Page 300 - Discrimination at Work The Psychological and Organizational Bases

11. PERSONALITY-BASED UNFAIR DISCRIMINATION
267
validity problem. As Stone-Romero (1994) demonstrated, many measures
of NA have items that deal with stressor-induced psychological and phys­
ical strain. Strain is "a set of psychological and physiological responses to
stressors, including long-term problems with physical and mental well­
being" (Stone-Romero, 1994, p. 460).
One of the many examples of an NA measure that has considerable
strain content is Taylor's (1953) Manifest Anxiety Scale. Samples of items
in this measure are "I have diarrhea once a month or more," "I work under a
great deal of tension," "I have a great deal of stomach trouble," and "Life is
a strain for me much of the time." Items such as these overlap considerably
with items in measures of role-related stress and strain (e.g., the measures
of Caplan, Tripathi, & Naidu, 1985; House & Rizzo, 1972; Karasek, 1979;
Patchen, 1970). Thus, it is clear that many measures that purport to index
NA have suspect construct validity because they include items that index
stress and/or strain. To the extent that they actually index strain, the use of
such measures for selection purposes would appear to violate the spirit, if
not the letter, of the Americans with Disabilities Act.
Interestingly, some researchers have noted that responses to a host of
measures typically used to index psychopathology may represent nothing
more than the presence of what Frank (1973) called demoralization, defined
as a set of psychological symptoms that develop when individuals find
themselves in environments that are stressful and that cannot be escaped
(Dohrenwend & Dohrenwend, 1981). The importance of this is that strain
is a consequence of exposure to stressors, as opposed to an immutable
trait (Stone-Romero, 1994). In support of this argument, the next section
considers literature that is relevant to the view that strain is a consequence
of exposure to stressors, not an unchangeable property of individuals. In
addition, it demonstrates how the use of personality measures for selection
purposes may lead to unfair discrimination.

SOCIAL CAUSATION MODELS OF WELL-BEING

A rather large body of literature exists on the social causation perspec­
tive on mental health or psychological well-being (Dohrenwend, 1975;
Dohrenwend & Dohrenwend, 1974; Taylor, Repetti, & Seeman, 1997). It
argues that the chronic stress and strain that is experienced by many
individuals is, in large measure, a function of the characteristics of their en­
vironments (eg., work, home, community; see Dohrenwend, 1975; Dohren­
wend & Dohrenwend, 1974, 1981; Kessler, 1979; Link, Dohrenwend, &
Skodol, 1986; Link, Lennon, & Dohrenwend, 1993; Taylor et al., 1997;
Williams & Collins, 1995; Yu & Williams, 1999). This is a crucial issue