Who gets the benefit of the doubt? The impact of causal reasoning depth on how violations of gender stereotypes are evaluated

• Date: 01 March 2018
• Published date: 01 March 2018
• DOI: http://doi.org/10.1002/job.2218

RESEARCH ARTICLE

Who gets the benefit of the doubt? The impact of causal

reasoning depth on how violations of gender stereotypes are

evaluated

Steffen Keck

1

|Linda Babcock

2

1

Department of Business Administration,

University of Vienna, Vienna, Austria

2

Department of Social and Decision Sciences,

Carnegie Mellon University, Pittsburgh,

Pennsylvania, U.S.A.

Correspondence

Steffen Keck, Department of Business

Administration, University of Vienna, Oskar‐

Morgenstern Platz, Vienna 1090, Austria.

Email: steffen.keck@univie.ac.at

Summary

A large body of research demonstrates that women encounter severe penalties for violating gen-

der stereotypes. In this paper, we explore the conditions under which the reverse is true—when

being subject to a stereotype can actually benefit a woman compared to a man who is not subject

to the same stereotype. In particular, we suggest that in situations of causal ambiguity—

uncertainty about the reasons that a behavior occurred—differences in how men and women

are evaluated will be moderated by the extent to which observers engage in a low or a high level

of deliberative causal reasoning. In 3 experimental studies, participants were asked to make judg-

ments about an employee who violated a female gender stereotype by failing to provide help to a

coworker when asked to do so, but the reasons for this behavior were unclear. When participants

were prompted to engage in deliberative causal reasoning, women were evaluated more positively

than men, but not in the absence of such a prompt. Moreover, when participants did engage in

deliberative causal reasoning, the more positive evaluations of women compared to men were

driven by participants0beliefs that women0s behavior was due more to situational constraints than

the same behavior by men.

KEYWORDS

causal attributions, gender stereotypes,social accounts

1|INTRODUCTION

Gender stereotypes are pervasive, easily activated, and affect

individuals0judgments even outside of their conscious control

(Dunning & Sherman, 1997; Rudman & Glick, 2001). Prior research

has shown that women who violate female gender stereotypes in the

workplace encounter backlash in the form of more severe social and

economic penalties than men who exhibit the same behavior. For

example, female employees who behave in a dominant manner

(Bowles, Babcock, & Lai, 2007; Eagly, Makhijani, & Klonsky, 1992;

Lewis, 2000) or fail to carry out communal actions such as helping a

fellow coworker (Heilman & Chen, 2005) are judged more harshly than

male employees. In this paper, we examine conditions under which the

reverse is true—that is, when being subject to a stereotype can actually

benefit a woman who engages in nonstereotypical behavior, relative to

a man who is not subject to the stereotype.

In particular, we demonstrate that in situations of causal ambiguity

—when there is uncertainty about the reasons that a behavior occurred

—evaluations of noncommunal behavior by men and women will

strongly depend on the extent to which observers engage in either a

low or a high level of deliberative causal reasoning. We conducted

three experiments that manipulated depth of causal reasoning either

directly or indirectly by enhancing the salience of the causal ambiguity.

Our findings show that when observers engage in more deliberative

causal reasoning, women who behave in a noncommunal manner are

judged more positively than men who behave in the same manner. This

is because, when observers think more deeply about the reasons for

the noncommunal behavior, they make more situational attributions

for women than for men. In other words, women are more likely than

men to get the “benefit of the doubt.”In contrast, when observers

engage in a low level of deliberative causal reasoning, their attributions

are predominately dispositional, regardless of the actor0s gender. In

this case, our findings are more in line with those of previous research,

where women were judged more harshly than men.

Our findings have several important implications. As shown by a

large body of prior research, causal ambiguity and the need to make

causal attributions are pervasive in the workplace and have the poten-

tial to strongly shape how different types of workplace behavior are

Received: 5 March 2015 Revised: 26 June 2017 Accepted: 7 July 2017

DOI: 10.1002/job.2218

276 Copyright © 2017 John Wiley & Sons, Ltd. J Organ Behav. 2018;39:276–291.wileyonlinelibrary.com/journal/job

perceived (e.g., Burton, Taylor, & Barber, 2014; Kim, Dirks, Cooper, &

Ferrin, 2006; LePine & Van Dyne, 2001; Martinko & Gardner, 1987).

However, despite the considerable attention paid to causal attribu-

tions in the workplace on the one hand and backlash against violations

of gender stereotypes on the other, the role of causal reasoning depth

has not been integrated into studying these phenomena. Our study fills

this gap by exploring how causal reasoning depth impacts the evalua-

tion of counter‐stereotypical behavior in the presence of causal ambi-

guity. In doing so, our research makes an important contribution to the

literature on gender stereotypes by building a more comprehensive

account of the impact of stereotypes on observer evaluations and

outlining the moderating role of causal reasoning processes, which

can even lead to a reversal of previous findings. Moreover, our findings

add to previous work on the effect of attributions on perceptions of

workplace behavior in general, by establishing different types of causal

reasoning processes as an important moderating factor in this context.

As a practical matter, our findings are of importance for individuals

who might violate normative stereotypes, by helping them to under-

stand the conditions under which they may or may not be penalized

for such behavior. Moreover, our research has implications for how

managers implement evaluation systems in the workplace, as assess-

ments of employees will depend critically on the intersection of gender

stereotypes and causal reasoning processes.

2|ATTRIBUTIONS UNDER CAUSAL

AMBIGUITY AND TYPES OF CAUSAL

REASONING PROCESSES

In practice, individuals in organizations often cannot clearly attribute

the behavior of other people to one precise reason. Instead, they form

subjective beliefs about the likely causes of another person0s actions

(e.g., Heider, 1958; Kelley, 1967; Martinko, Douglas, & Harvey, 2006;

Weiner, 2008). In general, such causal ambiguity can occur when there

are potential internal as well as external explanations for a behavior or

when there is a general lack of information about likely explanations (e.

g., Ahn, Kalish, Medin, & Gelman, 1995). In these situations of causal

ambiguity, the subjective causal attributions that are made in organiza-

tions have a strong influence on how employees respond to important

issues, such as leader behavior (Martinko & Gardner, 1987), apologies

after acts of workplace deviance (Kim et al., 2006), low‐performing

peers (LePine & Van Dyne, 2001), abusive supervision (Bowling &

Michel, 2011; Breaux, Tepper, Carr, & Folger, 2010; Burton et al.,

2014), new human resource practices (Nishii, Lepak, & Schneider,

2008), or job termination (Barclay, Skarlicki, & Pugh, 2005).

Importantly, even when initial causal ambiguity is later resolved when

more information is available, observers0initial impressions are likely

to have a large influence on how that additional information is proc-

essed and will thus continue to strongly shape individuals0judgments

(e.g., Luchins, 1957; Ross, Lepper, & Hubbard, 1975).

Causal attributions have been classified along various dimensions,

such as locus of control, temporal stability, and general controllability

(e.g., Weiner, 1979). However, prior research (e.g., Heider, 1958;

Kelley & Michela, 1980; Ross & Fletcher, 1985; Trope, 1986) suggests

that the most central dimension of causal attributions is the distinction

between attributions to internal, personal dispositions, such as

personality traits or attitudes, and attributions to external, situational

causes, such as task characteristics or social pressure. Importantly, pre-

vious work on the correspondence bias (for an overview, see, e.g., Gil-

bert & Malone, 1995) has found that observers often fail to consider a

broad range of possible explanations for an observed behavior and

instead make predominantly dispositional attributions. This correspon-

dence bias towards dispositional attributions is frequently driven by a

lack of awareness of possible external explanations and by the ten-

dency of observers to initially draw inferences about the traits of an

observed person and to then (insufficiently) adjust for the possibility

of external explanations (e.g., Gilbert & Malone, 1995).

However, the correspondence bias is strongly mitigated, and

individuals do consider a broader range of possible situational and

dispositional explanations when they engage in a high level of delib-

erative causal reasoning (Blumberg & Silvera, 1998; Gilbert, Pelham,

& Krull, 1988; Gilbert, Krull, & Pelham, 1988). For example, in their

seminal paper on causal reasoning, Gilbert, Pelham, et al. (1988)

argue that once observers fully identify an action taken by a target

person, they go through two steps when making causal attributions.

First, they categorize—that is, they draw dispositional inferences

about the actor—and then they correct these inferences by consider-

ing information about possible situational constraints. Importantly,

Gilbert et al. also suggest that “Characterization (making dispositional

attributions) is generally a relatively automatic process that requires

little effort or conscious attention, whereas correction (taking into

account possible situational factors) is a more deliberate, relatively

controlled process that uses a significant portion of the perceiver0s

processing resources”(p. 734). Consistently, the empirical findings

by Gilbert, Pelham, et al. (1988) showed that observers who

employed only a low level of deliberative causal reasoning made

mostly dispositional attributions. However, when observers invested

sufficient cognitive resources to think more deliberatively about an

observed action, they considered information about possible situa-

tional constraints, and their attributions were both dispositional

and situational, depending on the characteristics of the observed

behavior. This distinction between attributions that are based on

either a high or a low level of deliberative causal reasoning is also

consistent with research on dual process theory, which suggests that

individuals in general make judgments using two different reasoning

types: (a) a default process that is fast, automatic, and effortless or

(b) a process that is slow, deliberative, and effortful and which is

used only when individuals invest sufficient cognitive recourses (for

an overview of dual process theory, see, e.g., Evans & Frankish,

2009). Importantly, these two processes are usually thought of not

as fully mutually exclusive but as a continuum with purely automatic

judgments on one side and purely deliberative judgments on the

other (e.g., Evans & Frankish, 2009).

One important factor that is likely to affect an observer0s level of

deliberative causal reasoning is the salience of causal ambiguity in a

particular situation (Tobin & Weary, 2008; Weary & Jacobson, 1997;

Weary, Vaughn, Stewart, & Edwards, 2006). In particular, if causal

ambiguity is very salient, observers are likely to think more delibera-

tively and thus more carefully about possible internal or external expla-

nations. In contrast, when the salience of causal ambiguity is low,

observers are likely to make mostly dispositional attributions as

KECK AND BABCOCK 277