AGE DISCRIMINATION

O R I G I N A L P A P E R

Development and Validation of the Workplace Age
Discrimination Scale

Lisa A. Marchiondo1 • Ernest Gonzales2 • Shan Ran3

Published online: 9 December 2015

� Springer Science+ Media New York 2015

Abstract

Purpose Workplace age discrimination research is prolifer-

ating, but researchers lack a valid measure with which to capture

targets’ discriminatory experiences. We developed a measure of

perceived workplace age discrimination that assesses overt and

covert forms of discrimination and then compared older, mid-

dle-aged, and younger workers’ experiences.

Design/Methodology In Study 1, we developed the

Workplace Age Discrimination Scale (WADS) based on

older workers’ experiences using a deductive approach, a

qualitative study, and two quantitative surveys. In Study 2,

we validated the measure among young employees using a

qualitative and two quantitative surveys. In Study 3, we

tested the WADS among middle-aged workers and tested

models of invariance between age groups.

Findings Participants frequently endorsed covert dis-

criminatory experiences, which the WADS reflects. The

WADS contains convergent and discriminant validity, high

reliability, and a unidimensional structure across age groups.

It demonstrates criterion-related validity among older and

younger workers but not middle-aged workers, given their

low experiences of age discrimination. Age discrimination

frequency follows a U-shaped pattern across age groups.

Implications Researchers can use the WADS to identify

long-term outcomes of age discrimination and to further

compare workers’ discriminatory experiences. Practition-

ers and policymakers can use the measure to develop

interventions to ameliorate workplace age discrimination

and inform policymaking.

Originality/Value The WADS is the first validated mea-

sure of targets’ perspectives of workplace age discrimina-

tion. Our results challenge assumptions that only older

workers experience age discrimination (younger workers’

means were highest) and that age discrimination is usually

overt in nature (it is often covert).

Keywords Ageism � Age discrimination � Measurement �
Older workers � Middle-aged workers � Young workers �
Modern discrimination

… I wonder whether employers, or whether the
public generally realizes that age discrimination is

illegal…
Stuart Ishimaru, Chairman, U.S. Equal Employment

Opportunity Commission (EEOC 2009)

The opening quote illustrates the degree to which age

discrimination is tacitly accepted within U.S. society. This

phenomenon is not unique to the U.S. though; reports of

age discrimination abound globally (e.g., Balch 2015;

Hock 2015; Medhora 2015). The number of workplace age

discrimination claims through federal human rights agen-

cies is growing, reflecting the negative climates that a

growing number of workers face (e.g., EEOC 2013).

& Lisa A. Marchiondo
[email protected]

Ernest Gonzales

[email protected]

Shan Ran

[email protected]

1
Anderson School of Management, University of New

Mexico, MSC05 3090, 1 University of New Mexico,

Albuquerque, NM 87131-0001, USA

2
School of Social Work, Human Behavior Department, Boston

University, 264 Bay State Road, Boston, MA 02215, USA

3
Department of , Wayne State University, 5057

Woodward Avenue, Detroit, MI 48202, USA

123

J Bus Psychol (2016) 31:493–513

DOI 10.1007/s10869-015-9425-6

http://crossmark.crossref.org/dialog/?doi=10.1007/s10869-015-9425-6&domain=pdf

http://crossmark.crossref.org/dialog/?doi=10.1007/s10869-015-9425-6&domain=pdf

This prevalence may not be surprising, given that age-

ism is often not considered as offensive or unjust as other

highly researched forms of prejudice such as racism and

sexism (Deal et al. 2010; Levy and Banaji 2002).

Employees hold numerous, pervasive stereotypes about

older adults, including beliefs that they are less competent

and adaptable than other workers (Cuddy et al. 2005;

Posthuma and Campion 2009). Stereotypes such as these

ultimately manifest in discriminatory behavior, including

lower rates of hiring and promoting older workers (Post-

huma et al. 2012). Given estimates that by 2020 one in four

U.S. workers will be age 55 or older (Hayutin et al. 2013)

and one in three U.K. workers will be over age 50

(Department for Work and Pensions 2013), the prevalence

of and tolerance for ageism is concerning, as more workers

may become targets.

The ability to adequately study and address age dis-

crimination is hampered by several limitations in the lit-

erature. First, coworkers’ and supervisors’ (i.e.,

‘‘perpetrators’’’) negative age stereotypes and discrimina-

tory behavior are well documented, but the literature

speaks less to older employees’ (i.e., ‘‘targets’’’) experi-

ences of such discrimination. Targets’ perspectives are

important—often essential—to capture in order to under-

stand the impact of discrimination on targets’ productivity

and well-being. Second, the literature attends primarily to

overt forms of discrimination (e.g., refusal to hire) and

often ignores low-intensity, covert discriminatory behav-

iors (e.g., social exclusion). Consistent with modern dis-

crimination theories, such as interpersonal discrimination

(Hebl et al. 2002) and selective incivility (Cortina 2008),

manifestations of workplace discrimination are not as

explicit as in times past (Marchiondo et al. 2015). Covert

discrimination more closely reflects the experiences of

many stigmatized groups and occurs more frequently than

overt discrimination (Madera and Hebl 2013), warranting

its inclusion in age discrimination research. Finally,

although emerging scholarship has begun exploring older

workers’ experiences of discrimination, a validated mea-

sure of workplace age discrimination from the target’s

perspective does not exist. Proxy and convenience items

prevent accurate measurement of targets’ experiences.

Given these limitations, our initial objective was to develop

a reliable and valid measure that captures older employees’

experiences of workplace age discrimination in its many

forms.

Though sparse, increasing literature suggests that

younger workers also face discriminatory treatment (e.g.,

Snape and Redman 2003). Despite their seeming dissimi-

larities, younger and older workers alike possess less

influence and fewer resources than middle-aged workers

(North and Fiske 2012). Both groups are subject to more

negative stereotypes (Finkelstein et al. 2013), which may

give rise to discriminatory treatment. Further, the nature of

age stereotypes (and discrimination) might be changing,

with younger workers being judged more negatively than

older workers (Bertolino et al. 2012; Finkelstein et al.

2013; Weiss and Maurer 2004). As such, the second

objective was to apply our measure of age discrimination to

younger workers to capture their experiences.

We contend that ageism operates dynamically across the

working lifespan and that vulnerable out-groups include

older and younger workers, with middle-aged workers

constituting the in-group. To test this idea, the third

objective was to administer our age discrimination measure

to middle-aged workers and test a curvilinear pattern of

discrimination across age groups.

Negative Attitudes and Discrimination Toward
Older Employees

Workforces worldwide are ‘‘graying.’’ Increasing longevity

is extending the average work life, as employees are

physically able to work longer and prefer to do so for social

and psychological benefits (Freedman 2008). Employees

also work longer as the financial landscape of retirement

evolves in many countries and more employees are finan-

cially unprepared for retirement (Munnell and Sass 2008).

These forces, among others, are contributing to the growth

of the older workforce, underscoring the importance of

understanding older workers’ experiences.

Ample research has attended to attitudes toward older

employees. Stereotypes of older adults in general are

ubiquitous, overwhelmingly negative, and resistant to

change (Cuddy et al. 2005). Stereotypes of older employ-

ees include perceptions that they have lower performance,

possess lower potential for training and development,

refrain from change (Posthuma and Campion 2009; Rosen

and Jerdee 1976), have poorer interpersonal skills (Bal

et al. 2011), and are less healthy and trusting of others (Ng

and Feldman 2012). Measures designed to capture ageist

attitudes toward older adults emerged at least 60 years ago

(e.g., the Old People Questionnaire by Tuckman and Lorge

1953) and have undergone numerous iterations since (e.g.,

the Aging Semantic Differential by Rosencranz and

McNevin 1969; the Fraboni Scale of Ageism by Fraboni

et al. 1990). Thus, scholars have long recognized the need

to validly capture attitudes toward older adults.

Many researchers, too, have devoted attention to

employers’ and coworkers’ discriminatory behavior. For

instance, employers assign lower ratings to older adults for

hiring (Avolio and Barrett 1987), advancement (Bal et al.

2011), promotion and training (Rosen and Jerdee 1976),

and economic worth to organizations (Finkelstein and

Burke 1998). They also provide harsher recommendations

494 J Bus Psychol (2016) 31:493–513

123

following poor performance (Rupp et al. 2006). It is clear

that older adults often face unreceptive climates at work.

Significantly less is known about older workers’ per-

ceptions of discrimination—perspectives that may drive

their job attitudes and success, mental and physical well-

being, and retirement plans. The age discrimination liter-

ature exists in stark contrast to research on sex- and race-

based discrimination in this respect, the latter of which

have largely centered on targets’ experiences and subse-

quent well-being. For instance, targets of sexual harass-

ment experience greater job withdrawal and lower

psychological and physical health (Fitzgerald et al. 1997).

Racial discrimination, too, is associated with adverse

physical and mental health (Williams et al. 1997; Williams

et al. 2003). Most of these deleterious effects have not been

linked to age discrimination or have been done so using

invalidated measures. We echo calls by Ruggs et al. (2013)

and North and Fiske (2012) for more research on age

discrimination. We extend these calls by highlighting the

need for greater attention to targets’ perspectives, which we

address in this paper.

Negative Attitudes and Discrimination Toward
Younger Employees

Researchers and policymakers primarily attend to the

treatment of older employees (Finkelstein et al. 2013),

given their increasing numbers and legal protection in the

U.S. However, the prevalence of age discrimination toward

older workers may be changing (Weiss and Maurer 2004),

as age stereotypes appear to favor them over younger

workers (Bertolino et al. 2012; Finkelstein et al. 2013).

Expanding the scope of the ageism literature to capture

discrimination toward younger employees is important

given contentions that we have an ‘‘ageist ageism litera-

ture’’ that centers on the experiences of older adults

(Rodham 2001).

Although nascent, research on younger employees has

clearly demonstrated that they, like older employees, are

subject to a host of negative stereotypes. These include

assumptions that they are disloyal, inexperienced, unmo-

tivated, immature, irresponsible, and selfish (cf. Finkelstein

et al. 2013; Snape and Redman 2003). Younger workers

are rated lower than older adults on desirable personality

traits, such as conscientiousness, emotional stability, and

agreeableness, as well as on performance-related variables,

such as organizational citizenship behavior (Bertolino et al.

2012). They are also perceived less favorably than older

workers in terms of initiative, stability, and work experi-

ence (Gibson et al. 1993). Negative attitudes toward

younger people are not new; Baby Boomers and subse-

quent generations were all targeted with stereotypes of

entitlement and laziness when they were young (Deal et al.

2010). Middle-aged employees are particularly likely to

hold these negative stereotypes, assigning significantly

more undesirable stereotypic traits to younger workers than

positive ones (Finkelstein et al. 2013).

These negative stereotypes lead to discriminatory prac-

tices. Younger workers face more denials of promotion,

fewer opportunities for training and development, dispro-

portionately lower pay and benefits, restricted freedom and

responsibility, and increased vulnerability to layoffs

(Duncan and Loretto 2004; Loretto et al. 2000; Snape and

Redman 2003). Thus, researchers are uncovering the sur-

prisingly prevalent negative attitudes and treatment of

younger workers. These findings broaden the call to cap-

ture targets’ perceptions of age discrimination by attending

to younger employees’ perspectives too.

Targets’ Perceptions of Workplace Age
Discrimination

Although small in number, some studies have attended to

employees’ personal experiences and outcomes of age

discrimination at work. For instance, among workers age

50 and older in the Midlife in the United States II (MIDUS

II) dataset, 81 % reported at least one instance of age

discrimination in the past year (Chou and Choi 2011). For

older targets, age discrimination correlates negatively with

variables such as job satisfaction, organizational commit-

ment, life satisfaction, and job involvement, and correlates

positively with turnover intentions (Minnotte 2012; Orpen

1995; Redman and Snape 2006). This work has drawn

attention to the importance of studying employees’ per-

ceptions of age discrimination.

A small related body of work has addressed older

workers’ perceptions of organizational climates of ageism.

For example, Furines and Mykletun (2010) developed a

measure of Nordic employees’ perceptions of discrimina-

tory age climates with regard to formal aspects of the job

(e.g., training, promotion). Kunze et al.’s (2011) measure

of macro-level age discrimination aggregates employees’

perceptions of discrimination stemming from the organi-

zation (e.g., with regard to performance assessment). In

contrast to these measures of organizational climate, we

sought to capture employees’ personal experiences as tar-

gets of age discrimination. An individually-focused mea-

sure not only sheds light on discrete experiences of ageism,

but permits meaningful research on targets’ outcomes of

age discrimination.

Despite calls for greater research on age discrimination

(Ruggs et al. 2013), researchers lack a measure with which

they can validly capture employees’ experiences of it.

Numerous methodological issues exist. First, studies often

J Bus Psychol (2016) 31:493–513 495

123

contain convenience items, not developed based on theory

(a deductive approach) or respondents’ experiences (an

inductive approach). Many contain single-item measures,

which do not capture the breadth of age discrimination

(i.e., content adequacy). Second, few measures have

undergone validation procedures (for exceptions, see

measures of discriminatory age climate by Furines and

Mykletun 2010 and Kunze et al. 2011). Third, some mea-

surement intended to capture targets’ personal experiences

is confounded by the inclusion of items that assess

respondents’ perceptions of discriminatory climates as well

as of coworkers’ discriminatory experiences (Arvey and

Cavanaugh 1995). For instance, items such as ‘‘Young/old

adults as a group have been victimized,’’ ‘‘The best jobs

here are reserved for younger workers,’’ and ‘‘[There is a]

lower chance of promotion for older workers’’ capture

respondents’ beliefs about the discriminatory experiences

of age group members as a whole. Combining these items

with ones about a respondent’s personal experience (e.g.,

‘‘I have been overlooked here because of my age’’) makes

it difficult to determine whether responses are based on

one’s personal experience as a target, observations of

coworkers as targets, or general rumors of discrimination

within a company. This imprecision precludes clear and

accurate operationalization of the construct. Fourth, most

surveys do not specify a timeframe in which respondents

recall discriminatory experiences (e.g., ‘‘How frequently in

the last year…’’). The absence of a timeframe allows some
participants to focus on recent experiences and others to

draw on experiences from decades past, thereby introduc-

ing varying degrees of memory bias and measurement

error.

Fifth, some age discrimination items rely on participants

to define age discrimination (e.g., ‘‘I have experienced age

discrimination’’). This practice introduces significant

measurement error, as participants’ definitions of age dis-

crimination may vary widely (Hardy and Ford 2014) or

encompass only egregious or commonly discussed events

(Fuegen and Biernat 2000). If targets do not comprehend a

term in the same way as the researcher, lexical compre-

hension suffers, undermining the measure’s validity (Hardy

and Ford 2014). Indeed, targets can report experiencing the

same mistreatment behaviors as one another but differen-

tially define them (Fitzgerald and Shullman 1993). To

ensure consistency of interpretation and ultimately validity,

mistreatment items should concern specific behaviors and

avoid the inclusion of perceptual labels such as ‘‘age dis-

crimination’’ (Arvey and Cavanaugh 1995; Fitzgerald and

Shullman 1993).

Finally, most age discrimination items address only for-

mal aspects of the job, such as selection, promotion, and

training. While important, this lens ignores insidious forms

of discrimination that some workers may experience with

greater frequency. According to modern discrimination

theories, such as interpersonal discrimination (Hebl et al.

2002) and selective incivility (Cortina 2008), discriminatory

treatment has shifted toward more subtle manifestations.

Underhanded behaviors, such as social exclusion and gossip,

can be defended as non-discriminatory and as such, skirt the

radar of legal and organizational policies. For instance,

selective incivility theory states that lower-intensity, disre-

spectful acts serve to marginalize and undermine work-

ers belonging to stigmatized groups, constituting a less

explicit, ‘‘modern’’ method of discrimination (Cortina

2008). Although modern discrimination is less overt than

formal discrimination, its negative effects are of comparable

magnitude (Jones et al. 2013), highlighting the importance of

this perspective given that employers may not treat it as

seriously as formal discrimination.

Modern discrimination theories have primarily been

applied to uncovering the experiences of women and

people of color but should be extended to explain workers’

experiences based on age (Marchiondo et al. 2015). Older

workers might be especially vulnerable to covert discrim-

ination due to stereotypes that older adults generally are

warm but less competent (Cuddy et al. 2005). Members of

‘‘pitied’’ groups (those perceived as warm but incompetent)

are more likely to experience interpersonal isolation and

other covert discrimination (Cuddy et al. 2007). In the U.S.,

older workers might also experience heightened covert

discrimination given that they (but not younger workers)

are legally protected from age discrimination. s may

quell the prevalence of formal discrimination, yet negative

stereotypes can continue to manifest through covert dis-

crimination that operates outside legal boundaries. We

draw on modern discrimination theories during scale

development, ensuring that covert experiences of age dis-

crimination are captured.

Comparing Age Discrimination Between Age
Groups

We contend that a validated measure of age discrimination

among those most likely to experience it (younger and

older workers) is needed. We recognize that middle-aged

workers could face age discrimination as well, albeit likely

with lower frequencies based on the age stereotype litera-

ture. Employees at either end of the age spectrum are most

vulnerable to age-based stereotypes and discrimination

(Duncan and Loretto 2004; Gee et al. 2007; North and

Fiske 2012). Snape and Redman (2003) found that younger

workers experience age discrimination with similar if not

greater frequency as older workers. This result is consistent

with evidence of the changing landscape of age stereotypes

in which younger workers are viewed less positively than

496 J Bus Psychol (2016) 31:493–513

123

older workers (Bertolino et al. 2012; Finkelstein et al.

2013). In contrast, employees hold the most positive (and

fewest negative) stereotypes of middle-aged workers

(Finkelstein et al. 2013), possibly exposing them to the

least age discrimination. Middle-aged workers’ privilege in

this respect can be explained by their possession of greater

status, resources, and wealth than other age groups

(Garstka et al. 2004; North and Fiske 2012). Middle-aged

workers constitute an ‘‘idealized standard against which

other age groups are judged’’ (Finkelstein et al. 2013, p. 21)

and are posited to hold the highest social status with regard

to age (Garstka et al. 2004). Thus, a U-shaped pattern of

age discrimination likely exists, with younger and older

workers being prime targets. In Study 3, we administer our

age discrimination measure to middle-aged workers and

test the following hypothesis:

Hypothesis Age discrimination experiences are curvi-

linear across age groups, such that younger and older

workers experience the highest frequencies and middle-

aged workers experience the lowest frequency.

The Present Studies

We conducted three studies based on participant age group

(older, younger, middle-aged), each of which contained an

iterative series of qualitative and quantitative phases, to

create and validate the Workplace Age Discrimination

Scale (WADS). Study 1 addressed older workers’ experi-

ences of age discrimination. Through four phases, we

deductively generated a pool of items, inductively devel-

oped items using a qualitative survey, reduced the item list

several times, and tested confirmatory factor structure, as

well as multiple types of validity. Study 1 also included an

experimental design to determine an appropriate timeframe

for the measure’s stem. In Study 2, we extended the WADS

to younger workers. Through three phases, we used

inductive item generation and exploratory factor analysis to

determine if the items adequately captured younger work-

ers’ experiences. We also tested the measure’s confirma-

tory factor structure and validity among this group. In

Study 3, we administered the WADS to middle-aged

workers and tested confirmatory factor structure and mul-

tiple types of validity. We then conducted measurement

invariance tests to compare the WADS between the three

age groups.

The use of self-report was appropriate—indeed, neces-

sary—given our emphasis on employees’ perceptions of

age discrimination targeted at them. We followed four

recommendations for reducing common method bias

(Conway and Lance 2010; Podsakoff et al. 2003, 2012),

which can strengthen or weaken relationships between

variables (Chan 2009). First, we selected established

measures with high reliability and validity. Next, all sur-

veys were anonymous, which reduces pressure to respond

in a consistent and socially desirable manner. Third,

mistreatment measures were placed later in the survey than

non-mistreatment measures so that recollection of

mistreatment did not influence responses to other con-

structs. Finally, scale formats (e.g., scale type, anchor

labels, polarity) varied between measures, reducing

anchor- and end-point biases.

A Note About Study Samples

We collected samples via crowdsourcing (Amazon

Mechanical Turk; MTurk) for several important reasons.

First, the diversity of an organization affects employees’

reports of discrimination (Avery et al. 2008) and emotional

conflict (Pelled 1996). Organizational age distributions,

specifically, can influence communication (Zenger and

rence 1989) and emotional conflict (Pelled et al. 1999)

between employees. As a result, we avoided sampling

single organizations due to their unique organizational

demographies that could shape participants’ experiences of

age discrimination. Crowdsourcing addresses these issues

by providing samples of workers across hundreds of

organizations (and age distributions, industries, etc.),

thereby increasing the generalizability of the measure

(Aguinis and al 2012). This decision was reinforced by

literature demonstrating that organizational samples are

vulnerable to similar convenience sampling limitations as

crowdsourcing (Landers and Behrend 2015) and that

MTurk is a valid and reliable source of data collection

(Barger et al. 2011; Buhrmester et al. 2011). Large and

diverse samples are also recommended for scale develop-

ment (Clark and Watson 1995). Finally, organizational

leaders are often unwilling to allow researchers to conduct

discrimination-related studies within their companies,

fearing reputational harm and lawsuits.

Study 1: Older Workers

Phase 1: Deductive Item Generation

Drawing on Hinkin’s (1998) deductive approach to item

generation, the Phase 1 goals were to define our concep-

tualization of age discrimination and to generate items by

evaluating the literature. First, each author independently

defined age discrimination and identified its essential fea-

tures. The authors discussed their definitions until they

agreed on the following operationalization of workplace

age discrimination: (a) it is a behavioral manifestation of

J Bus Psychol (2016) 31:493–513 497

123

prejudice and negative stereotypes; (b) the behaviors are

unjust, disrespectful, and/or unfavorable; (c) one or mul-

tiple parties (targets, observers, perpetrators) can define the

behavior as unjust, disrespectful, or unfavorable (consistent

with definitions of other mistreatment constructs, such as

workplace incivility; Andersson and Pearson 1999); (d) the

behaviors occur in the work context (for pay or volunteer)

and may stem from supervisors, coworkers, customers, or

any other member of one’s work environment; and (e) the

behaviors can be overt but, consistent with modern dis-

crimination theories, may also be covert. Although

numerous age ranges have been used to define ‘‘older

employees,’’ we selected age 50 and older, consistent with

cutoffs that aging experts have used in nationally repre-

sentative and federally funded studies (e.g., the Health and

Retirement Study). Further, studies of age stereotyping

have defined older adults as being over ages 50–52 (e.g.,

Finkelstein et al. 2013; Fritzsche and Marcus 2013).

Next, each author generated survey items by reviewing

the literatures on ageism and workplace discrimination, as

well as by drawing on his/her expertise to produce new

items. We compared our lists, removing items that were

redundant or did not meet the facets of the construct’s

definition. This approach yielded 55 items.

Phase 2: Inductive Item Generation and Initial Item

Reduction

Because limited research was available on age discrimi-

nation from targets’ perspectives, the next goal was to

inductively generate items (Hinkin 1998) by collecting and

examining qualitative data from older employees. This

approach ensured that items encompassed the breadth of

their experiences, particularly with regard to covert age

discrimination.

Participants and Procedure

We recruited N = 96 U.S. employees age 50 and older (all

currently working for pay) from MTurk. We prescreened

participants by asking them to select ranges for their age,

hours of work per week, and country of employment. At

the end of the survey, participants provided their specific

ages and work hours per week, providing another eligibility

check.
1

The average age was 56.5 years (SD = 4.5 years),

average organizational tenure was 15 years (SD =

7.3 years), 75 % worked 35 or more hours per week,

44.8 % were male, 88.5 % were White, 11.5 % were

African American, and 8.3 % were Latino (in all studies in

this paper, participants could select more than one race).

Participants read a definition of age discrimination:

‘‘Differential workplace treatment based on age, which

impairs fairness of treatment or opportunity.’’ Based on this

definition, participants described in as much detail as

possible a time they personally experienced age discrimi-

nation at work. Fourteen participants left the question

blank.

Analytic Method and Results

The authors reviewed each participant’s story and dis-

cussed whether items …