cause behaviors to be condoned while intentions related to selWshness, aggressive
tendencies toward resolving conXict, and a lack of integrity might create blame.
Some organizational (change) practices can attempt to control these motivational
factors, but selection devices may also help by identifying people with character-
istics that moderate or neutralize the factors naturally.
In an early meta-analysis, Ones et al. ( 1993 ) showed that integrity tests predict
counterproductive work behaviors, although validities were better for broader
outcome measures than theft alone. More recent work has also shown that integrity
tests are generally valid predictors of counterproductive outcomes (e.g. Fortmann
et al. 2002 ). Bennett et al. ( 2005 ) describe moral identity as a self-regulating
mechanism that prevents the pursuit of unethical behaviors. At the same time,
Sackett and Wanek ( 1996 ) point out a number of issues that may limit the
successful use of integrity tests, including overlap with personality constructs and
legal rights to privacy.
As suggested by Robinson and Bennett ( 1997 ), visible deviance could also serve as
an emotional or instrumental expression aimed at remedying an injustice (Neuman
2004 ). Spector and Fox ( 2002 , 2004 ) introduced formal models in which aVect,
stress, and perceptions of control mediate the inXuence of aVective dispositions and
situational factors on counterproductive behaviors. Penney and Spector ( 2002 )
found evidence supporting a model of trait anger expressed as aggression, where
trait anger was triggered by threats to one’s narcissistic view. However, college
students who were not necessarily working comprised their sample, necessitating
future investigations. Marcus et al. ( 2002 ) validated a counterproductive behavior
scale and later used it to show that self-control (i.e. the higher-level ability to
consider and weigh short- and long-term consequences and to delay gratiWcation)
predicted general counterproductive behaviors in a German organization (Marcus
and Schuler 2004 ). In any case, the implication of such research is that organizations
may wish to select individuals who are able and willing to express themselves in a
constructive or mature manner, and who will remedy injustices through formally
sanctioned means (e.g. communicating with the supervisor directly).
Other predictors of negative work behaviors include attitudes toward risk-taking
and career orientation (Harris and Ogbonna 2002 ). In applying the General
AVective Aggression Model (Anderson 1997 ) to work behaviors, Neuman ( 2004 )
suggests that personality, self-monitoring, beliefs and values regarding aggression,
and self-esteem are indirect determinants of aggression. Although aspects of
personality have typically shown small relationships with deviance (Bennett and
Robinson 2003 ), Colbert et al. ( 2004 ) found, in four samples, that conscientious-
ness, agreeableness, and emotional stability moderated the eVect of environmental
work factors (i.e., developmental opportunities and support) on organizational
deviance. Interestingly, Bennett and Robinson ( 2003 ) propose that ethnic cultural
variables related to ethnocentrism, cooperation, and collectivism might identify
employees who avoid deviant behaviors.
selection decision-making 313