.com/ qolkemphtml.

Kemp, B., & Mosqueda, L. (2004). Aging with a disability: What the clinician needs" to
know. Baltimore, MD: John Hopkins University Press.

Kennedy, M. (2000). Gaining access. American School & University, 73(4), 14-19.

Kerlinger, F. (1986). Foundations of Behavioral Research (3rd ed.). New York: Holt,
Rinehart & Winston.

Kim, H. & DeVaney, S. A. (2005). The selection of partial or full retirement by older
workers. Journal of Family and Economic Issues, 26(3), 371-394.

Kinjerski, V.M. and Skrypnek, B.J. (2004) Deﬁning spirit at work: Finding common
ground, Journal of Organizational Change Management, 17(1), 26-42.

Kraus, L. E. and Stoddard, S. (1991). Chart book on work disability in the United States.
An Info Use Report. Washington, DC: US. Department of Education, National
Institute on Disability and Rehabilitation Research.

Lahey, J .N. (2005). Do older workers face discrimination? Issue Brief, Center for
Retirement Research, No. 33. Boston, MA: Boston College, Center for
Retirement Research.

LaPlante, MP. (1996). Deﬁning Disability, presented at the annual meeting of the
American, Public Health Association in New York, NY.

Larkin,V.M., Alston, R.J., Middleton, R.A., & Wilson, KB. (2003). Underrepresented
ethnically and racially diverse aging populations with disabilities: Trends and
recommendations. Journal of Rehabilitation, 69(2), 26—3 1.

Lee S, Saito T, Takahashi, M., & Kai, I. (2008). Volunteer participation among older
adults in Japan: An analysis of the determinants of participation and reasons for
non-participation. Archives of Gerontology and Geriatrics, 47 (2), 177-187.

Leedy, P., & Orrnrod, J. (2001). Practical Research (7* ed). Merrill Prentice Hall.

Lewis, K. (1989). Persons with disabilities and the aging factor. Journal of
Rehabilitation, 55(4), 133-145.

Lockwood, NR. (2003). The aging workforce: The reality of the impact 0f older workers
and eldercare in the workplace. HR Magazine, 48(1), 1-11.

Long, J. S. (1997). Regression models of categorical and limited dependent variables.
Sage, CA:.Thousand Oaks

115

Loprest, P., & Maag, E. (2001). Barriers and supports for work among adults with
disabilities: Results from the NHIS-D. Washington, DC: The Urban Institute.

Loprest, P., Rupp, K., Sandell, SH, 1995, Gender, Disabilities, and Employment in the
HRS, Journal of Human Resources 30(11), 298-318.

Lyness, J.M,, Noel,T.K., Cox, C., King, D.A., Conwell, Y., & Caine, ED. (1997).
Screening for depression in elderly primary care patients. Archives of Internal

Medicine, 157(4), 449-453.

Maddala, G S. (1983). Limited—dependent and qualitative variables in econometrics.
Cambridge: University Press.

Maestas, N. (2007). Back to work: Expectations and realizations of work after retirement.

Rand Working Paper (WR-196-2). Retrieved July 5, 2007, from
http://www.rand.org/mibs/workingjapgrs/2007/RAND_WR l 96-Q3df

May, V., & Vieceli, L. (1983). Barriers to placement: strategies and resolution. Journal of
Rehabilitation, 49(3), 43—46.

McKenna, M.T., Michaud, C.M., Murray, C.J., & Marks, J .S.(2005). Assessing the
burden of disease in the United States using disability-adjusted life years.

American Journal of Preventive Medicine. 28(5), 415-423.

McNaught, W., & Barth, M. (1992). Are older workers “good buys”?: A case study of
Days Inns of America. Sloan Management Review, 33(3), 53-63.

McNeil, J.M. (1993). Americans with Disabilities: I991-92: Data from the Survey of
Income and Program Participation, Current Population Reports: Household
Economic Studies. Washington, DC: Bureau of the Census, Economics and

Statistics Administration.

McNeil, J.M. (1997). Americans with disabilities: 1994-95. US Bureau of the Census
Current Population Reports, Washington, DC: US Government Printing Ofﬁce.

Menard, S. (2002). Applied logistic regression analysis, 2nd Edition. Thousand Oaks,
CA: Sage Publications. Series: Quantitative Applications in the Social Sciences.

Menec, V. (2003). The relationship between everyday activities and successful aging: A
6-year longitudinal study. Journal of Gerontology: Social Sciences, 588(2), 74-82.

Michigan Office of Services to the Aging (2004). State of Michigan senior employment
services coordination plan. The United States Department of Labor - Division of

Older Worker Programs, Retrieved August 30, 2009, from
hnp://wwwmicliigan.gov/documents/miseniors/SCSEPPlan03 04F i nal_1 72796_L

116

pd_f.

Minter, S. G (2002). Ergonomic challenge: The aging workforce. Occupational Hazards.
Retrieved November 6, 2009, from
htgpz/lwwwprinchpalhealthnews.com/artic1e/bellhowel1/103057203.

 

Mitchell, J ., Adkins, R., & Kemp, B. (2006). The effects of aging on employment of
people with and without disabilities, Rehabilitation Counseling Bulletin, 49(3),
157-165.

Moon, M., & Hushbeck, J. (1989). Options for Extending Work Life. Generations, 8(2),
27-30.

Moon, M., & Juster, F. T. (1995). Economic status measures in the Health and Retirement
Study. Journal of Human Resources, 30(5), 138— 157.

Moore, C. L., Feist-Price, S., & Alston, R. (2002). Competitive employment and mental
retardation: Interplay among gender, race, secondary psychiatric disability and
rehabilitation services. Journal of Rehabilitation, 68 (1), 14-19

Mor-Barak, M. & Tynan, M. (1993). Older workers and the workplace: A new challenge
for social work, Social Work, 38 (1), 45-55.

Morton, L., Foster, L., & Sedlar, J. (2005). Managing the mature workforce. New York:
NY: The Conference Board.

Munnell, A. H., & Perun, P. (2006). An Update on Private Pensions. Issue in Brief, 50.
Chestnut Hill, MA: Center for Retirement Research at Boston College.

Munnell, A. H., Sundén, A., & Lidstone, E. (2002). How Important Are Private
Pensions? Issue in Brief, 8 (February), Chestnut Hill, MA: Center for Retirement
Research at Boston College.

Myers, J. E. (1992). Career and retirement counseling for older workers. In D. H.
Montross & C. J. Shinkman (Eds), Career development: Theory and practice (pp.
279-290). Springﬁeld, IL: Thomas.

Nagi $2. (1969). Disability and rehabilitation: Legal, clinical and self concepts and
measurement, Ohio University Press, Washington DC.

Nagi, 8.2. (1966). Some conceptual issues in disability and rehabilitation. In M. D.
Sussman (Ed), Sociology and Rehabilitation (pp. 100—113). Chicago, IL:

American Sociological Association.

National Institute of Disability Management and Research. (2005). Disability
management success: A global corporate perspective. Victoria, BC: National

117

Institute of Disability Management and Research

National Center for Health Statistics (2006). Health, United States, 2006 with chart-book

on trends in the health of Americans. Washington, DC: US Government Printing
Ofﬁce.

National Institute on Aging (2007). Growing older in America: The health & retirement
stuay. Bethesda, MD: Author.

Neumark, D. & Stock, W. A. (1999). Age discrimination laws and labor market efficiency.
Journal of Political Economy, 107(5), 1081-1125.

Oi, W. & Andrews, ES. (1992). A Theory of the Labour market for Persons with
Disabilities. Report for the US Department of Health and Human Services.

Penninx, B. W., Deeg, D. J., van Eijk, J. T., Beekman, A. T., & Guralnik, .l. M. (2000).
Changes in depression and physical decline in older adults: A longitudinal
perspective. Journal of Affective Disorders, 61(1), 1—12.

Petty, M., & Fussel, l. E. (1997). Employer attitudes and satisfaction with supported
employment. Focus on Autism and Other Developmental Disabilities, 9(1), 15-22.

Pitt-Catsouphes, M.& Matz-Costa, C. (2008), The multi-generational workforce:

workplace ﬂexibility and engagement, Community, Work & Family, 11(2), 215-
229.

Platt, R.K. (2001). The aging work force: Angst for organizations, baby boomers.
Workspan, 44(1), 26-29.

Poulos, S. & Nightingale, S. D. (1997). The Aging Baby Boom: Implications for
Employment and Training Programs, Retrieved November 6, 2009, from
http://www.urban.org/aging/abb/agingbabvhtml.

Purcell, P. (2009), Older workers: Employment and retirement trends, Journal of
Deferred Compensation, 14(2), 85-102.

Quinn, J. (1997). The Role of Bridge Jobs in the Retirement Patterns of Older Americans,
In Philip DeJong and Theodore Marmor (eds), The labour market (pp. 91-116).
Social Security and London: Ashgate Publishing Ltd.

Quinn, J. (2002). Changing Retirement Trends and Their Impact on Elderly Entitlement
Programs. In S. H. Altman and D. Shactman (Eds), Policies for on Aging Society
(pp. 29915). Baltimore, MD: Johns Hopkins University Press.

Quinn, J., Burkhauser, R. V., Cahill, K. & Weathers, R. (1998). Microeconometric
analysis of the retirement decision: United States. OECD Economic Department

118

Working Papers (Publication Number 203). Retrieved April . 5, 2010 from
hmx/lwwwoecdong/eco/eco.

Quinn, J., & Kozy, M. (1996). The role of bridge jobs in the retirement transition: gender,
race, and ethnicity. The Gerontologist, 36(3), 363-372.

Radloff, L. S. (1997). The CES-D scale: a self-report depression scale for research in the
general population, Applied Psychological Measurement. 1(3), 385—401.

RAND. (2006). The National Institute on Aging and the Social Security Administration.
RAND HRS Data, Version F.

Reimers, C., & Honig M. (1996). Responses to Social Security by men and women:
Myopic and far-sighted behavior. The Journal of Human Resources, 31(2), 359-
382.

Reiner, M. L. (1974). Aging—what does it mean? Journal of Rehabilitation, 40(5), 14-15.

Remenyi, A. (1995). Older Workers and Rehabilitation Counseling. The Australian
Journal of Rehabilitation Counseling, 1(1), 46-60.

Reynolds, 8., Ridley, N., & Horn, C. (2005). A work-filled retirement: Workers ’changing
views on employment and leisure (Survey No. 8.1). New Brunswick, NJ: John J.
Heldrich Center for Workforce Development, Rutgers University. Retrieved
August 30, 2009, from

http://www.heldrich.rutgers.edu/uploadedFiles/Publications/WT1 6 .pdf.

Reynolds, 8., Ridley, N ., & Van Horn, C. (2005). A work-filled retirement: Workers’

changing views on employment and leisure (Survey No. 8.1). New Brunswick,
NJ: John J. Heldrich Center for Workforce Development, Rutgers University.
Retrieved August 30, 2009, from
mpJ/wwwheldrich.rutgers.edu/uploadedFiles/Publications/WT16.pdf.

Rix, S. (2004). Aging and work—A view from the United States. Washington: AARP.
Retrieved August 30, 2009, from www.aarp.org/ppi.

Robert Wood Johnson Foundation (2006). Wisdom at work: the importance of the older
and experienced nurse in the workplace. Retrieved August 30, 2009, from
http://www.rwif.org/ﬁles/publications/other/wisdomatworkpdf.

Rogowski, J. & Karoly. L.(2000). Health Insurance and Retirement Behavior: Evidence
from the Health and Retirement Survey. Journal ofHealth Economics, 19(4), 529-
539.

Rosenthal, D. A., Hursh, N., Lui, J., Isom, R., & Sasson, J. (2007). A survey of current
disability management practice: Emerging trends and implications for

119

certiﬁcation. Rehabilitation Counseling Bulletin, 50(2), 76—86.

Ruhm, C. (1995). Secular Changes in the Work and Retirement Patterns of Older Men,
Journal of Human Resources, 30(2), 36595.

Santiago, A.M. & Muschkin, C. G (1996). Disentangling the effects of disability status
and gender on the labor supply of Anglo, Black and Latino older workers. The
Gerontologist, 36(3). 299 - 310

Saxon, J. P., & Spitznagel, R.J. (1992). Age bias: An Intervening variable in vocational
rehabilitation feasibility decision making? Journal of Applied Rehabilitation
Counseling, 23(1), 246.

Schechter, E. S. (1999). Industry, occupation and disability insurance beneﬁciary work
return, Social Security Bulletin, 62(1), 10-22.

Schwarz, N. (1998). Accessible content and accessibility experiences: The interplay of
declarative and experiential information in judgment. Personality and Social
Psychology Review, 2(2), 87—99.

Seeman, T. E., Unger, J. B., McAvay, G., & Mendes de Leon, C. F. M. (1999). Self-
effrcacy beliefs and perceived declines in functional ability: MacArthur studies of

successful aging. Journal of Gerontology: Psychological Sciences, 54B(5), 214-
222.

Shaw, L.R. & Betters, C. (2004). Rehabilitation counseling and consultation in the private
sector. In T.F. Riggar & D.R. Maki (Eds). The Handbook of Rehabilitation
Counseling (pp. 236-251). New York: Springer Publishing Co.

Shrey, D.E., & Lacerte, M.M.D. (1995). Principles and Practices of Disability
Management in Industry. Winter Park, FL: GR Press, Inc.

Sickles R.C. & Taubman, P. (1986). An Analysis of the Health and Retirement Status of
the Elderly, Econometric, 54(6), 1339-1356.

Social Security Administration. (2006). Income of the Population 55 and Older,
Washington, DC: Office of Policy.

SPSS Inc. (2006). SPSS Complex Samples 15.0. Chicago IL: SPSS Inc.

Stefﬁck D. Ann Arbor, MI: University of Michigan Survey Research Center; 2000.
Documentation of affective functioning measures in the Health and Retirement
Study (HRS/AHEAD Documentation Report No. DR-005). Retrieved August 30,
2009, from http://www.umich.edu/~hrswww/docs/userg/index.html.

Stensrud, R., & Gilbride, DD. (2004). Placement. In T. Riggar & D. Maki (Eds),

120

Handbook of rehabilitation counseling (pp. 218-235). New York, NY: Springer
Publishing Company, Inc.

Stern, S. (1989). Measuring the effect of disability on labor force participation. Journal of
Human Resources. 24(3), 361-395.

Stock, W.A., & Beegle, K. (2004). Employment protections for older workers: Do
disability discrimination Laws matter? Contemporary Economic Policy, 22(1),

111-126.

Swett, E. A., & Bishop, M. (2003). Mental health and the aging population: Implications
for rehabilitation counselors. Journal of Rehabilitation, 69(2), 178-189.

Szinovacz, M., & Davey, A. (2004). Honeymoons and joint lunches: Effects of retirement
and spouse’s employment on depressive symptoms. Journal of Gerontology:

Psychological Sciences, 59(5), 233-245.

Szymanski, E. M. (1985). Presidential address. Rehabilitation Counseling: A profession
with a vision, an identity, and a future. Rehabilitation Counseling Bulletin, 29(1),

2-5.

Szymanski, B, Ryan, C., Merz, M., Trevino, B., & Johnston-Rodriguez, S. (1996).
Psychosocial and economic aspects of work: Implications for people with
disabilities. In E. M. Szymanski & R. M. Parker (Eds), Work and disability (pp.

9—38). Austin: PRO-ED.

Tabachnick, B. G, & Fidell, L. S. (2007). Using Multivariate Statistics, 5th ed. Boston:
Allyn and Bacon.

Turvey, C. L., Wallace, R. B., & Herzog, R. (1999). A revised CES-D measure of
depressive symptoms and a DSM-based measure of major depressive episodes in

the elderly. International Psychogeriatrics, 11(2), 139-148.

The United States Government Accountability Ofﬁce (2001). Older Workers:
Demographic Trends Pose Challenges for Employers and Workers (Publication
Number 02-85). Government Accountability Office Reports Opiie via GPA
Access, Retrieved August 30, 2009, from Igg//www4.gg).gﬁov/cgi-

bin/gerggevo-oz-ss.

The United States Government Accountability Office (2003). Older Workers:
Employment Assistance Focuses on Subsidized Jobs and Job Search, but Revised

Performance Measures Could Improve Access to Other Services (Publication
Number 0950). Government Accountability Office Reports Opiie via GPA Access,
Retrieved August 30, 2009, from http://www.gao.gov/new.items/d03350.pdf.

Toossi, M. (2004). Labor force projections to 2012: the graying of the US. workforce.

121

Monthly Labor Review, 127(2), 37- 57.

Trends Research Brief (2008, June). Trends in health and disability among adults
approaching retirement age: Summary of evidence presented at a meeting of the
TRENDS Network, Kalamazoo, University of Michigan, and Retrieved March 12,
2010, from mpd/trendspsc.isr.umich.edu/rxl@ubs/TRENDSResBrief-02-20-

09.pdf.

Trends in Health and Disability among Adults Approaching Retirement Age (2008).
Center for the Demography of A ging, University of Michigan, Retrieved April, 24,
2010, from http://trends.psc.isr.umich.edu/.

Turner, J ., & Turner, R. (2004). Physical disability, unemployment, and mental health.
Rehabilitation Psychology, 49(3), 241-249.

Turvey C.L., Wallace, R.B., & Herzog, R. (1999). A revised CES-D measure of
depressive symptoms and a DSM based measure of major depressive episodes in
the elderly. International Psychogeriatrics, 11(2), 139- 148.

US. Census Bureau. (2007). US. census press releases: New census bureau data reveal
more older workers, homeowners, non-English speakers. Washington, DC: US.
Census Bureau. Retrieved November 6, 2009 from http://www. census. gov/Press-
Release/www/releases/archives/american_community_survey_acs/010601 .html.

US. Department of Labor, Bureau of Labor Statistics. (1996). BLS Reports on the

Amount of Formal and Informal Training Received by Employees. News USDL
96-515.

U.S. Equal Employment Opportunity Commission. (1997). Facts about age
discrimination. Retrieved October 21, 2009 from http://wwweeocgo.

 

United States Government Accountability Ofﬁce (2001). Older workers: demographic
trends pose challenges for employers and workers (Publication Number 02-85)
Retrieved April 15, 2005, from Government Accountability Ofﬁce Reports Opiie
via GPA from http://wiﬂo.ggv/cgi-bin/get_r;ﬁGAO-02-85.

 

UnumProvident Company. (2005). Health and productivity in the aging American
workforce: Realities and opportunities. Chattanooga. TN: UnumProvident
Company.

Unutzer, J., Patrick, D. L., Simon, G. D., Grembowski, E., Walker, C. R., & Katon, W.
(1997). Depressive symptoms and the cost of health services in HMO patients

aged 65 Years and older. A 4-year prospective study. Journal of the American
Medical Association, 277(20), 1618-1623.

Vargo, T.J., & Grzanowicz, G. W. (2002). Strategies for effective disability management.

122

Employee Benefit Plan Review, 57(6), 24-28.

Wadsworth, J & Kampfe, C. M. (2004). The characteristics of senior applicants for
vocational rehabilitation services. Rehabilitation Counseling Bulletin, 47(2), 104-

112.

Wallace R, Herzog AR, Ofstedal MB, Stefﬁck D, Fonda S, Langa K. (2000).
Documentation of aﬂective functioning measures in the health and retirement
study. University of Michigan; Ann Arbor, MI.

Walls, R. T., Misra, S., & Majumder, R. K. (2002). Trends in vocational rehabilitation:
1978, 1988, 1998. Journal of Rehabilitation, 68(3), 4—10.

Wray, L. A. & Blaum, CS. (2001). Explaining the role of sex on disability: A population-
based study. Gerontologist. 41(4), 499—5 10.

Zola, I. K. (1989). Aging and disability: toward a uniﬁed agenda. Journal of
Rehabilitation, 55(4), 6-8.

Zwerling, C. W., Paul, S., Sprince, N. L., Davis, C. 8., Wallace, R. B., Blanck, P. D., &

Heeringa, S. G (2002), Workforce participation by persons with disabilities,
Journal of Occupational and Environmental Medicine, 44(4), 358-364.

123

|
l
|
I

3