Sage Journals: Discover world-class research

Abstract

In an era of intense competition for talent, turnover among organizational newcomers—particularly from underrepresented groups—remains a persistent challenge. While organizational allyship programs are known to reduce turnover, their frequent association with diversity, equity, and inclusion (DEI) efforts has led to implementation resistance. This study addresses gaps in the research regarding allyship by testing communication-based strategies for promoting participation in allyship and examining factors contributing to freedom threat or participation willingness. Participants (N = 312) were randomly exposed to one of four experimental conditions: a control message, a narrative advocacy message, a directive advocacy message, or a combination of both. Results showed that both the narrative message and the combined message positively influenced attitudes toward allyship through transportation. Additionally, a second-stage moderation analysis showed that participants’ political identification influenced the impact of narrative transportation on allyship attitudes, with stronger effects for more conservative participants. Greater levels of psychological safety were also linked to more positive relationships between narrative messages and attitudes toward allyship. Overall, the study provides insights into the potential use of narrative persuasion as a tool to support allyship advocacy as well as a range of other organizational initiatives that might need to overcome resistance due to perceptions of freedom threat.

Keywords

allyship DEI freedom threat psychological reactance narrative persuasion transportation resistance recruitment retention turnover organizational communication

Employee turnover is a concerning issue in many organizations, particularly in an era when unemployment percentages are low and competition for talent is high, with the greatest rates of employee turnover to be found among newly hired employees, particularly those from under-represented minorities (Allen et al., 2010; Hofhuis et al., 2014; Liu et al., 2024). At a time in which an organization’s ability to recruit, retain and deploy a suitably talented, motivated and diverse workforce has become of mission-critical competitive importance, organizational allyship programs have been identified as a potentially effective means to reduce turnover and support talent-retention efforts (Brown & Ostrov, 2013; Ferdman, 2014; Miller & Katz, 2018). However, attempts to implement such programs have met with mixed success, often owing to resistance from organizational members (Dobbin & Kalev, 2016, 2018; Kawasaki & Zou, 2023).

Prior studies have examined what an allyship program should look like (Kiehne & Hafen, 2022; Ostrove et al., 2019) and how allies should behave (Janssen, 2024; Perkins et al., 2024), but few studies have addressed why organizational members would want to become allies, and why they might not. The current study examines allyship advocacy strategies from a resistance-reduction perspective. Beyond allyship, it offers insights into the underlying organizational communication factors that can promote and reduce resistance to a broad range of organizational initiatives.

As costly as it can be to recruit and train new hires, it can be even more costly to replace them. According to a recent major business publication report, onboarding costs average about $1,400 per new employee and it takes about 6 months for an organization to break even on a new hire (Snyder & Bottorff, 2023). Compounding the problem, about 30% of new hires leave their new jobs within 90 days, and cost estimates for replacing them run from about 33% (Snyder & Bottorff, 2023) to as much as 200% (Liu et al., 2024) of their annual salaries.

Newcomer turnover is a confounding problem in organizations (Liu et al., 2024), especially in regard to members of under-represented populations (Hofhuis et al., 2014; Miller & Gans, 2002; Shenoy, 2023). Even finding new employees from minority populations can be challenging. Research by a leading recruitment organization revealed that 50% of surveyed employers found it difficult to expand their workforce diversity through their recruitment processes, with 43% noting a lack of applicants of color and 42% noting a lack of female applicants (Pollack, 2023). Retention is also a challenge. Once hired, newcomers from minority populations tend to encounter more reasons to leave than their majority-population counterparts. Research has shown that, compared to their majority counterparts, minority employees experience more negative social interactions and fewer opportunities for career advancement, factors which have been shown to be predictive of turnover intentions and decisions (Allen et al., 2010; Hofhuis et al., 2014).

Addressing Gaps in the Study of Allyship

Often promoted as part of organizational diversity, equity and inclusion (DEI) programs, allyship has been widely portrayed as an effective strategy for supporting under-represented members of organizations and thereby reducing turnover (Ferdman, 2014; Miller & Katz, 2018) but also regarded with mixed feelings by some of those ally-supported members (Collier-Spruel & Ryan, 2024). Described in terms of the actions of members of dominant or majority groups with power and privilege in support or defense of members of disadvantaged or minority groups (Ostrove & Brown, 2018), allyship can manifest as support for individuals or groups who are marginalized or treated differently because of their race, gender, ethnicity, age, religion, sexual orientation, disability status or other aspect of identity (Ferdman, 2014; Reeves et al., 2024). Efforts to promote allyship at the societal level generally focus on social justice issues (Brown & Ostrove, 2013; Kiehne & Hafen, 2022; Ostrove & Brown, 2018). In organizations, allyship is often framed as beneficial for marginalized groups by promoting inclusion and retention (Ferdman, 2014; Graham et al., 2023; Reeves et al., 2024) and occasionally as advantageous for the organization itself (Allen et al., 2010). Research on allyship often centers on effective ally characteristics and best practices (Collier-Spruel & Ryan, 2024; Kiehne & Hafen, 2022; Janssen, 2024; Ostrove et al., 2019; Reeves et al., 2024). However, few studies explore how to persuade members of privileged groups to act as allies for minority newcomers or the importance of doing so (Perkins et al., 2024; Reeves et al., 2024). When members of privileged groups act as allies for newcomers, they not only provide support for the newcomers but also provide role models for other members of privileged groups as exemplars of desired workplace behaviors (Ferdman, 2014; Miller & Katz, 2018; Reeves et al., 2024). Despite the potential benefits offered by allyship programs and other efforts to build inclusion in organizational cultures, these efforts frequently encounter resistance (Dobbin & Kalev, 2016, 2018; Kawasaki & Zou, 2023) and yield mixed results (Collier-Spruel & Ryan, 2024).

The current study positions allyship as a strategic tool essential for fostering employee engagement and reducing turnover across diverse groups, particularly among underrepresented newcomers, but also prone to resistance. When effectively framed and communicated, allyship programs have the power to reshape workplace inclusivity and retention efforts, moving beyond traditional DEI approaches to serve as a foundation for broader organization-wide engagement strategies that contribute to more positive workplace cultures (Ferdman, 2014; Miller & Katz, 2018; Reeves et al., 2024).

Specifically, this study adopts a communication perspective and contributes to the organizational communication literature by examining the effects of persuasive messaging strategies designed to minimize resistance while promoting allyship. The study also explores the underlying factors that can lead to resistance or promote participation willingness in organizational allyship initiatives. Considering the importance and the discouraging rate of success of many organizational DEI initiatives (Al-Gharbi, 2020; Dobbin & Kalev, 2016; Gubler et al., 2022) as well as change efforts more broadly (Beer et al., 1990; Hughes, 2011; Knight et al., 2017), organizational leaders, change agents and scholars could benefit from a better understanding of the factors that contribute to the success or failure of their initiatives and interventions. This study aims to demonstrate that with the right communication approach, allyship can be reimagined as a core strategy for building inclusive, resilient organizational cultures that support organizational success as well as long-term employee engagement and retention.

Anticipating Reactance and Resistance

The challenges of designing successful organizational interventions can be greater than many organizational change agents appreciate. Failure rates of organizational change interventions are often estimated at 70% and higher (Bashein & Markus, 1994; Beer et al., 1990; Hughes, 2011; Kotter, 1995). Bashein and Markus (1994) suggest the reasons for failure often appear to be similar to the reasons the interventions were initiated in the first place, for example, problematic management styles, poor planning, competing priorities, fear, lack of optimism, and hostilities among colleagues. Beer et al. (1990) attribute much of this failure rate to programmatic approaches that attempt to “push change down through the organization” (p. 158), the likes of which other researchers have suggested can lead to resistance in the form of psychological reactance (Brehm, 1989; Burgoon et al., 2002; Miller et al., 2006).

Threat to Personal Freedom as a Cause of Resistance

The theory of psychological reactance (Brehm, 1989) suggests that explicit behavioral directives from figures of authority are often construed as threats to an individual’s sense of freedom and personal autonomy, which can lead to reactions such as “ignoring the persuasive attempt, derogating the source, and even producing even more of the undesired behaviors as a means of demonstrating choice or restoring attitudinal freedom” (Burgoon et al., 2002, p. 215). Scholars have established that strongly worded behavioral advocacy messages consistently arouse reactance (Rosenberg & Siegel, 2018), with messages carrying a high degree of authoritative threat having been shown to evoke greater levels of reactance than low-threat messages (Bensley & Wu, 1991).

Miller et al. (2007) suggest the level of authority of the source of the behavioral directive can influence the strength of the reaction such that “the more directive and controlling a persuasive message is perceived to be, the more likely its position is to be rejected” (p. 223). Whether or not they include allyship components, DEI initiatives are often enacted as “top-down” programs with problematic levels of resistance accruing to psychological reactance as a result (Dobbin & Kalev, 2016, 2018, 2021; Lipman, 2018). Therefore, organizational directives intended to persuade employees to engage in newcomer allyship behaviors, particularly those involving underrepresented minority populations, may encounter resistance.

This suggests the following hypothesis:

H1: For promoting newcomer allyship, a directive allyship advocacy message will result in greater levels of psychological reactance compared to a non-directive message.

A Narrative Approach to Reducing Resistance

In recent years, narrative persuasion has been proposed by increasing numbers of scholars and organizational change agents as a strategy to avoid or overcome resistance (Hall et al., 2023; Hamby et al., 2017; Lane, 2023; Ratcliff & Sun, 2020). Internal organizational communication strategies incorporating narrative persuasion and storytelling are gaining increasing traction in organizations (Beigi et al., 2019; Lane, 2023; Rhodes & Brown, 2005).

The effectiveness of narrative persuasion strategies is attributed by many scholars to the mechanism of transportation (Green & Brock, 2000; Hamby et al., 2017; van Laer et al., 2014), in which the target audience becomes immersed in the narrative story and can vicariously experience attitudes and behaviors they might argue against or never consider if they were not in a transported state (Moyer-Gusé, 2008; Moyer-Gusé & Nabi, 2010). Factors that influence the level of transportation experienced by the target audience include the salience of the situation portrayed in the story and the levels of sympathy for and identification with the characters portrayed in the story (Green & Brock, 2000, 2002; Slater et al., 2006).

Organizational storytelling puts the persuasive powers of narrative transportation to work in highlighting, reinforcing or changing organizational norms, values and cultures (Boyce, 1996; Brady & Haley, 2013). When advocated-for behaviors are modeled by story characters, persuasive intent is less likely to be perceived and the advocacy effort is less likely to provoke freedom threat or counter-arguing from message recipients (Moyer-Gusé, 2008; Moyer-Gusé & Nabi, 2010). In experimental settings, dissemination of organizational stories in which characters similar to the intended audience experience positive results from engaging in behaviors aligned with organizational initiatives has been shown to produce positive results (Gans & Zhan, 2023a, 2023b). In one study of organizational citizenship behaviors, mediation effects of transportation relating to an organizational storytelling message produced significantly greater supportive voice intentions compared to control and directive message conditions (Gans & Zhan, 2023a). In another study, mediation effects of freedom threat relating to DEI-training advocacy messaging were associated with significant differences in attitude toward DEI training (Gans & Zhan, 2023b).

Based on the foregoing and prior research in these areas, we suggest the following set of hypotheses:

H2: A narrative-based allyship advocacy message will result in greater levels of transportation compared to a control message.

H3: The influence of message strategy on attitude toward allyship will be mediated by transportation, such that a narrative-based message will induce greater levels of transportation compared to other messages, which in turn will lead to greater positive attitude change.

H4: The influence of message strategy on attitude toward allyship will be mediated by reactance, such that a narrative-based message will induce lower levels of reactance compared to other messages, which in turn will lead to greater positive attitude change.

The Influence of Political Identification

In recent years, DEI initiatives have been often accused of being “woke” programs that are supportive of everyone except white men (Lipman, 2018). Consequently, attitudes toward anything related to DEI—such as allyship—tend to be skewed by politics, with people who identify as more conservative harboring less-positive attitudes toward DEI-related behaviors and likely to express greater resistance to promotion of such behaviors (Gans & Zhan, 2023b).

Identification with conservative or liberal political ideologies has been shown to produce differences in attitudes on a range of different issue topics that seem to align with the egalitarian/communitarian perspectives often associated with liberals and the hierarchical/individualist perspectives associated with conservatives (Kahan et al., 2011), such that people who self-identify as conservatives seem much less favorable toward policies that provide social supports for people dealing with obesity (Niederdeppe et al., 2014), or that might promote vaccination against the human papillomavirus (Gans, 2014; Kahan et al., 2010) or that might advocate for action to address global warming or limit access to handguns (Kahan et al., 2011). As noted above, similarity with characters and values depicted in narratives tends to strengthen the transportative influence of the narratives (Moyer-Gusé, 2008; Moyer-Gusé & Nabi, 2010).

Based on the foregoing, we propose the following hypotheses regarding the influence of political identity on allyship advocacy:

H5: Political identity will moderate the mediation effects of transportation on the influence of the message conditions on attitude change, such that the more conservative one is, the more positive the relationship between transportation and attitude toward allyship.

H6: Political identity will moderate the mediation effects of reactance on the influence of the message conditions on attitude change, such that the more conservative one is, the more negative the relationship between reactance and attitude toward allyship.

The Influence of Perceived Organizational Climate: Psychological Safety

Organizational climate refers to employees’ shared perceptions and interpretations of workplace policies, practices, procedures, and the observed behaviors that are encouraged, rewarded, or expected (Schneider et al., 2013). This collective understanding shapes how individuals interpret their organizational environment and influences their willingness to engage with organizational initiatives, including allyship initiative programs. Organizational climate serves as an important contingency for narrative persuasion at the organizational level, particularly in fostering allyship, because it establishes the norms and expectations that influence whether employees feel empowered to take actions perceived as risky, such as “standing up for” or “speaking up for” disadvantaged colleagues (Detert & Edmondson, 2011).

Speaking up in defense of others, even when aligned with organizational values, is often perceived as risky, especially when it involves pointing out inequities (Detert & Edmondson, 2011; Edmondson & Besieux, 2021; Gans & Zhan, 2023a). For allyship behaviors to become normative, employees must not only see clear models of such behavior but also trust that their efforts will lead to positive outcomes (Gans & Zhan, 2023a). This highlights the crucial role of psychological safety—a shared belief among employees that it is safe to take interpersonal risks in the workplace (Edmondson, 1999). In organizations whose employees experience low levels of psychological safety, individuals may hesitate to engage in allyship for fear of repercussions or awkwardness in adopting new behaviors. In other words, those employees are reluctant to take interpersonal risks to advocate for their colleagues or engage in allyship behaviors because they fear potential negative consequences for themselves (Perkins et al., 2024).

Psychological safety creates an environment where employees feel comfortable navigating the learning curve associated with adopting allyship behaviors. Mistakes and initial discomfort are inevitable when learning any new behavior, but a psychologically safe climate allows employees to take these risks without fear of judgment or retaliation (Edmondson, 1999; Miller & Katz, 2018). Thus, fostering psychological safety within an organization is not only a key element of a supportive climate but also a prerequisite for empowering employees to act as allies. In this context, psychological safety becomes an essential organizational climate factor that enables narrative persuasion strategies to succeed by encouraging engagement with allyship initiatives.

In sum, psychological safety, an organizational-level climate factor, may influence the effectiveness of communication strategies. Therefore, we propose this final hypothesis, focusing on the influence of psychological safety as an organizational-level climate factor:

H7: Psychological safety will moderate the influence of the message conditions on attitude toward allyship, wherein greater levels of psychological safety will amplify the persuasive effects of the messages.

As outlined above, the study is designed to examine the effects of a range of allyship advocacy message strategies on attitude toward allyship as well as the mediating and moderating effects of freedom threat, narrative transportation, political identity and psychological safety. These relationships are presented in Figure 1. The associated hypotheses are presented in Table 1.

Figure 1.

Experimental model: Effects of mediating and moderating factors on allyship attitude.

Table 1.

Hypotheses, as Outlined in Figure 1.

Effects of directive vs. narrative strategies
H1	A directive allyship advocacy message will produce greater levels of freedom threat than a non-directive message
H2	A narrative-based allyship advocacy message will produce greater levels of transportation than a control message
H3	The effects of message strategy on attitude toward allyship will be mediated by transportation, such that a narrative-based message will induce greater levels of transportation compared to other messages, which in turn will lead to greater positive attitude change
H4	The effects of message strategy on attitude toward allyship will be mediated by freedom threat, such that a narrative-based message will induce lower levels of freedom threat compared to other messages, which in turn will lead to greater positive attitude change
Effects of political identity
H5	Political identity will moderate the mediating effects of transportation on the influence of message strategy on allyship attitude such that the more conservative one is, the more positive the relationship between transportation and allyship attitude.
H6	Political identity will moderate the mediating effects of freedom threat on the influence of message strategy on allyship attitude such that the more conservative one is, the more negative the relationship between freedom threat and allyship attitude.
Effects of psychological safety
H7	Psychological safety will moderate the influence of the message conditions on attitude toward allyship, wherein greater levels of psychological safety will amplify the persuasive effects of the messages.

Method

Sample

Participants were recruited from Prolific.com, which is a widely used research platform that facilitates data collection. Prolific upholds stringent standards to ensure data integrity. For example, their automated algorithm, called Protocol, incorporates more than 35 different checks to detect and remove bad actors, bots, and noticeably low-quality participants, and measures participant contributions over time to make sure they continue to meet quality standards (Prolific, 2024). With those preventative measures in place, Prolific has demonstrated superior data quality over Amazon’s MTurk in various dimensions (Stanton et al., 2022).

The questionnaires and experimental stimuli for the current study were hosted on the QuestionPro platform. Acknowledging the potential challenges associated with online participant pools, we adhered to established best practices recommended for communication researchers during the design and execution of online experiments (Sheehan, 2018). Participants spent an average of 9 min 35 s completing the survey and received $1.75 as compensation, resulting in an average hourly wage-equivalent of $11.96, which is greater than the minimum compensation rate of $8.00 per hour required by Prolific.

Following the removal of incomplete responses and those that failed attention checks, a total of 312 valid responses remained (N = 312). Among these, 57.7% identified as female, 40.4% as male, 1.6% as non-binary, and 0.3% chose not to disclose their gender. Participants’ ages ranged from 20 to 81 years, with a mean age of 39.76 (SD = 12.91). The majority of participants, constituting 67.9%, had completed college and 21.5% had some college, while 56.6% reported an annual income of at least $60,000. In terms of employment status, a majority of the participants had full-time or part-time jobs, with 79.5 employed full-time and 15.4% employed part-time. Regarding race and ethnicity, 60.9% identified as White, 10.3% as Asian or Pacific Islander, 17.6% as Black, 6.4% as Latino/a or Hispanic, 3.8% as multiracial, and 0.3% as Middle Eastern.

Study Procedures

Upon joining the study through Prolific, participants were furnished with a hyperlink leading them to the online survey hosted on QuestionPro. They were directed to an informed consent statement and required to provide consent before proceeding further. The study, survey instrument, and informed consent components were approved for human subjects research by a research university’s Office of Regulatory Services (IRB Protocol Number 2024-0147.1) on February 9, 2024.

The research adopted a between-subjects experimental design. Participants were prompted to envision themselves as receiving specific content in a work email from their current employer. They were then randomly allocated to one of four message conditions: a control condition, a narrative condition, a directive condition, or a narrative-plus-directive condition. Following this, participants were assessed for their attitudes toward newcomer allyship, alongside evaluations of other variables including freedom threat, transportation, political identity as well as standard demographics.

Control Condition

In the control condition, participants encountered a message regarding how Americans feel about their work. While the message was intentionally crafted to mirror the style and length of messages in other conditions, its content carried no references to newcomer allyship and no behavioral recommendations.

Narrative Condition

For participants assigned to the narrative condition, the message took the form of a personal narrative shared by a long-time member of the organization. This narrative portrayed the individual’s journey starting as a newcomer and grappling with challenges alongside others. The individual recounts being fortunate, as a newcomer, to have received assistance from a veteran within the company. Subsequently, inspired by this act of kindness, the individual transitioned into embracing the role of mentor and extending allyship support to newcomers because helping them succeed “is how we help us all succeed.” The narrative was accompanied by a stock photo depicting a blue-collar worker.

Directive Condition

For participants assigned to the directive condition, the message took the form of a representation of a break-room wall poster that encouraged employees to be allies for newcomers. It had specific suggestions such as “speak up for newcomers,” “defend newcomers,” “empower newcomers,” and “be an ally.” Notably, the messaging emphasized the benefits of such behaviors to those who acted as allies and to the organization as a whole, not just the newcomers. The text was accompanied by a stock photo depicting workers working on their laptops to ensure similarity with the narrative message format. (See Appendix for Control, Narrative, and Directive message conditions.)

Narrative plus Directive Condition

In this condition, participants were exposed to both experimental conditions, seeing the narrative message first and then the directive message.

Measures

We evaluated all attitude variables by asking participants to indicate their level of agreement with various statements using 7-point Likert-type scales, ranging from 1 (Strongly Disagree) to 7 (Strongly Agree).

Attitude toward Allyship

After message exposure, attitude toward allyship was measured using a 9-item measure adapted from Collier-Spruel and Ryan’s (2024) allyship measure. Sample items included “it’s important to advocate for minority newcomers when they are treated worse than others” and “it’s important to advocate for minority newcomers to receive opportunities.” The Cronbach’s alpha reliability score for this measure was .94. The McDonald’s omega was .95.

Transportation

Participants’ levels of narrative transportation were assessed using Slater et al.’s (2006) 5-item measure. Sample items included statements like “I could vividly imagine myself in the scene,” “I felt curious about how events would unfold,” and “I was fully engaged mentally in the narrative.” The Cronbach’s alpha reliability score for this measure was .82. The McDonald’s omega was .83.

Freedom Threat

Drawing on Moyer-Gusé and Nabi’s (2010) study, we utilized Dillard and Shen’s (2005) 4-item measure of autonomy threat to assess of this precursor to reactance. Sample items included statements like “the content’s underlying message felt like a threat to my freedom of choice” and “the content’s underlying message seemed manipulative.” The Cronbach’s alpha for this measure was .86, as was the McDonald’s omega.

Political Identity

Political identity was assessed using a single question item: “Conservative or Liberal? Which best describes you politically?” Participants indicated their position on a scale ranging from 1 (very strongly conservative) to 9 (very strongly liberal).

Psychological Safety

We measured psychological safety by using Edmondson’s (1999) measure. Sample items included “members of this team are able to bring up problems and tough issues” and “it is safe to take a risk on this team.” The Cronbach’s alpha for this measure was .85, as was the McDonald’s omega.

Manipulation Checks

As per O’Keefe (2003), it is appropriate to examine whether different message conditions result in variations in intended psychological states. In our study, the two message conditions of interest are a non-directive narrative-based advocacy message, which was intended to stimulate transportation, and a directive advocacy message, which was intended to stimulate freedom threat. These expectations were proposed as Hypotheses 1 and 2, and results of the hypothesis tests are detailed in the Results section.

Measurement Validity

Before conducting the analysis, we assessed the skewness and kurtosis statistics of all measures, finding them to be within acceptable limits of univariate normality (Hancock et al., 2010), with absolute values not exceeding 2. Consequently, the original data were utilized without transformation. Significance level was set at p < .05, but exact p values are reported where appropriate.

To ensure construct validity, a confirmatory factor analysis (CFA) was performed using Mplus 8.0. Attitude toward allyship, along with freedom threat, transportation, psychological safety were included in the model. During CFA, error covariance paths were added for items sharing similar wordings to address potential method effects. Results from the CFA indicated robust and statistically significant factor loadings for all constructs. While the Chi-square test yielded significance (χ2 = 539.78, df = 266, p < .001), goodness-of-fit indices (RMSEA = 0.057, CFI = 0.94, SRMR = 0.058) indicated a satisfactory fit between the data and the model (Hu & Bentler, 1999).

Analytical Strategies

In this study, we utilized SPSS 29.0 to conduct the necessary analyses for testing the proposed hypotheses. Mediation and moderation analyses were carried out using the well-established PROCESS 4.2 macro in SPSS, developed by Hayes (2022). Descriptive statistics for the measured variables, including means, standard deviations, and bivariate correlations were computed and are presented in Table 2.

Table 2.

Means, Standard Deviations, and Bivariate Correlations.

Variable	Mean	SD	1	2	3	4
1 Transportation	26.57	4.87
2 Freedom threat	10.20	5.33	−.16**
3 Psychological safety	34.4	7.63	.05	−.17**
4 Attitude toward allyship	56.27	7.15	.47**	−.38**	.13
5 Political ideology	6.03	2.17	.07	−.29**	.02	.36**

p < .01.

Results

Hypothesis 1: Directive Messages and Reactance

Hypothesis 1 predicted that directive-based messages would lead to greater levels of freedom threat compared to narrative-based messages in promoting newcomer allyship. We conducted an independent samples t-test to test the hypothesis. The results indicated a significant difference in participants’ levels of freedom threat between those exposed to the narrative message, M = 8.58, SD = 5.03, and those exposed to directive messages, M = 10.85, SD = 5.51; t (151) = −2.63, p = .005 (one-sided), Cohen’s d = −.427, thereby supporting Hypothesis 1.

Hypothesis 2: Narrative Messages and Transportation

H2 stated that narrative-based messages will result in greater levels of transportation compared to a control message in promoting newcomer allyship. We conducted an independent samples t-test to test the hypothesis. The results indicated a significant difference in participants’ levels of transportation between those exposed to the narrative message, M = 27.29, SD = 5.24, and those exposed to control messages, M = 25.43, SD = 4.93; t (141) = 1.86, p = .015 (one-sided), Cohen’s d = .366, thereby supporting Hypothesis 2.

Hypothesis 3: Transportation as a Mediator

H3 stated that the relationship between message strategy and attitude toward allyship is mediated by transportation, such that narrative-based messages will induce greater levels of transportation compared to other messages, which in turn will lead to greater attitude change. The mediation analysis examined the relationships among message conditions (IV), transportation (M), and attitude regarding newcomer allyship (DV), using the PROCESS macro (i.e., Model 4; see Hayes, 2022). Message conditions were dummy coded, where the control condition is set up as the referent, X1 is the narrative message condition, X2 is the directive message condition, and X3 is the narrative-plus-directive message condition. A summary of results of the mediation analysis is presented in Table 3.

Table 3.

Mediation Analysis Result: Transportation as Mediator.

	Transportation (SE)	Attitude (SE)	Indirect effect [95% CI]
Constant	25.43*** (0.56)	36.29 (1.98)***
Narrative	1.86* (0.81)	2.81 (1.04)***	1.21* [0.09, 2.37]
Directive	1.03 (0.77)	3.60 (0.99)***	0.67 [−0.28, 1.78]
Narrative + Directive	1.66* (0.77)	3.82 (0.99)***	1.08* [0.09, 2.30]
Transportation		0.65 (0.07)***
R ²		0.27***

Note. Unstandardized coefficients. DV = Attitude toward Newcomer Allyship; Referent: Control Message.

p < .05. **p < .01. ***p < .001.

Initially, we assessed the direct effect of message conditions on attitude change. The results indicated a significant direct effect of message conditions on attitude change (b_narrative = 2.81, p < .01; b_directive = 3.60, p < .001; b_combined = 3.82, p < .001), suggesting that compared to the control condition, the narrative, directive, and narrative-plus-directive message conditions all had a significantly greater impact on attitude toward newcomer allyship. Next, we investigated the indirect effect of message conditions on attitude toward newcomer allyship through transportation. The results revealed significant indirect effects of the narrative message condition (b_narrative = 1.21, p < .05) and narrative-plus-directive message condition (b_combined = 1.09, p < .05) on attitude through transportation, but not the directive message condition (b_directive = 0.67, p > .05). This suggests that transportation partially mediated the relationship between the narrative and narrative-plus-directive message conditions and attitude toward newcomer allyship. Additionally, we examined the direct effect of transportation on attitude toward newcomer allyship. The results showed a significant direct effect of transportation on attitude change (b = 0.65, p < .001), indicating that greater levels of transportation were associated with more positive attitude regarding newcomer allyship. Therefore, Hypothesis 3 is supported.

Hypothesis 4: Freedom Threat as a Mediator

H4 stated that the relationship between message strategy and attitude toward allyship is mediated by freedom threat, such that narrative-based messages will induce lower levels of freedom threat compared to other messages, which in turn will lead to more positive attitudes toward allyship. This mediation analysis examined the relationships among the message conditions (IV), freedom threat (M), and attitude regarding newcomer allyship (DV). This time the message conditions were dummy coded with the directive condition as the referent, X1 as the narrative message condition, X2 as the control message condition, and X3 as the narrative-plus-directive message condition. The mediation analysis was done using the PROCESS macro (i.e., Model 4; see Hayes, 2022). Results of this mediation analysis are presented in Table 4.

Table 4.

Mediation Analysis Result: Freedom Threat as Mediator.

	Reactance (SE)	Attitude (SE)	Indirect effect [95% CI]
Constant	10.84*** (0.58)	62.89*** (1.02)
Narrative	−2.27** (0.86)	−1.45 (1.04)	1.19* [0.31, 2.12]
Control	−0.71 (0.84)	−4.65*** (1.01)	0.38 [−0.50, 1.27]
Narrative + Directive	0.07 (0.81)	0.67 (0.98)	−0.04 [−0.94, 0.86]
Freedom threat
R ²	0.03*	0.22***

Note. Unstandardized coefficients. DV = Attitude toward Newcomer Allyship; Referent: Directive Message.

p < .05. **p < .01. ***p < .001

The relative direct effect of message conditions on attitude was found to be significant only for the difference between the directive (referent) message and control (X2) message (b_narrative = −1.45, p > .05, b_control = −4.65, p < .01, b_combined = 0.67, p > .05), indicating that after controlling for the mediating effect of freedom threat, participants exposed to the directive message exhibited significantly more positive levels of attitude toward newcomer allyship compared to the control condition. The relative indirect effect for the difference between the directive and narrative (X1) message conditions on attitude was significant (b_narrative = 1.20, p < 0.05) but the indirect effects of the differences between the other conditions were not (b_control = 0.38, p > .05, b_combined = −0.04, p > .05).

To evaluate the sensitivity of the analyses for the non-significant indirect effects, a post-hoc power analysis was conducted. For the indirect effects of control versus directive condition, the power to detect this effect was 0.997, indicating that the non-significant result is likely reflective of a true null effect. For the combined versus directive condition, the power was 0.079, indicating that the non-significant result for this condition may be attributable to insufficient sensitivity.

In summary, the results revealed a significant indirect effect of the message conditions on attitude toward allyship through freedom threat for the difference between narrative message and directive message only (b = 1.20, 95% CI [0.46, 0.31], p < .05), indicating mediation. However, the indirect effects were not significant for the difference between directive message and control and narrative-plus-directive conditions. These findings suggest that the mediation process operates selectively. Thus, Hypothesis 4 is partially supported.

Hypothesis 5: Political Identity in the Moderated Mediation Model for Transportation

H5 stated that political identity will moderate the mediating effects of transportation on the influence of the message conditions on attitude toward allyship, such that the more conservative one is, the more positive the relationship between transportation and attitude toward allyship. We conducted a second-stage moderated mediation analysis, as outlined by Hayes' PROCESS macro model 14, to examine the moderated mediation model in which message condition (IV) indirectly influences attitude toward allyship (DV) through transportation (M), moderated by political identity. Results of this analysis of moderated mediation are presented in Table 5.

Table 5.

Moderated Mediation Analysis: Effects of Message on Attitude Mediated by Transportation, Moderated by Political Identity.

	Transportation (SE)	Attitude toward allyship (SE)	Moderated mediationM*W
Constant	25.43** (.56)	6.32 (4.81)
Message Conditions
Narrative (vs. Control)	1.95* (0.81)	1.97* (0.94)
Directive (vs. Control)	1.03 (0.77)	2.51** (0.89)
Directive + Narrative (vs. Control)	1.66* (0.77)	2.69** (0.89)
Transportation (M)		1.57** (0.18)
Political identity (W)		5.36** (0.78)
Transportation × PID (M*W)			−0.16** (0.03)
R ²	.02	.42**	.06**

Note. PID = Political Identity; M = Mediator (Transportation); W = Moderator (Political Identity); Unstandardized coefficients are reported. The control condition serves as the referent group.

p < .05. **p < .01. ***p < .001.

The results revealed significant pathways in this model. First, we found evidence of the mediation effect of transportation, as found in hypothesis 3. Furthermore, the analysis revealed a significant second-stage moderation effect, indicating that the strength of the relationship between transportation and attitude toward allyship varied depending on the level of political identification (b = −0.16, p < .001) but there was no moderation effect on the relationship between message conditions and attitude toward allyship (see Figure 2). Therefore, Hypothesis 5 is supported.

Figure 2.

Second-order moderated mediation model of transportation and political identity.

Hypothesis 6: Political Identity in the Moderated Mediation Model for Freedom Threat

H6 stated that political identity will moderate the mediating effects of freedom threat on the influence of the message conditions on attitude toward allyship, such that the more conservative one is, the more negative the relationship between freedom threat and attitude toward allyship. We utilized a second-stage moderated mediation analysis, following the framework outlined in Hayes’ PROCESS macro Model 14, to investigate these intricate relationships. The impact of message condition (IV) on attitude toward allyship (DV) through freedom threat (M) was explored, with political identity serving as the moderator. The results of this analysis are summarized in Table 6.

Table 6.

Moderated Mediation Analysis: Effects of Message on Attitude Mediated by Freedom Threat, Moderated by Political Identity.

	Reactance (SE)	Attitude toward allyship (SE)	Moderated mediationM*W
Constant	10.13*** (0.61)	56.46*** (2.38)
Message conditions
Narrative (vs. Control)	−1.52 (0.89)	2.89** (1.05)
Directive (vs. Control)	0.71 (0.84)	4.13*** (0.99)
Directive + Narrative (vs. Control)	0.78 (0.84)	4.78*** (0.99)
Freedom threat (M)		−0.73*** (0.18)
Political identity (W)		.021 (0.36)
Freedom threat × PID (M*W)			0.05 (0.02)
R ²	.03*	.28***	0.01

Note. PID = Political Identity; M = Mediator (Reactance); W = Moderator (Political Identity); Unstandardized coefficients are reported. The control condition serves as the referent group.

p < .05. **p < .01. ***p < .001.

Our finding regarding mediation is as discussed in the results section of Hypothesis 4. However, the analysis did not find a notable second-stage moderation effect, suggesting that the association between freedom threat and attitude toward allyship did not vary contingent upon one’s level of political identification (b = .05, p = .07). Therefore, Hypothesis 6 is not supported. To assess the sensitivity of the analyses for the moderation effects, a post-hoc power analysis (Cohen, 1988) was conducted for the non-significant moderation effect of political identification on the relationship between freedom threat and attitude. The power was 0.096, indicating that the non-significant result is more likely due to insufficient sensitivity to detect small effects rather than the absence of a true effect.

Hypothesis 7: Psychological Safety as a Moderator

Hypothesis 7 posited that psychological safety moderates the impact of message conditions on attitude toward allyship, with higher levels of psychological safety enhancing the persuasive effects of the messages. We tested H7 by conducting a moderation analysis using Hayes’ (2022) PROCESS macro Model 1 to explore the influence of psychological safety on the relationship between message conditions and attitude toward allyship. The model summary, shown in Table 7, revealed that the overall model accounted for a significant amount of variance in attitude toward allyship (R² = 0.12, F (7, 304) = 5.79, p < .001), suggesting a moderately strong association between the predictor variables and the outcome.

Table 7.

Coefficients: Message Condition and Safety as Predictors of Allyship Attitude.

Predictors	B	SE (B)	t	LLCI	ULCI	Sig
Intercept	39.79	4.38	9.09	31.17	48.41	<.001***
X1 (Narrative)	9.12	5.51	1.66	−1.71	19.96	.098
X2 (Directive)	16.50	5.55	2.97	5.58	27.41	.003**
X3 (Narr + Directive)	16.05	5.55	2.89	5.14	26.96	.004**
Safety	0.36	0.12	3.04	0.03	.13	.003**
X1 × Safety	−0.12	0.15	−0.81	−0.43	0.18	.421
X2 × Safety	−0.34	0.15	−2.20	−0.64	−0.04	.029*
X3 × Safety	−0.31	0.15	−1.99	−0.61	−0.003	.043*

Note. Referent: Control Message; X1 = Narrative; X2 = Directive; X3 = Narrative + Directive; Analysis: Hayes’ PROCESS macro Model 1 (Hayes, 2022); DV = Attitude Toward Allyship.

R = .343, R² = .118, MSE = 46.11, F (7, 394) = 5.79, p < .001.

p < .05. **p < .01. ***p < .001.

The coefficients for the predictor variables indicated significant effects on attitude toward allyship. Specifically, the directive message (b = 16.50, p = .003) and narrative-plus-directive message (b = 16.05, p = .004) produced significantly more positive attitudes toward allyship compared to the control message. Additionally, the moderator variable, psychological safety, exhibited a significant positive effect on attitude toward allyship (b = 0.36, p = .003), indicating that greater levels of psychological safety were associated with more favorable attitudes toward allyship. These can be seen graphically in the differences in slopes of the lines in Figure 3.

Figure 3.

The moderating role of psychological safety.

Further analysis revealed significant interaction effects between message conditions and psychological safety. The interaction between X2 (difference between directive and control) and psychological safety demonstrated a significant negative association with attitude toward allyship (b = −0.34, p = .028), while the interaction between X3 (the difference between narrative-plus-directive and control) and psychological safety also showed a significant negative association (b = −0.31, p = .047). However, the interaction between X1 (the difference between narrative and control) and psychological safety did not reach statistical significance (b = −0.12, p = .421). The test of highest order unconditional interaction (X*W) yielded a non-significant result (R² change = 0.02, F (3, 304) = 2.20, p = .088), suggesting that the interaction between predictor variables and the moderator did not significantly improve the model’s predictive power beyond the main effects. A post-hoc power analysis (Cohen, 1988) was conducted to evaluate the sensitivity of the analyses for the interaction effects. For the interaction between X1 (the difference between narrative and control) and psychological safety, the power to detect this effect was 0.322, indicating insufficient sensitivity to detect an effect of this magnitude. For the test of the highest-order interaction, the power was 0.057, suggesting that the non-significant finding may reflect low statistical power. Future research with a larger sample size may be needed to adequately test these interaction effects.

Overall, these results highlight the importance of considering psychological safety as a moderator in understanding attitudes toward allyship, with certain predictor variables demonstrating varying effects on attitude toward allyship depending on levels of psychological safety. Therefore, Hypothesis 7 is partially supported.

A summary of the hypotheses and results is presented in Table 8.

Table 8.

Results of Hypothesis Testing.

Effects of directive vs. narrative strategies
H1	A directive allyship advocacy message will produce greater levels of freedom threat than a non-directive message	Directive > Non-Directive, t (151) = 2.63, p = .005 (one-sided) SUPPORTED
H2	A narrative-based allyship advocacy message will produce greater levels of transportation than a control message	Narrative > Control, t (141 = 1.86, p = .015 (one-sided) SUPPORTED
H3	The effects of message strategy on attitude toward allyship will be mediated by transportation, such that a narrative-based message will induce greater levels of transportation compared to other messages, which in turn will lead to greater positive attitude change	b = 0.65, p < .001 SUPPORTED
H4	The effects of message strategy on attitude toward allyship will be mediated by freedom threat, such that a narrative-based message will induce lower levels of freedom threat compared to other messages, which in turn will lead to greater positive attitude change	Narrative > Directive (b_x1 = 1.20, p < .05); Narrative vs Other (b_x2 = 0.38, p > .05, b_x3 = −0.04, p > .05). partially supported
Effects of political identity
H5	Political identity will moderate the mediating effects of transportation on the influence of message strategy on allyship attitude such that the more conservative one is, the more positive the relationship between transportation and allyship attitude.	Trans – Pol Id – Allyship Att b = −0.16, p < .001MStrat – Pol Id - Allyship Att NS. partially supported
H6	Political identity will moderate the mediating effects of freedom threat on the influence of message strategy on allyship attitude such that the more conservative one is, the more negative the relationship between freedom threat and allyship attitude.	b = .05, p = .07not supported
Effects of Psychological Safety
H7	Psychological safety will moderate the influence of the message conditions on attitude toward allyship, wherein greater levels of psychological safety will amplify the persuasive effects of the messages.	Psy. Safety – Allyship Att b = 0.36, p = .003MStrat – Psafety – Allyship Att b = −0.12, p = .421. partially supported

p < .05. **p < .01. ***p < .001.

Discussion

In the context of the changing multicultural landscape with large numbers of newcomers joining the workforce (Gans & Zhan, 2023b) and increased recruitment and retention pressures on organizations (Hofhuis et al., 2014; Liu et al., 2024), this study sheds light on the efficacy and mechanisms of narrative message strategies in promoting newcomer allyship, as well as the nuanced influence of individual and organizational factors on these mechanisms. We found that narrative-based messages elicited lower levels of freedom threat—identified as a pre-cursor to psychological reactance (Rosenberg & Siegel, 2018)—compared to directive-based messages while inducing greater levels of transportation compared to control messages. Furthermore, our results revealed that transportation and freedom threat served as key mechanisms in influencing the effects of narrative messages on attitude toward allyship. Specifically, narrative-based messages facilitated greater levels of transportation and lower levels of freedom threat, thereby enhancing people’s attitude toward newcomer allyship. Our findings also underscored the influence of political identification. Political identity was found to moderate the mediation effects of transportation on attitude change, indicating a more positive relationship between transportation and attitude toward allyship among participants who identified as Conservative. Furthermore, organizational context significantly influenced the effectiveness of message interventions. We found that perceived organizational climate, specifically psychological safety, moderated the influence of message conditions on attitude toward allyship, such that greater levels of psychological safety were associated with amplified persuasive effects of the messages.

In line with theories regarding narrative persuasion and psychological reactance, the results demonstrated that directive-based messages elicited greater levels of freedom threat, potentially triggering defensive responses among recipients. This aligns with studies conducted in the persuasion-reactance literature that messages perceived as controlling or manipulative may lead to resistance and reduced persuasion because they tend to be perceived as freedom threatening (e.g., Rains & Turner, 2007). Furthermore, consistent with Green and Brock’s (2000) results, narrative-based messages, with their immersive storytelling format, engaged individuals on a deeper level, fostering transportation and thereby enhancing persuasion. This supports narrative persuasion theory’s contention that narratives are more effective in promoting attitude change by appealing to emotions and obfuscating persuasive intention (Escalas, 2007; Quick et al., 2013). By evoking empathy, identification, and emotional resonance, narrative messages may activate transportation and overcome resistance, leading to more favorable attitudes toward an advocated-for behavior such as newcomer allyship. Thus, our findings underscore the importance of considering message-processing mechanisms in designing persuasive communication interventions aimed at promoting attitude change in organizations.

Our findings also highlight a nuanced link between political identity and the effectiveness of narrative messages in shaping attitudes toward newcomer allyship, underlining the importance of considering political differences in organizational persuasive efforts, especially initiatives that may be politically charged. Specifically, our study revealed that narrative transportation was a more important factor for promoting positive attitudes toward allyship among conservative individuals than among those who were more liberal, with conservatives who were deeply transported by narrative messages more likely to adopt attitudes aligned with the message content and exhibiting a more positive association between transportation and allyship attitude compared to those who are more liberal. This suggests that narrative messages, with their immersive and emotionally compelling nature, have the potential to counteract the negative influence of a conservative political identity on attitudes toward minority newcomer allyship (cf. Niederdeppe et al., 2014). By overcoming these kinds of individual-difference barriers, narrative messages facilitate transportation, thereby promoting more favorable attitudes. These findings emphasize the need for tailored persuasive strategies that resonate with diverse political perspectives within organizational contexts.

The observed moderation effect of psychological safety on the effectiveness of message interventions aligns with research in organizational climate. Perceptions of organizational climate can influence employee attitudes and behaviors (Schneider et al., 2013). Individuals who perceive their organizational environment as psychologically safe are more likely to feel secure in expressing themselves and engaging with organizational initiatives (Edmondson, 1999; Miller & Katz, 2018). A psychologically safe climate, characterized by trust, openness, and mutual respect, creates a supportive context for communication and encourages employees to engage with organizational messages (Jena & Pradhan, 2020; Miller & Katz, 2018). This sense of safety may foster a conducive environment for message reception and processing, thereby enhancing the persuasive impact of the organizational messages. Consequently, when individuals perceive higher levels of psychological safety, they may be more receptive to the persuasive messages promoting allyship and feel more comfortable engaging in allyship behaviors, leading to more positive attitudes toward allyship. Thus, the moderation effect of psychological safety underscores the importance of organizational context in influencing the effectiveness of persuasive organizational communication strategies aimed at promoting inclusive allyship behaviors within the workplace.

Theoretical Implications

This study contributes to the organizational communication literature in several ways. First, the findings contribute to the understanding of persuasive communication within organizational contexts by shedding light on the effectiveness of different messaging strategies and their interaction with individual and organizational factors in both producing and overcoming resistance to pro-social behavioral recommendations. Virtually every behavioral recommendation produces some degree of freedom threat. The study contributes to the narrative persuasion literature by demonstrating the effects and mechanisms of narrative-based messages’ effect in reducing freedom threat while promoting more positive attitudes toward allyship. Additionally, the identification of moderators such as political identity and perceived organizational climate highlights the importance of considering individual and contextual factors in organizational communication interventions. Moreover, the study underscores the dynamic interplay between individual beliefs, organizational contexts, and persuasive communication strategies, advancing theoretical understanding of how attitudes are formed, changed, and influenced within organizational settings. Overall, the theoretical implications of the study contribute to a deeper understanding of the complexities of persuasive communication in organizations and provide a foundation for further theoretical development and empirical inquiry in this area.

Practical Implications

The findings offer practical insights for organizational and employee communication professionals. First, employee communication professionals can incorporate insights from the study to emphasize narrative-based communication strategies when designing employee newsletters and other forms of employee communication materials to promote allyship toward minority employees. Narrative-based messages are effective in promoting transportation and reducing freedom threat, which in turn contributes to more positive attitudes toward organizational initiatives and priorities such as allyship. For example, storytelling elements that highlight successful allyship efforts can create emotional connections, making the message more relatable and impactful. Furthermore, the study suggests that strategic communicators’ planning related to employee communication should consider their target audiences’ individual differences, such as political identities, to ensure messages effectively engage diverse audiences. For instance, employees with more conservative political identities may respond differently than those who are more liberal, necessitating careful considerations of employee political orientation to maximize resonance and reduce potential resistance. Additionally, cultivating a psychologically safe organizational climate, characterized by trust and mutual respect, enhances the persuasive impact of messages promoting attitudes toward allyship and inclusion. Psychological safety encourages employees to engage with organizational messaging more openly, reducing defensiveness and fostering a greater willingness to adopt positive attitudes toward allyship and other diversity-related initiatives.

Limitations and Future Directions

While this study provides valuable insights into persuasive organizational communication, several limitations should be acknowledged. First, the study relied on several self-report measures, which are susceptible to biases such as social desirability (Pedregon et al., 2012). Future research could employ more objective measures or incorporate multiple data sources to enhance the robustness of the findings. Additionally, the study prompted participants to think about the organization that they are working for and there may be other confounding factors related to the organizations. Future research could explore these dynamics in different organizational contexts or cultural settings to provide a more comprehensive understanding of the factors influencing attitudes toward allyship. Furthermore, while the study examined the influence of political identity on message effectiveness, it did not delve into other individual differences such as personality traits or demographic factors, which could also play a role in shaping responses to persuasive messages. Future studies could explore the interplay of these variables to provide a more nuanced understanding of attitude change processes. Lastly, the study utilized cross-sectional data, which limits the ability to examine longitudinal effects over time. Future research could employ longitudinal designs to better understand the long-term effects of message interventions on attitudes toward allyship in organizations (Parker et al., 2022).

Overall, the results of this study suggest that advocating for allyship produces a perception of freedom threat, and that the most effective strategies for such advocacy must address, avoid or overcome those perceptions.

Footnotes

Appendix: Directive,Narrative,and Control Message Conditions

Directive:

Narrative:

Control:

Author’s Note

The current manuscript is an original work of research that is not under consideration and has not been published elsewhere.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research presented in this article was supported in part by a grant from the UNniversity of Tex at Arlington’s College of Liberal Arts.

ORCID iDs

Roger Gans

Mengqi Monica Zhan

Author Biographies

Roger Gans, is an Assistant Professor in the Department of Communication at the University of Texas at Arlington, where he teaches courses in strategic communication and integrated marketing communications from the perspective of a working professional with more than 25 years’ experience as a copywriter and creative director in the advertising industry. His research focuses on persuasion strategies for promotion of pro-social behavior in the areas of health, the workplace and the community.

Mengqi Monica Zhan, is an Associate Professor in the Department of Communication at the University of Texas at Arlington, where she teaches organizational communication and research methods. Her research interests center around how interpersonal communication processes and social networks influence team information elaboration processes, team performance, and organizational crisis management.

References

Al-Gharbi

(2020, December 5). Research shows diversity training is typically ineffective. RealClear Science. https://www.realclearscience.com/articles/2020/12/05/research_shows_diversity_training_is_typically_ineffective_652014.html

Allen

D. G.

Bryant

P. C.

Vardaman

J. M.

(2010). Retaining talent: Replacing misconceptions with evidence-based strategies. The Academy of Management Perspectives, 24(2), 48–64. https://doi.org/10.5465/amp.24.2.48

Bashein

B. J.

Markus

M. L.

(1994). Preconditions for BPR success. Information Systems Management, 11(2), 7–13. https://doi.org/10.1080/10580539408964630

Beer

Eisenstat

R. A.

Spector

(1990). Why change programs don’t produce change. Harvard Business Review, 68(6), 158–166. https://hbr.org/1990/11/why-change-programs-dont-produce-change

Beigi

Callahan

Michaelson

(2019). A critical plot twist: Changing characters and foreshadowing the future of organizational storytelling. International Journal of Management Reviews, 21(4), 447–465. https://doi.org/10.1111/ijmr.12203

Bensley

L. S.

(1991). The role of psychological reactance in drinking following alcohol prevention messages. Journal of Applied Social Psychology, 21(13), 1111–1124. https://doi.org/10.1111/j.1559-1816.1991.tb00461.x

Boyce

M. E.

(1996). Organizational story and storytelling: A critical review. Journal of Organizational Change Management, 9(5), 5–26. https://doi.org/10.1108/09534819610128760

Brady

W. H.

Haley

(2013). Storytelling defines your organizational culture. Physician Executive, 39(1), 40–43. https://pubmed.ncbi.nlm.nih.gov/23437755/

Brehm

J.W.

(1989). Psychological reactance: Theory and applications. Advances in Consumer Research, 16(1), 72–75. https://www.acrwebsite.org/volumes/6883/volumes/v16/NA-16

10.

Brown

K. T.

Ostrove

J. M.

(2013). What does it mean to be an ally? The perception of allies from the perspective of people of color. Journal of Applied Social Psychology, 43(11), 2211–2222. https://doi.org/10.1111/jasp.12172

11.

Burgoon

Alvaro

Grandpre

Voloudakis

(2002). Revisiting the theory of psychological reactance: Communicating threats to attitudinal freedom. In Dillard

Pfau

(Eds.), The persuasion handbook: Theory and practice (pp. 213–232). Thousand Oaks, CA: Sage. https://dx.doi.org/10.4135/9781412976046.n12

12.

Cohen

(1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Lawrence Erlbaum Associates.

13.

Collier-Spruel

L. A.

Ryan

A. M.

(2024). Are all allyship attempts helpful? An investigation of effective and ineffective allyship. Journal of Business and Psychology, 4, 1–26. https://doi.org/10.1007/s10869-022-09861-9

14.

Detert

J. R.

Edmondson

A.C.

(2011). Implicit voice theories: Taken-for-granted rules of self-censorship at work. Academy of Management Journal, 54(3), 461–488. https://doi.org/10.5465/amj.2011.61967925

15.

Dillard

J. P.

Shen

(2005). On the nature of reactance and its role in persuasive health communication. Communication Monographs, 72(2), 144–168. https://dx.doi.org/10.1080/03637750500111815

16.

Dobbin

Kalev

(2016). Why diversity programs fail. Harvard Business Review. https://hbr.org/2016/07/why-diversity-programs-fail

17.

Dobbin

Kalev

(2018). Why doesn’t diversity training work? The challenge for industry and academia. Anthropology Now, 10, 48–55. https://doi.org/10.1080/19428200.2018.1493182

18.

Dobbin

Kalev

(2021). The civil rights revolution at work: What went wrong. Annual Review of Sociology, 47, 281–303. https://doi.org/10.1146/annurev-soc-090820-023615

19.

Edmondson

(1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999

20.

Edmondson

A. L.

Besieux

(2021). Reflections: Voice and silence in workplace conversations. Journal of Change Management, 21(3), 269–286 https://doi.org/10.1080/14697017.2021.1928910

21.

Escalas

J. E.

(2007). Self-referencing and persuasion: Narrative transportation versus analytical elaboration. Journal of Consumer Research, 33(4), 421–429. https://doi.org/10.1086/510216

22.

Ferdman

B. M.

(2014). The practice of inclusion in diverse organizations: Toward a systemic and inclusive framework. In Ferdman

B. M.

Deane

B. R.

(Eds.), Diversity at work: The practice of inclusion (pp. 3–54). Jossey-Bass/Wiley. https://doi.org/10.1002/9781118764282.ch1

23.

Gans

(2014). The politics of HPV vaccination advocacy: Effects of source expertise on effectiveness of a pro-vaccine message. Proceedings of the New York State Communication Association: Vol. 2013, Article 3. https://docs.rwu.edu/nyscaproceedings/vol2013/iss2013/3

24.

Gans

Zhan

(2023a). A story about speaking up: Mediation effects of narrative persuasion on organizational voice intentions. International Journal of Business Communication, 60(3), 865–891. https://doi.org/10.1177/2329488422109127

25.

Gans

Zhan

(2023b). Let’s influence that attitude before it’s formed: Inoculation against reactance to promote DEI training. International Journal of Business Communication, 0(0),1–25. https://doi.org/10.1177/23294884231216952

26.

Graham

Hodsdon

Busse

Crosby

M. P.

(2023). BIPOC voices in ocean sciences: A qualitative exploration of factors impacting career retention. Journal of Geoscience Education, 71(3), 369–387. https://doi.org/10.1080/10899995.2022.2052553

27.

Green

M. C.

Brock

T. C.

(2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5), 701–721. https://doi.org/10.1037/0022-3514.79.5.701

28.

Green

M. C.

Brock

T. C.

(2002). In the mind’s eye: Transportation-Imagery model of narrative persuasion. In Green

M. C.

Strange

J. J.

Brock

T. C.

(Eds.), Narrative impact: Social and cognitive foundations (pp. 315–341). Erlbaum.

29.

Gubler

J. R.

Karpowitz

C. F.

Monson

J. Q.

Romney

D. A.

South

(2022). Changing hearts and minds? Why media messages designed to foster empathy often fail. Journal of Politics, 84(4), 2156–2171. https://doi.org/10.1086/719416

30.

Hall

K. R.

Harrison

D. E.

Obilo

O. O.

(2023). Building positive internal and external stakeholder perceptions through CSR storytelling. Journal of Strategic Marketing. 31(7), 1317–1338. https://doi.org/10.1080/0965254X.2021.1895289

31.

Hamby

Brinberg

Daniloski

(2017). Reflecting on the journey: Mechanisms in narrative persuasion. Journal of Consumer Psychology, 27(1), 11–22. https://doi.org/10.1016/j.jcps.2016.06.005

32.

Hancock

G. R.

Mueller

R. O.

Stapleton

L. M.

(2010). The reviewer’s guide to quantitative methods in the social sciences. Routledge.

33.

Hofhuis

Van der Zee

K. I.

Otten

(2014). Comparing antecedents of voluntary job turnover among majority and minority employees. Equality, Diversity and Inclusion: An International Journal, 33(8), 735–749. https://doi.org/10.1108/EDI-09-2013-0071

34.

Hayes

A. F.

(2022). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Publications.

35.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. https://doi.org/10.1080/10705519909540118

36.

Hughes

(2011). Do 70 per cent of all organizational change initiatives really fail? Journal of Change Management, 11(4), 451–464. https://doi.org/10.1080/14697017.2011.630506

37.

Janssen

J. D.

(2024). Diversifying aquatic husbandry: Creating a wake of equity and inclusion, Journal of Museum Education, 49(3), 296–305. https://doi.org/10.1080/10598650.2024.2386486

38.

Jena

L. K.

Pradhan

(2020). Workplace persuasion: Conceptual model, development and validation. Global Business Review, 21(2), 567–585. https://doi.org/10.1177/0972150918761086

39.

Kahan

D.M.

Braman

Cohen

Gastil

Slovic

(2010). Who fears the HPV vaccine, who doesn’t, and why? An experimental study of the mechanisms of cultural cognition, Law & Human Behavior, 34(6), 501–516. https://doi.org/10.1007/s10979-009-9201-0

40.

Kahan

D.M.

Jenkins-Smith

Braman

(2011). Cultural cognition of scientific consensus. Journal of Risk Research, 14(2), 147–174. https://doi.org/10.1080/13669877.2010.511246

41.

Kawasaki

Zou

(2023). A meta-analytic evaluation of diversity training outcomes across cultures. Social and Personality Psychology Compass, 45, e12741. https://doi.org/10.1111/spc3.12741

42.

Kiehne

Hafen

(2022). Advancing active allyship for social justice: Cross-group friendships, social empathy, and political engagement. Social Work Research, 46(3), 193–204. https://doi.org/10.1093/swr/svac017

43.

Knight

Patterson

Dawson

(2017). Building work engagement: A systematic review and meta-analysis investigating the effectiveness of work engagement interventions. Journal of Organizational Behavior, 38(6), 792–812. https://doi.org/10.1002/job.2167

44.

Kotter

J.P

(1995). Leading change: Why transformation efforts fail. Harvard Business Review, 73(2), 59-67. https://hbr.org/1995/05/leading-change-why-transformation-efforts-fail-2

45.

Lane

(2023) Towards a theory of organizational storytelling for public relations: An engagement perspective. Public Relations Review, 49(1), 1–10. https://doi.org/10.1016/j.pubrev.2023.102297

46.

Lipman

(2018, January 25). How diversity training infuriates men and fails women. Time Magazine. https://time.com/5118035/diversity-training-infuriates-men-fails-women/

47.

Liu

Watts

Feng

Yin

(2024). Unpacking the effects of socialization programs on newcomer retention: A meta-analytic review of field experiments. Psychological Bulletin, 150(1), 1–26. https://doi.org/10.1037/bul0000422

48.

Miller

C.H.

Burgoon

Grandpre

J.R.

Alvaro

E.M.

(2006). Identifying principal risk factors for the initiation of adolescent smoking behaviors: The significance of psychological reactance. Health Communication, 19(3), 241–252. https://doi.org/10.1207/s15327027hc1903_6

49.

Miller

C. H.

Lane

L. T.

Deatrick

L. M.

Young

A. M.

Potts

K. A.

(2007). Psychological reactance and promotional health messages: The effects of controlling language, lexical concreteness, and the restoration of freedom. Human Communication Research, 33(2), 219–240. https://doi.org/10.1111/j.1468-2958.2007.00297.x

50.

Miller

F. A.

Gans

(2002, January). There is nothing more diverse than “new.” Cultural Diversity at Work, 1-2. https://mavmatrix.uta.edu/communication_facpubs/1/

51.

Miller

F. A.

Katz

J. H.

(2018). Safe enough to soar: Accelerating trust, inclusion and collaboration in the workplace. Berrett-Koehler Publishers.

52.

Moyer-Gusé

(2008). Toward a theory of entertainment persuasion: Explaining the persuasive effects of entertainment-education messages. Communication Theory, 18(3), 407–425. https://doi.org/10.1111/j.1468-2885.2008.00328.x

53.

Moyer-Gusé

Nabi

R. L.

(2010). Explaining the effectives of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26–52. https://doi.org/10.1111/j.1468-2958.2009.01367

54.

Niederdeppe

Shapiro

M. A.

Kim

H. K.

Bartolo

Porticella

(2014). Narrative persuasion, causality, complex integration, and support for obesity policy. Health Communication, 29(5), 431–444. https://doi.org/10.1080/10410236.2012.761805

55.

Ostrove

J.M.

Brown

K.T.

(2018). Are allies who we think they are? A comparative analysis. Journal of Applied Social Psychology, 48(4), 195–204. https://doi.org/10.1111/jasp.12502

56.

Ostrove

J. M.

Kornfeld

Ibrahim

(2019). Actors against ableism? Qualities of nondisabled allies from the perspective of people with disabilities. Journal of Social Issues, 75(3), 924–942. https://doi.org/10.1111/josi.12346

57.

Parker

K. A.

Ivanov

Matig

Dillingham

L. L.

Peritore

(2022). Inoculation booster messages: Frequency, content, and timing. The Journal of Communication and Media Studies, 7(1), 1. https://doi.org/10.18848/2470-9247/CGP/v07i01/1-19

58.

Pedregon

C. A.

Farley

R. L.

Davis

Wood

J. M.

Clark

R. D.

(2012). Social desirability, personality questionnaires, and the “better than average” effect. Personality and Individual Differences, 52(2), 213–217. https://doi.org/10.1016/j.paid.2011.10.022

59.

Perkins

Merola

R. H.

Ghosh

Aragon

(2024). ‘I’m a pilot first, female second': Why flight deck gender imbalance persists and the case for allyship. Journal of Aviation/Aerospace Education & Research, 33(2), 258–292. https://doi.org/10.58940/2329-258X.2025

60.

Pollack

(2023, October 11). The top 10 talent acquisition challenges of 2023. ZipRecruiter Blog. https://www.ziprecruiter.com/blog/talent-trends-report/

61.

Prolific. (2024, May 10). Participant pool. Prolific.com. https://www.prolific.com/participant-pool

62.

Pung

L.-X.

Goh

Y.-S.

(2017). Challenges faced by international nurses when migrating: An integrative literature review. International Nursing Review, 64(1), 146–165. https://doi.org/10.1111/inr.12306

63.

Quick

B. L.

Shen

Dillard

J. P.

(2013). Reactance theory and persuasion. In Dillard

J. P.

Shen

(Eds.), The SAGE handbook of Persuasion: Developments in theory and practice (2nd ed., pp. 167–183). SAGE. https://doi.org/10.4135/9781452218410.n11

64.

Rains

S. A.

Turner

M. M.

(2007). Psychological reactance and persuasive health communication: A test and extension of the intertwined model. Human Communication Research, 33(2), 241–269. https://doi.org/10.1111/j.1468-2958.2007.00298.x

65.

Ratcliff

C. L.

Sun

(2020). Overcoming resistance through narratives: Findings from a meta-analytic review. Human Communication Research, 46(4), 412–443. https://doi.org/10.1093/hcr/hqz017

66.

Reeves

Loughhead

Teague

Halpin

M. A.

Procter

(2024). Lived experience allyship in mental health services: Recommendations for improved uptake of allyship roles in support of peer workforces. International Journal of Mental Health Nursing, 33(5), 1191–1601. https://doi.org/10.1111/inm.13322

67.

Rhodes

Brown

A. D.

(2005). Narrative, organizations and research. International Journal of Management Reviews, 7(3), 167–188. https://doi.org/10.1111/j.1468-2370.2005.00112.x

68.

Rosenberg

B. D.

Siegel

J. T.

(2018). A 50-year review of psychological reactance theory: Do not read this article. Motivation Science, 4(4), 281–300. https://doi.org/10.1037/mot0000091

69.

Schneider

Ehrhart

M. G.

Macey

W. H.

(2013). Organizational climate and culture. Annual Review of Psychology, 64, 361–388. https://doi.org/10.1146/annurev-psych-113011-143809

70.

Sheehan

K. B.

(2018). Crowdsourcing research: Data collection with Amazon’s Mechanical Turk. Communication Monographs, 85, 140–156. https://doi.org/10.1080/03637751.2017.1342043

71.

Shenoy

(2023). Purple hiring: A revolution in workplace inclusivity: How industries are embracing persons with disabilities (PWD) through innovative practices. Strategic HR Review, 22(6), 216–218. https://doi.org/10.1108/SHR-10-2023-0056

72.

Slater

M. D.

Rouner

Long

(2006). Television dramas and support for controversial public policies: Effects and mechanisms. Journal of Communication, 56(2), 235–252. https://doi.org/10.1111/j.1460-2466.2006.00017.x

73.

Snyder

Bottorff

(2023, May 17). Key HR statistics and trends in 2024. Forbes Advisor. https://www.forbes.com/advisor/business/hr-statistics-trends/

74.

Stanton

Carpenter

R. W.

Nance

Sturgeon

Villalongo Andino

(2022). A multisample demonstration of using the Prolific platform for repeated assessment and psychometric substance use research. Experimental and Clinical Psychopharmacology, 30(4), 432–443. https://doi.org/10.1037/pha0000545.supp (Supplemental)

75.

van Laer

De Ruyter

Visconti

L. M.

Wetzels

. (2014). The extended transportation- imagery model: A meta-analysis of the antecedents and consequences of consumers’ narrative transportation. Journal of Consumer Research, 40(5), 797–817. https://doi.org/10.1086/673383

Advocating for Allyship: Understanding the Effects of Narrative Transportation,Freedom Threat,and More on Minority-Newcomer Retention Strategies

Abstract

Keywords

Addressing Gaps in the Study of Allyship

Anticipating Reactance and Resistance

Threat to Personal Freedom as a Cause of Resistance

A Narrative Approach to Reducing Resistance

The Influence of Political Identification

The Influence of Perceived Organizational Climate: Psychological Safety

Method

Sample

Study Procedures

Control Condition

Narrative Condition

Directive Condition

Narrative plus Directive Condition

Measures

Attitude toward Allyship

Transportation

Freedom Threat

Political Identity

Psychological Safety

Manipulation Checks

Measurement Validity

Analytical Strategies

Results

Hypothesis 1: Directive Messages and Reactance

Hypothesis 2: Narrative Messages and Transportation

Hypothesis 3: Transportation as a Mediator

Hypothesis 4: Freedom Threat as a Mediator

Hypothesis 5: Political Identity in the Moderated Mediation Model for Transportation

Hypothesis 6: Political Identity in the Moderated Mediation Model for Freedom Threat

Hypothesis 7: Psychological Safety as a Moderator

Discussion

Theoretical Implications

Practical Implications

Limitations and Future Directions

Footnotes

Appendix: Directive,Narrative,and Control Message Conditions

Author’s Note

Declaration of Conflicting Interests

Funding

ORCID iDs

Author Biographies

References