Being tolerated as a minority group member: an experimental study with virtual teams

ABSTRACT

Research on the experience of being tolerated has focused on single events, ignoring the important question of whether the experience of being tolerated depends on previous experiences. We examined whether the experience of being tolerated has a negative impact on minority team members depending on whether they had previously been rejected or fully accepted. In a pre-registered study with 440 participants, we used a recently developed experimental paradigm to simulate workstyle minority status in virtual teams. These participants were randomly assigned to experience rejection or acceptance followed by being tolerated. Experiencing tolerance after rejection was strongly positive, reducing negative well-being, increasing positive future expectations about interactions with majority team members, and reducing people’s intention to withdraw from their teams. By contrast, experiencing tolerance after acceptance was weakly but consistently negative, with increased negative well-being, increased negative future expectations, and increased withdrawal intentions. Lastly, despite tolerance being more harmful than acceptance, that harmfulness did not translate into greater willingness to raise one’s voice and express discontent about not being valued.

KEYWORDS:

• Diversity
• minorities
• tolerance
• virtual teams
• workplace

Companies, organizations and teams are increasingly made up of people who have different backgrounds and worldviews. There are various forms of organizational and team diversity which have been theorized and found to be beneficial for functioning, performance, and innovation (Ng & Stephenson, 2015). Under the right circumstances, diverse teams can work together congenially and form socially cohesive and productive units. However, so-called deep-level diversity in personalities, values, and working styles (Bell, 2007) can also be a source of tension, which makes it difficult for team members to get along (e.g., Van Knippenberg & Schippers, 2007). People have their own ways of doing things, their own views and convictions and they are unlikely to equally appreciate and value all habits, beliefs, and behaviors of others. Indeed, efforts to bring everyone’s personal values into alignment is likely to cause more conflict than it resolves (e.g., Danışman, 2010).

However, what can be expected is that people be tolerant in showing forbearance and not exclude or behaviorally reject those who think and work differently. Tolerance is widely considered an indispensable ingredient for allowing deep-level differences to co-exist and teams and communities to function (e.g., Verkuyten, 2023; Von Bergen et al., 2012). Whereas intolerance causes tensions and hostilities between individuals and groups (e.g., Sekulić et al., 2006), tolerance makes it possible to live and work together despite contrasting and conflicting perspectives about what people (ought to) think and do. Tolerance thus offers an important answer to the question of how to manage meaningful differences in communities, organizations, and teams despite the inevitable disagreements, dislikes, and disapprovals that will exist.

Forbearance tolerance arises when people recognize that they strongly disagree with or disapprove of what others think and do, but simultaneously acknowledge the importance of not ignoring, silencing, or excluding these others (Cohen, 2004; King, 2012; Verkuyten et al., 2022). Forbearance tolerance recognizes that the social environment will always include people with very different styles, values, and views that one does not agree with, but has to endure for many diverse people to co-exist and work together. Tolerance is especially relevant for those who are in a position to indulge others. For example, majority members can tolerate minorities while the question of tolerance is less relevant for minority members because they are typically not in the position to negatively interfere with the behavior of the majority.

Building tolerance in organizations and teams is considered a neglected but essential part of making members get along and enhance cooperation between them, despite their disagreements, disapprovals, and even dislikes (Gebert et al., 2017; Von Bergen et al., 2012). However, while the benevolent side of tolerance protects the expression of personal identities and enables deep-level workplace diversity, tolerance may also have a malevolent side in signaling to minority members that their beliefs, practices, and working style are less appreciated and valued and merely put up with (“mere tolerance”). Being tolerated might be offensive and hurtful because it implies disapproval of what one values, believes, and practices. Theoretically it has been proposed that the experience of being tolerated as begrudging inclusion can have a negative impact on minority members’ psychological well-being and expectations about future interactions with team members. Furthermore, the presumed benevolence of tolerance may make people feel less comfortable and willing to express their dissatisfaction about the disapproval that one faces, thus cutting off one of the most effective means of gathering support for positive change (Verkuyten et al., 2020).

Previous research on the implications of being tolerated for minority members suggests that the experience of being tolerated is a more positive one compared to being behaviorally rejected, but less positive compared to being appreciated and fully accepted (Adelman et al., 2023, Cvetkovska et al., 2021). However, this research relied on survey data and correlational designs and people’s own definitions of the experience of tolerance, making it hard to draw firm conclusions of how being tolerated is experienced (Bagci et al., 2020; Cvetkovska et al., 2021). The limited experimental work conducted on the topic has focused on single experiences and used between-subjects experimental design that yields average differences between experimental conditions (Adelman et al., 2023; Cvetkovska et al., 2021). However, average experimental effects between random groups of individuals do not necessarily reflect changes that can occur within individuals in response to prior experiences (Hamaker, 2011). Thus we do not know how tolerance is experienced in light of previous experiences that a person would have, despite the fact that an important aspect of work environments is the repeated nature of encounters and exposure, each of which can inform the experience of the next. Specifically, being tolerated after first having experienced behavioral rejection might have a different meaning and impact than being tolerated after first having experienced full acceptance. It is likely that in the first situation being tolerated is comparatively a more positive experience than the second situation because the former one implies an increase in positive valence and the latter one a decrease.

In order to make a novel contribution and to further our understanding of the social psychological implications of being tolerated, we aimed to examine the experience of being tolerated as a function of previous experiences of being accepted or rejected. We conducted this research using a newly developed experimental manipulation (Adelman et al., 2023) with workstyle minority members being placed into two virtual teams, one after the other, either experiencing acceptance or rejection followed by being tolerated, and, for comparative purposes, being tolerated followed by being accepted or being rejected. Virtual teams in which members use electronic communication to collaborate from distant locations are increasingly common and an established reality in many companies, organizations, and teams.

Being tolerated

Forbearance tolerance has been characterized as being “intermediate between wholehearted acceptance and unrestrained opposition” (Scanlon, 2003, p. 187), and there is evidence that the subjective experience of being tolerated falls in between that of full acceptance and complete rejection (Adelman et al., 2023; Cvetkovska et al., 2021). In the literature, the constructs of “acceptance,” “rejection” and “tolerance” have diverse meanings and are used in different and sometimes overlapping ways (see Verkuyten, 2023). One useful way of making a clear conceptual distinction between these three constructs is to adopt the common theoretical distinction between attitude and behavior. Acceptance, rejection, and tolerance each can be considered to comprise of two components: the underlying attitude and the corresponding behavior. In the first two cases, which are often presented as the main contrasting responses to difference and diversity, there is no gap between the attitude and the behavior. Acceptance can be conceptualized as being welcoming in attitude and inclusive in behavior, such as someone who values and welcomes cultural minority trainees within their team. Rejection is when you are attitudinally negative and exclusionary in behavior, such as someone who dislikes trainees and excludes them in their professional life. By contrast, tolerance sits in between these with the tolerating attitude involving the negativity of rejection that is combined with the inclusion in behavior, such as someone who dislikes trainees, but nonetheless equally includes them in their team. This unique balance is what makes tolerance as powerful a tool as it is: it does not necessitate changing the mind and internal values, but rather seeks to block the link between disapproving attitude and the corresponding behavior with the value of noninterference in other peoples’ lives. Thus, just as tolerance falls in-between acceptance and rejection, the implications of being tolerated can be expected to be in-between being accepted and being rejected because tolerance shares with acceptance the behavioral inclusion, but differs in underlying attitude, and it shares with rejection the underlying negative attitude, but differs in behavior.

A theoretical process model of being tolerated has been proposed that focuses on the nature of being tolerated as a minority member and the implied devaluation of minority practices and beliefs (Verkuyten et al., 2020). This model argues that the experience of being tolerated as begrudging inclusion can have a negative impact: simply knowing that as a minority member one is being tolerated, and thus objected to but not excluded, can still be offensive and hurtful: it implies disapproval and devaluation of what minorities believe, value, and practice, and people typically do not find it desirable to be begrudgingly included, but rather want to be valued and respected (Bergsieker et al., 2010; Muragishi et al., 2023). The process model argues that forbearance tolerance can be considered as patronizing and condescending to those who are being tolerated, and might also prevent those who are tolerated from feeling morally or practically comfortable with raising their voice and, for example, express suggestions for improving work-related issues. Therefore, although tolerance protects against behavior exclusion, it often comes with the subjective awareness that one is being endured and “put up with” which might negatively affect minority members’ well-being, their expectations of how future interactions with majority members might go, and their willingness to speak out and indicate discontent and strive for change.

To experimentally test these possible effects, we utilized an adaptation of the well-established cyberball paradigm (Hartgerink et al., 2015; Williams & Jarvis, 2006), to simulate the experience of being a minority virtual team member interacting with majority group teammates who express acceptance, toleration, or rejection. We opted to test the impact of being tolerated in the context of a study of online work environments which not only are increasingly common but in which rejection can occur and acceptance is feasible. An online team setting offers a reasonable cover story for participants’ experiences. In doing so, we are able to manipulate experimentally workstyle minority experiences of being tolerated vs. accepted vs. rejected and test its effects on the three key dependent outcomes for minority group members that have been proposed in the theoretical process model (Verkuyten et al., 2020): psychological well-being (Pascoe & Smart Richman, 2009; Schmitt et al., 2014) which captures the possible negative feelings experienced by a target of tolerance; interpersonal motivation and expectations about future participation and engagement with majority team members (Smart Richman & Leary, 2009), which captures the possibility that being merely tolerated might lead minority group members self-exclude from interactions with the majority; and willingness to raise one’s voice through offering participants the opportunity to express their discontent and take action against the majority group members disapproval (e.g., Bashshur & Oc, 2015), which captures the possibility that tolerance might lead to an unwillingness to push for full acceptance.

We predicted, first, that being tolerated would have a more negative effect on well-being and expectations of future treatment compared to acceptance and, importantly, especially after an experience of acceptance. By contrast, being tolerated is expected to have a less negative effect on well-being and future expectations compared to being rejected, and especially after first being rejected. When it comes to willingness to raise one’s voice and express discontent, we expected to find evidence for this particularly pernicious element of tolerance, that despite the negative implications of being endured, people are unwilling to complain. Thus, we unsurprisingly expect that, compared to being rejected and especially after first being rejected, tolerance leads to decreased willingness to complain. However, and more importantly, despite our prediction of being tolerated reducing well-being and future expectations compared to and after acceptance, we predict that people who are tolerated will nonetheless be unwilling to raise their voice and complain about the disapproval that they face.

Across these three types of outcomes, we therefore generally expect that being tolerated would be a negative experience following acceptance, but a rather positive experiencing after being first rejected: the former situation involves a decrease in positive valence (from acceptance to tolerance), and in the latter case there is an increase in valence (from rejection to tolerance). Similarly, when first experiencing tolerance, we expected acceptance to be positive and rejection to be especially negative. Ethical approval for the study was obtained by the corresponding author at their institution before data collection, and the study was preregistered and all measures and manipulations in the study are disclosed.¹

Method

Participants

We recruited 440 participants (10.7% 18–25 years-old, 38.2% 26–35, 27.7% 36–45, 14.5% 46–55, 6.1% 56–65, 2.3% 66–74, 0.5% 75+) from Cloud Research/MTurk for this study: 44.5% female, 54.5% male; 0.9% other; 74.5% white, 9.5% black, 5.7% Asian, and 6.4% mixed.²

In the absence of good priors of expected effect size, we conducted the study aiming for at least 100 participants per each of four conditions (see condition details below). Using the parameters of the present sample of 407 participants, with α = .05 and power of .80, we achieved sufficient sensitivity to consistently find at least small to medium effects, η_p² = .026.

Materials

To investigate the experience of being tolerated, and if this is impacted by previous experiences of being accepted or rejected, and also how experiences of tolerance impact later experiences of acceptance or rejection, we used materials developed in previous experimental research on the implications of being tolerated (Adelman et al., 2023). One important change we made was that participants were assigned to two virtual teams, repeating the same exercises and measures twice.

As the three experiences of being accepted, rejected, and tolerated reflect three different combinations of attitude and behavior, the experiment was designed to create those experiences while maintaining a believable cover story for participants to generate meaningful data. The experimental manipulation thus had two primary goals: (a) that participants experience being a minority team member, and (b) that participants believe that their virtual workgroup teammates either appreciate and include them (acceptance condition), disapprove of them but nevertheless equally include them (tolerance condition), or disapprove of them and exclude them (rejection condition).

Procedure

To accomplish these two goals, participants first completed a set of eight items purportedly designed to test their work-related personality style as a people- vs. task-oriented person, after which they all received bogus feedback that they were people-oriented. Next, participants were randomly shown a set of three comics depicting a people-oriented character either accepting, tolerating, or rejecting a task-oriented character and asked which of the three best matched their preferred approach. This was done primarily to prepare participants to find out about their (task-oriented) teammates’ preferences about working with them (a person-oriented team member). Then they were introduced to their three alleged workgroup teammates who were all task-oriented, leaving the participant in a work-related minority position. Further, the three teammates all had indicated on the comic task that they either accepted, tolerated, or rejected people-oriented teammates, such as the participant, for the task ahead. Finally, participants engaged in a cyberball game, based on the classic ostracism manipulation (Williams & Jarvis, 2006), where they were asked to imagine being together in real life with their teammates while passing a ball between them. In the acceptance as well as the tolerance conditions, the participant was behaviorally fully included in the game, while in the rejection condition, the participant was fully excluded after a first round of ball throwing. Thus, we created an online setting to mirror virtual working situations where minority group members might feel and experience being behaviorally accepted, tolerated, or rejected by having interactions with majority group virtual teammates who attitudinally either accepted their difference (being people-oriented among task-oriented people), tolerated their difference, or rejected it.

Design

In order to test the experiences of being tolerated in the context of previous experiences, we constructed four experimental conditions: (a) participants first experienced being tolerated and then being rejected (n = 101), (b) first tolerated and then being accepted (n = 125), (c) first being accepted and then tolerated (n = 111), or (d) first being rejected, and then tolerated (n = 103). This enables us to answer three questions: (1) What is the effect of being tolerated compared to being accepted and being rejected? (2) How do experiences of being first accepted or rejected affect a later experience of being tolerated? (3) How does being tolerated first affect later experiences of being accepted or rejected? As part of the set of questions that participants answered for each of the two virtual teams they were placed with, they also answered a single item manipulation check asking how participants felt during the team exercises using 11-point scales from complete rejected (1) to tolerated (6) to fully accepted (11; see Adelman et al., 2023).

The participants responded to all outcome variables twice: once following the first “team” manipulation, and once following the second. The variables involved the three expected outcomes of being tolerated (Verkuyten et al., 2020), namely psychological well-being, future interaction expectations, and willingness to express discontent.

For psychological well-being, participants completed 11-item measures of negative personal well-being (M = 3.53; SD = 1.51; α = .93), which drew on elements of belongingness, sense of control, self-esteem, and meaningfulness used in previous research (e.g., Bagci et al., 2020; Van Beest & Williams, 2006). For measuring future interaction expectations, they completed a four-item measure (M = 3.95; SD = 1.58; α = .95) indicating how much they would expect their teammates to be open to their opinions and to value, incorporate, and solicit their suggestions in hypothetical future activities (see Adelman et al., 2023). We also measured expectations of future withdrawal from the team (M = 3.26; SD = 1.23; α = .70) using four items which asked whether participants expected to participate less and be less likely to engage in future activities with the team. Finally, participants indicated their willingness to raise their voice in relation to their teammates (M = 1.90; SD = 1.08; r = .66), using two-items which asked whether they had trouble with their teammates, and whether they wanted those teammates to be excluded from future teamwork activities (Cvetkovska et al., 2021). The same variables were measured for the second team manipulation: negative well-being (M = 4.09; SD = 1.83; α = .97), future expectations (M = 3.95; SD = 1.63; α = .95), expectations of personal behavior in the team (M = 3.26; SD = 1.49; α = .81), and willingness to raise voice (M = 1.97; SD = 1.10; r = .58).

Lastly, we included three vignettes of potential workplace dilemmas in which participants were asked to make a dichotomous choice which is partially based on their trust of each of their three teammates. These prisoner’s dilemma trust items were included in both the first and the second manipulation sets and served primarily to maintain the cover story of a study on teamwork.

Analysis

Analyses were conducted with mixed ANOVAs in SPSS to test how people experienced being tolerated, depending on whether they had first experienced acceptance or rejection, and how first being tolerated affects interpretations of subsequently being accepted or rejected. The labels “first experience” and “second experience” that we use below in discussing the results refer to the order of the experimental manipulations in the within-subjects experimental design.

Results

Manipulation check for first experience

We first analyzed the four experimental condition differences across the first manipulation that participants experienced (i.e., two groups which had experienced tolerance, one group which experienced rejection, and one group which experienced acceptance). To ensure that the manipulations worked as intended, we examined the single-item continuous manipulation check measure where higher scores a indicate greater feeling of being accepted and lower scores indicate a greater feeling of being rejected with tolerance in the middle. The finding showed that the manipulation performed as intended, F(3, 435) = 141.877, p < .001, η_p² = .495, as the different conditions were distinct from each other. Specifically, and as shown in Table 1, Tukey-corrected post-hoc analyses found that those in the acceptance condition indicated relatively higher levels of acceptance than those in the rejection condition and those in the two tolerance conditions, while those in the rejection condition indicated lower feelings of acceptance (and more rejection) than those in the tolerance conditions. Having found a successful effect of the conditions on the manipulation check, we then sought to answer the first question, of how being tolerated differs from being accepted and rejected.

Table 1. Key outcome variable contrasts for effects in comparing participants’ experiences in the first experimental conditions of the within-subjects model (with two tolerance conditions).

	Experimental conditions
	Rejection	Tolerance A	Tolerance B	Acceptance
Manipulation check	3.37 (3.03)a	8.09 (1.92)b	8.21 (2.01)b	9.17 (1.88)c
Negative well-being	1.23 (.78)a	−.26 (.72)b	−.28 (.75)b	−.61 (.64)c
Future expectations	−.78 (.97)a	.07 (.92)b	−.05 (.89)b	.70 (.59)c
Future withdrawal	.46 (1.11)a	.03 (.94)b	−.04 (.91)b	−.43 (.87)c
Raising voice	3.06 (.91)a	1.68 (.96)b	1.50 (.84)b	1.45 (.74)b

Analyses on negative well-being, future expectations, and future withdrawal were performed on factor scores. Different subscripts signify significantly different mean scores across the rows (at p < .05; calculated with Tukey post-hoc correction).

Outcomes after the first experience

For the outcome measures, there were condition effects for negative well-being, F(3, 435) = 135.964, p < .001, η_p² = .484, future expectations, F(3, 435) = 53.339, p < .001, η_p² = .269, future withdrawal, F(3, 435) = 15.219, p < .001, η_p² = .095, and willingness to raise their voice, F(3, 435) = 81.002, p < .001, η_p² = .358. Using Tukey-corrected post-hoc analyses (see Table 1), we found that effects of being tolerated on negative well-being, future expectations, and future withdrawal all fall between the scores for being rejected and being accepted, which is consistent with previous research (Adelman et al., 2023; Cvetkovska et al., 2021) and our expectations. Additionally, and again consistent with our hypotheses, despite these negative effects of being tolerated, people were no more willing to complain and raise their voice after being tolerated as they were after being accepted: thus, there was no difference despite the fact that the negative effects of being tolerated (compared to being fully accepted) should give them more reason to complain than those in the acceptance condition who have no reason to complain. This suggests that being tolerated can prevent people from expressing their dissatisfaction with their majority teammates.

In sum, the results of analyses for the first experience in the within-subjects experimental design indicate that being tolerated is better than being rejected but worse than being accepted, and that being tolerated may prevent people from expressing dissatisfaction with their treatment.

Next, we looked into the second and third questions: How do experiences of having first been accepted or rejected affect later experiences of being tolerated, and how do experiences of having first been tolerated affect later experiences of being rejected or accepted?

Manipulation check within and across conditions

We first looked at the manipulation check items and used a mixed ANOVA analysis to test whether the effect of experimental condition (the experiences of tolerance, acceptance, or rejection) on responses to the manipulation check differed based on the order of the conditions (which experience came first and which came second). The analysis showed a significant interaction effect, F(3, 436) = 303.035, p < .001, η_p² = .676, which implies that participants’ manipulation check responses indicated that the experiences of being tolerated, accepted, or rejected depend on the context of what people experienced first.

We started to unpack this interaction by looking at whether the tolerance condition differed from the acceptance and rejection conditions in the expected direction among the two experiences that each participant had (within-subjects differences). We found significant within-subject differences for all four conditions (see Figure 1), with people showing a large decrease toward the “feeling rejected” side of the 11-point scale when experiencing rejection after tolerance, F(3, 436) = 402.896, p < .001, η_p² = .480, and a large increase toward the “feeling accepted” side of the scale when experiencing tolerance after rejection, F(3, 436) = 490.177, p < .001, η_p² = .529. Smaller differences emerged for participants who experienced a combination of tolerance followed by acceptance, with a slight increase toward the acceptance side of the scale when acceptance followed tolerance, F(3, 436) = 4.288, p = .039, η_p² = .010, and a slightly stronger decline away from the acceptance end of the scale when tolerance followed acceptance, F(3, 436) = 11.958, p = .001, η_p² = .027. As Figure 1 shows (comparing within condition cluster), the experience in the tolerance condition was consistently felt to be less as full acceptance compared to the acceptance condition, and much less as rejection compared to the rejection condition, providing further evidence that tolerance is a distinct experience (Cvetkovska et al., 2021).

Figure 1. Figure showing manipulation checks findings of whether an experience felt more like rejection (1), tolerance (6), or acceptance (9) on an 11-point scale. Acc. = acceptance, tol = tolerance, and rej = rejection. Higher scores indicate stronger feeling of being accepted. Error bars indicate standard errors. Full scale (1–11) not shown to improve figure accessibility.

Next, we took another approach to answering questions two and three. Using the manipulation check item, we examined whether being tolerated was experienced as less or more accepting depending on whether it was preceded by being accepted or being rejected, and vice versa. To access this, we compared feelings in the second condition that participants experienced, and we found significant effects of the preceding condition, F(3, 436) = 218.524, p < .001, η_p² = .601. Specifically, participants experiencing being tolerated after having first been rejected felt the most accepted, with those who experienced acceptance after tolerance and tolerance after acceptance feeling less accepted by comparison and not differing from each other. Rejection after tolerance was felt to be close to being fully rejected (see Figure 1, comparing across the four light gray bars).

Negative well-being

We took the same analytic approach with the different outcome variables. Participants’ self-reported negative well-being was affected by the order in which they experienced being accepted, tolerated, and rejected in the within-subjects conditions, F(3, 434) = 234.554, p < .001, η_p² = .619. To understand this interaction, we first looked within all four conditions, and found significant within-subject differences between first experiencing tolerance followed by rejection, F(3, 436) = 276.401, p < .001, η_p² = .389, rejection followed by tolerance, F(3, 436) = 409.825, p < .001, η_p² = .486, and acceptance followed by tolerance, F(3, 436) = 17.268, p < .001, η_p² = .038. No difference emerged for tolerance followed by acceptance, F(3, 436) = .168, p = .682, η_p² = .000. All significant effects were in the predicted directions, with the being tolerated condition leading to more negative well-being compared to the acceptance condition and less negative well-being compared to the rejection condition (see Figure 2).

Figure 2. Negative well-being by experimental condition and order of the experience (first/second). Higher scores indicate higher negative well-being: acc. = acceptance, tol = tolerance, and rej = rejection.

We then looked across first and second experiences for all four conditions. A significant effect across the first experiences (four dark gray bars), F(3, 436) = 135.695, p < .001, η_p² = .484, was found. The most positive well-being emerged in the acceptance condition, followed by the two tolerance conditions which did not differ from one another, and with the most negative well-being in the rejection condition. However, the significant effect across the second experiences (four light gray bars), F(3, 436) = 170.844, p < .001, η_p² = .541, indicated that the least negative well-being emerged when an experience of being tolerated followed a first experience of rejection, Further, rejection following an earlier experience of tolerance leads to the most negative well-being, and tolerance following acceptance and acceptance following tolerance falling between the two.

Interestingly, when comparing both across and within conditions, we see that whereas being tolerated following acceptance increased negative well-being, being tolerated following rejection decreased negative well-being, leading to two very different experiences of being tolerated depending on the first experience that one has. Furthermore, the experience of being accepted was more negative after rather than before experiencing being tolerated, and being rejected generated similar negative well-being whether it came before or after tolerance (see Figure 2).

Future treatment expectations and withdrawal behavior

Participants’ expectations of future treatment showed a similar pattern of effects. The significant mixed ANOVA interaction, F(3, 435) = 160.999, p < .001, η_p² = .526, indicated that significant differences emerged between first and second experiences and all in the expected direction (see Figure 3): being tolerated followed by rejection, F(3, 435) = 135.019, p < .001, η_p² = .237; rejection followed by tolerance, F(3, 435) = 250.710, p < .001, η_p² = .366; tolerance followed by acceptance, F(3, 435) = 19.446, p < .001, η_p² = .043; and acceptance followed by tolerance, F(3, 435) = 77.822, p < .001, η_p² = .152.

Figure 3. Future expectations by condition and order of the experience (first/second). Higher scores indicate more positive future expectations of treatment. Acc. = acceptance, tol = tolerance, and rej = rejection.

Looking across first experiences, F(3, 435) = 39.415, p < .001, η_p² = .270, we again found the most positive expectations in the acceptance condition, followed by the two tolerance conditions which did not differ from each other, and then the rejection condition. But when we look across the second experiences, F(3, 435) = 39.415, p < .001, η_p² = .270, we find that being tolerated after being rejected produces much more positive expectations than after being accepted. Furthermore, acceptance after (vs before) tolerance led to lower positive expectations, and rejection after (vs before) tolerance also led to lower positive experiences.

Similarly, the mixed ANOVA for participants’ expected withdrawal behavior in future interactions with their teammates was significant, F(3, 434) = 86.036, p < .001, η_p² = .373, with differences emerging between all first and second experiences, and all in the expected directions of most withdrawal in the rejection condition followed by the tolerance condition and with the least withdrawal in the acceptance condition: tolerance followed by rejection, F(3, 434) = 112.666, p < .001, η_p² = .206; rejection followed by tolerance, F(3, 434) = 114.260, p < .001, η_p² = .208; tolerance followed by acceptance, F(3, 434) = 13.297, p < .001, η_p² = .030; and acceptance followed by tolerance, F(3, 434) = 17.688, p < .001, η_p² = .039. Once again, when looking across first and second experiences, the first experiences (four dark gray bars, Figure 4) were consistent with previous studies, F(3, 434) = 15.539, p < .001, η_p² = .097, with the most withdrawal in the rejection condition, then equal withdrawal in the two tolerance conditions, and the least withdrawal in the acceptance condition (Adelman et al., 2023). However, when looking at the second experiences (four light gray bars, Figure 4), F(3, 434) = 66.784, p < .001, η_p² = .316, being tolerated after being rejected produced significantly less withdrawal than after being accepted. Additionally, acceptance after (vs before) tolerance led to somewhat higher withdrawal, and rejection after (vs before) tolerance also led to higher withdrawal.

Figure 4. Future withdrawal behavior by experimental condition and order (first/second). Higher scores indicate higher likelihood of withdrawal. Acc. = acceptance, tol = tolerance, and rej = rejection.

Raising voice

The results for willingness to raise one’s voice to express dissatisfaction with the majority teammates indicated that the experience of being tolerated produced a lower willingness. The mixed interaction effect was again significant, F(3, 436) = 190.309, p < .001, η_p² = .567, and the within-subjects simple effects indicated no significant difference in willingness to complain between the tolerance and acceptance conditions, regardless of whether tolerance, F(3, 436) = .603, p = .438, η_p² = .001, or acceptance, F(3, 436) = 1.712, p = .191, η_p² = .004, came first. However, those in the rejection condition were more willing to raise their voice, again regardless of whether it preceded, F(3, 436) = 267.284, p < .001, η_p² = .380, or followed, F(3, 436) = 303.606, p < .001, η_p² = .410, being tolerated. Comparing across first, F(3, 436) = 81.410, p < .001, η_p² = .359, and second, F(3, 436) = 88.774, p < .001, η_p² = .379, experiences, we found the same pattern of effects for both: people were equally unwilling to raise their voice in the acceptance and tolerance conditions, while those in the rejection condition were much more willing to complain (Figure 5).³

Figure 5. Likelihood of raising one’s voice to express dissatisfaction with one’s teammates by experimental condition and order (first/second experience). Higher score indicates a higher likelihood. Acc. = acceptance, tol = tolerance, and rej = rejection.

Discussion

Organizations, companies, and institutions increasingly comprise of people that differ in all sorts of ways, including deep-level differences in personalities, values, and working styles (Bell, 2007). Tolerance is considered an indispensable ingredient in the workplace (Von Bergen et al., 2012) and a neglected dimension in organizational diversity training (Gebert et al., 2017). Tolerance is considered important because people need to endure what they object to in order to let others express their alternative values, beliefs, and ways of doing things (Cohen, 2004; Verkuyten, 2023). However, tolerance also contains the implication of disapproval and dislike and it has been argued that the experience of being tolerated is inescapably patronizing and condescending, and therefore an inadequate substitute for the appreciation and full acceptance that minority members seek (Bergsieker et al., 2010; Verkuyten et al., 2020).

In trying to make a novel contribution and further our understanding of what it means to be tolerated, we delved more deeply into the question of how being tolerated is experienced as a function of prior experiences. In real life, experiences follow upon each other and thereby can influence each other, but this has not been considered in the existing research on the implications of being tolerated (Adelman et al., 2023; Cvetkovska et al., 2021). Specifically, being tolerated after first having experienced rejection might have a more positive meaning and impact than being tolerated after first having experienced full acceptance.

Using an experimental design and focusing on different working styles in virtual teams, we found evidence for the pre-registered pattern of effects we proposed for negative well-being, future expectations and withdrawal, and willingness to raise one’s voice.⁴ While the experience of being tolerated in a workgroup setting tended to fall in between being accepted and being rejected when measured first – as has been found in previous experimental research (Adelman et al., 2023; Cvetkovska et al., 2021) – that is no longer true when studied depending on what it follows. The findings demonstrate that being tolerated is experienced relatively more positively when it follows an initial experience of rejection. In fact, being tolerated after being rejected seems to be equally positive to being fully accepted, while tolerance following acceptance does not seem to have a comparable negative impact. Additionally, the findings further suggest that when being tolerated turns into rejection, the negative consequences of being rejected are enhanced, compared to rejection absent of prior tolerance. Thus, being begrudgingly included (being tolerated) was a rather negative experience for minority team members in comparison to full acceptance and it did not matter much whether one was first accepted. However, being tolerated was much better for minority members’ well-being and expectations of future treatment than being rejected, and even more so after first having experienced rejection. Thus, being tolerated had more positive implications than being rejected and especially after first experiencing rejection, but it fell short of being fully accepted. Furthermore, the results for raising one’s voice indicated that minority team members were no more willing to do so when being endured on the basis of their personal working style than those who were fully accepted. This suggests that despite the negative implications, being tolerated makes it difficult to speak out against the attitudinal disapproval and dislike that is implied in toleration: it seems more difficult to question someone’s attitude than their behavior.

The findings demonstrate that the experience of being tolerated is not necessarily a positive one: tolerance contains the message that one’s beliefs and practices are disapproved of and that one’s behavioral inclusion is dependent on the goodwill of the tolerator (Verkuyten et al., 2020). An important implication of the current research is that the differences between being accepted and being tolerated emerged from the simple awareness that one’s team members disapprove of their working style, but yet included them in the team tasks and did so equally in their ball passing behavior. This is important because while extensive research demonstrates the psychological effects of behavioral inclusion versus exclusion using the cyberball game (Hartgerink et al., 2015), part of our findings emerge from comparing participants who were identically treated during the game but either objected to (tolerance) or welcomed (acceptance) for their differences.

Furthermore, this finding is also relevant in light of the microaggression literature in which it is proposed that the actual intention or attitude behind the microaggression does not matter for the victim to feel microaggressed (Freeman & Stewart, 2021; Mekawi & Todd, 2021). However, the difference in attitude is exactly what matters for the distinction between being tolerated and being accepted. Specific for the experience of being tolerated is that the behavior is equally inclusive but the attitude is not. Being tolerated, therefore, differs from forms of microaggression and future research might consider to empirically examine these differences more systematically, and also differences between being tolerated and forms of microinclusion (Muragishi et al., 2023) and workplace incivility (Andersson & Pearson, 1999). Further, the experience of being tolerated differs from forms of rejection and discrimination in which victims are behaviorally excluded which has various negative psychological and social implications (Pascoe & Smart Richman, 2009; Schmitt et al., 2014). Some of the strongest experimental effects that we found were between being rejected and being tolerated which suggests that facing actual negative behavioral treatment is much more painful and problematic than knowing that others endure your disapproved off habits, views and beliefs without treating you differently.

Limitations and future research

It could be argued that being tolerated in the virtual team situation has negative implications because full acceptance of differences tends to be normative within these settings. The language of diversity and the valuing and appreciation of differences have become dominant in liberal societies and in most organizational and work contexts in these countries. Thus, being endured might have negative implications because it is considered rather inappropriate, which is indicated by the use of the term “mere tolerance.” Future research could investigate this interpretation, for example, by examining the implications of being tolerated in a context where appreciation and acceptance of differences is not normative (e.g., an authoritarian organizational context where hierarchy and inequality are normative).

Additionally, this interpretation also implies that the impact of being tolerated can be different in contexts or in relation to differences for which forbearance tolerance rather than appreciation is normative. For example, whereas people may expect to be fully accepted despite their different working style (e.g., person-oriented), someone who is less conscientious or who expresses a lack of motivation for the team’s objective might expect tolerance at best. Thus, there might be contexts and differences for which appreciation and full acceptance is less feasible or even not desirable, and this might matter for the implications that the experience of being tolerated has. Future research could examine whether tolerance is normative in a given context or not, and whether this distinction has a psychological and social impact on the person being tolerated.

Furthermore, we focused on the difference between people- and task-oriented working style for identifying someone as in the minority position, which is meaningful for a working context. However, people can also be tolerated for other differences that cannot be changed or learned, such as race, gender, disability, and sexual orientation. It might be that some of our findings are magnified for these sorts of characteristics that may constitute central aspects of minority identities. Being tolerated as an ethnic or sexual minority team member might have more negative implications and this could be examined in future research.

Implications and conclusions

A diverse, equal and inclusive setting depends on people’s willingness to allow others to express their own opinions, beliefs and personalities. It requires that people put up with things they disapprove of or even dislike, also with things that are antithetical and incompatible with their own convictions, commitments, and ways of doing things. In short, it requires tolerance as forbearance which is considered an indispensable component of managing diverse contexts (see Verkuyten, 2023), including in organizational and workplace settings (Gebert et al., 2017; Von Bergen et al., 2012). A focus on tolerance is important because it allows people to work together without having to discard or compromise on their own personalities and strongly held personal values and beliefs. However, we have demonstrated that there can be negative side effects to being tolerated, depending on the prior experiences that people have.

These possible effects should be acknowledged and considered in finding balanced and productive ways for dealing with forms of diversity in companies, organizations and institutions. Organizations and institutions which recognize the importance of tolerance but also pay attention to minority members’ feelings are more likely to reap the potential benefits that are associated with diversity (Ng & Stephenson, 2015; Van Knippenberg & Schippers, 2007). It clearly is important that minority individuals are treated equally, but it also matters whether they feel that their contributions are appreciated and valued to prevent, for example, a silencing effect on willingness to advocate for better treatment. Hence, an organization promoting tolerance may want to create advocacy structures to encourage engagement by those who are being tolerated.

Furthermore, this study uniquely identifies that tolerance does not exist in a vacuum, but that the experience of being tolerated will both be affected by and in turn also impact other workplace experiences that people have. Introducing a tolerance-based perspective is much more beneficial in a workplace with a culture of (micro)exclusions than in a setting in which appreciation and full acceptance predominates. However, tolerance is an indispensable ingredient in diverse workplaces and being tolerated a common experience (Gebert et al., 2017; Von Bergen et al., 2012). This makes it important to be aware of the possible negative consequences for the targets of tolerance and to consider methods of alleviating reductions in well-being, perceptions of future belonging and success, and willingness to engage with management or teammates to improve the workspace. Similarly, an experiment in appreciation and acceptance which ultimately fails may reduce the relatively positive experiences people might have in a tolerant environment.

Ethical approval

The research was ethically approved by the Ethics Committee of the Faculty of Social and Behavioral Sciences, Utrecht University.

Open scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered. The data and materials are openly accessible at https://osf.io/gr7kw.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The study has been preregistered with study design, hypotheses and measures: See Open Science https://osf.io/se2h7/. Additional documentation, full materials and measures, and data and syntax are also available: https://osf.io/gr7kw.

Additional information

Funding

This manuscript was supported by a European Research Council (ERC) Advanced Grant under the European Union’s Horizon 2020 research and innovation program [grant agreement No 740788].

Notes on contributors

Levi Adelman

Levi Adelman is a Post Doc researcher at Ercomer, Faculty of Social and Behavioral Sciences at Utrecht University, the Netherlands.

Maykel Verkuyten

Maykel Verkuyten is a full professor at Ercomer, Faculty of Social and Behavioral Sciences at Utrecht University, the Netherlands.

Kumar Yogeeswaran

Kumar Yogeeswaran is an associated professor at the School of Psychology, Speech and Hearing at the University of Canterbury in Christchurch, New Zealand.

Notes

1. See Open Science Framework for pre-registration of study design, hypotheses and measures: https://osf.io/se2h7/. Additional documentation, full materials and measures, and data and syntax are also available: https://osf.io/jakzm/.

2. Thirty-three participants failed an attention check question for at least one of the two conditions they were placed in (which asked them to correctly recall how often they had the ball shared with them as part of the cyberball exercise). While results were slightly strengthened by their exclusion, we opted to retain the full sample for the results reported (see below for more details on the exclusion measure).

3. Additionally, results from the prisoner’s dilemma trust items show a small effect of order on trust, F(3, 436) = 26.993, p < .001, η_p² = .157, which is driven by lower trust in the rejection condition compared to the tolerance condition within-subjects, regardless of whether rejection, F(3, 436) = 44.098, p < .001, η_p² = .092, or tolerance, F(3, 436) = 33.421, p < .001, η_p² = .071, came first. Similar findings also emerged when looking across second experiences, F(3, 436) = 18.820, p < .001, η_p² = .115, with rejection generating less trust than acceptance or tolerance. When looking at first experiences, F(3, 436) = 6.980, p < .001, η_p² = .046, however, the results were less clear, with less trust in the rejection condition compared to acceptance, but not to either tolerance condition, although less trust emerged in one of the tolerance conditions compared to acceptance.

4. Additional documentation, full materials and measures, and data and syntax are available on OSF: https://osf.io/jakzm/