CoChat
Sign Up
PreprintCited by 0Gender Diversity and InequalityJob Satisfaction and Organizational BehaviorCareer Development and Diversity

Included but undermined: How representation goals backfire

Open access
Osborne, Merrick Robinson, Roberson, Jacob W., Tost, Leigh·2025
DOI 10.31234/osf.io/cse5r_v2
Open this paper in CoChat — free accountOpen PDF

Abstract

In recent years, an increasingly common practice amongst organizations has been setting representation goals, or organizational mandates intended to increase the prevalence of underrepresented employees in visible work assignments. The work of meeting these goals often falls to managers, who must also weigh additional criteria – like the employees’ workloads or skillsets – in their task assignments. Typically, when faced with a task assignment decision, managers will refrain from assigning employees who are not available or who do not have the requisite skillset. However, leaning on Tetlock’s theory of value pluralism, we predict that these goals can lead managers to set aside traditional criteria of effective task assignment (specifically, underrepresented employees’ person-job fit and/or workloads) to prioritize achieving these organizational goals. In this way, underrepresented employees are more likely to be assigned to tasks that are sub-optimal for their occupational success (i.e., by being assigned to work that does not appropriately match their skillset or that leads them to be overworked). We examine these predictions in four studies (N = 2,598). We find that this decision-making is driven by managers’ increased diversity concerns (as a consequence of the representation goals), but can be mitigated by how the organization communicates those goals. We conclude by discussing both the theoretical and practical implications of our findings.

AI snapshot

Written from the paper’s full text.

TL;DR

Across four experiments (N = 2,598), representation goals led decision-makers to assign underrepresented employees to visible tasks even when those employees were already overworked or lacked person-job fit. This effect was driven by increased prioritization of diversity concerns and was reduced when leaders used balanced rather than singular endorsement language.

Key findings

  • Representation goals increased assignment of women despite overload and/or low fit: Study 1 participants were 2.25 times more likely to assign a woman (95% CI [1.74, 2.93]); Study 2 participants were 4.56 times more likely to choose a woman and 4.14 times more likely to assign the overworked woman Samantha; Study 3 women were 7.00 times more likely to choose the woman candidate.
  • The effect extended to low person-job fit candidates: Study 2 participants were 2.36 times more likely to select the low-fit woman Anna; Study 4 participants were 13.15 times more likely to select the Black candidate when representation goals were present.
  • Mediation by diversity concerns was supported across studies: e.g., Study 1 indirect effects via Likert diversity considerations = .35, SE = .06, 95% CI [.23, .47]; Study 4 indirect effect via Likert measure = .20, SE = .03, 95% CI [.15, .26].
  • Balanced endorsement language reduced the effect: Study 4 participants in the singular endorsement condition were 24.84 times more likely to select the Black candidate under representation goals, versus 4.37 times in the balanced endorsement condition; the interaction was significant (b = −1.74, SE = .67, p = .008).

Methods

  • Four preregistered vignette experiments with employed participants recruited via Prolific Academic and ConnectResearch; final samples were Study 1 N = 985, Study 2 N = 347, Study 3 N = 737 women, Study 4 N = 529 (total N = 2,598 after exclusions).
  • Participants made hypothetical task-assignment decisions under representation-goal versus control conditions; Study 2 crossed initiative with gender of the low person-job fit candidate, Study 3 crossed initiative with self-versus-other assignment, and Study 4 crossed representation goals with balanced versus singular endorsement language in a race context.
  • Main analyses used hierarchical logistic regression for binary assignment outcomes and PROCESS mediation/moderation models to test diversity considerations as mediator and endorsement language as moderator.

Limitations

  • The paper states that women and employees of color may accept short-term extra work if it brings promotion or skill benefits, that lack of fit may build longer-term skills, and that focusing only on workload, fit, and diversity goals may miss other productivity considerations; it also calls for future work on actual time/cognitive/emotional costs, decision-maker feelings, other diversity domains, and manager identity.
  • Evident from the text: all studies rely on hypothetical scenarios and self-reported intentions rather than observed workplace assignments, and samples come from online panels; Study 2 had many participants with fewer than five years of work experience (45.7% before exclusions), and Study 4 included part-time employees.
  • Some analyses were exploratory or not preregistered, and Study 2 had a programming error affecting eight participants; the paper reports supplemental analyses excluding those participants and states results did not change.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.31234/osf.io/cse5r_v2

Cites

0 citation errors · 0 not checkable

Statistics

68 tests recomputed · 1 with the reported and recomputed p on opposite sides of .05

Claims the paper makes

0 of 42 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    In Study 1, using the ranking variable, the indirect effect of representation goals on assigning a woman through diversity prioritization did not cross zero (indirect effect = .66, SE = .09, 95% CI [.49, .86]).

    “Using the ranking variable, we found that the confidence interval of the indirect effect also did not cross zero (indirect effect = .66,SE = .09,95% C.I.: [.49, .86]).”
    Quote in the paperSubstantial claim
  • Findingown data

    In Study 2, there was a significant indirect effect of representation goals on selecting Samantha, mediated by prioritizing diversity concerns (indirect effect = 1.34, SE = .24, 95% CI [.98, 1.91]).

    “We found a significant indirect effect of representation goals on selecting Samantha, mediated by prioritizing diversity concerns (indirect effect = 1.34,SE = .24,95% C.I.: [.98, 1.91]).”
    Quote in the paperSubstantial claim
  • Findingown data

    In Study 1, participants in the representation-goals condition were more likely to assign at least one woman candidate than participants in the control condition (B = .81, SE = .13, χ2(1) = 37.24, p < .001; OR = 2.25, 95% CI [1.74, 2.93]).

    “We found a significant effect of the representation-goals condition (B = .81,SE = .13,χ2 (1) = 37.24, p < .001)such that participants in the representation-goals condition were 2.25 times more likely (95% C.I.: [1.74, 2.93]) to assign a woman candidate than participants in the control condition.”
    Quote in the paperSubstantial claim
  • Findingown data

    In Study 1, the effect of representation goals on assigning a woman was mediated by prioritization of diversity concerns, using the Likert measure (indirect effect = .35, SE = .06, 95% CI [.23, .47]).

    “Using the Likert scale, the confidence interval did not cross zero (indirect effect = .35,SE = .06,95% C.I.: [.23, .47]),suggesting that the effect of our manipulation on assigning a woman was mediated by concerns about prioritizing diversity.”
    Quote in the paperSubstantial claim
  • Findingown data

    In Study 1, using the binary ranking variable, the indirect effect of representation goals on assigning a woman through diversity prioritization did not cross zero (indirect effect = .06, SE = .09, 95% CI [.05, .08]).

    “With the binary variable, we found that the confidence interval of the indirect effect did not cross zero (indirect effect = .06,SE = .09,95% C.I.: [.05, .08])foot_0 .”
    Quote in the paperSubstantial claim
  • May combine two claims. 21
  • Classified as probably not a substantial claim. 27
  • The quote only partly states this claim. 3

  • May combine two claims.

    Findingown data

    In Study 2, there was no effect of the gender of the unqualified candidate on assigning Samantha, and the interaction term was not significant (p = .62 and p = .76, respectively).

    “There was no effect of the gender of the unqualified candidate (p = .62)and the interaction term was not significant (p = .76).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    For all considerations except workload, there were no interactions between the manipulations on participants' considerations.

    “Additionally, for all considerations (except for workload), there were not any interactions between the manipulations on participant's considerations (all p's > .181).”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    Among participants with decision-making power, those in the representation-goals condition had stronger diversity considerations (M = 5.10, SD = 1.82) than those in the control condition (M = 4.32, SD = 2.01; t(376.81)= 3.94, d = .40).

    “Participants in the representation-goals condition had stronger diversity considerations (M = 5.10, SD = 1.82) than participants in the control condition (M = 4.32, SD = 2.01; t(376.81)= 3.94, d = .40).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There was a significant interaction between endorsement language condition and representation goals condition on selecting a Black candidate (b = -1.74, SE = .67, χ2(1) = 6.82, p = .008), such that the effect of representation goals was larger in the singular endorsement language condition than in the balanced endorsement language condition.

    “The interaction term was significant (b = -1.74,SE = .67,χ2 (1) = 6.82, p = .008),such that the effect of representation goals was larger in the singular endorsement language condition than in the balanced endorsement language condition (Hypothesis 3).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In Study 2, in the first step of hierarchical logistic regression, unqualified woman vs. unqualified man significantly predicted selection of the unqualified candidate (B = .66, SE = .19, χ2(1) = 12.06, p < .001).

    “In the first step of our hierarchical logistic regression, both of our independent variables significantly predicted the selection of the unqualified candidate (representation goals vs. control: B = .62,SE = .19,χ2(2) = 11.30,p < .001;unqualified woman vs. unqualified man: B = .66,SE = .19,χ2(1) = 12.06, p < .001).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    In the assign-self condition, there was not a significant difference in workload considerations between the strategic marketing and diversity conditions.

    “For workload considerations, there was not a significant difference in the assign-self condition in strategic marketing (M = 5.95, SD = 1.48) or diversity conditions (M = 5.86, SD = 1.30; d = .07,95% CI:[-0.13,.28]).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Tetlock's value pluralism model theorizes and empirically demonstrates that when people encounter value-based tradeoffs, they prioritize the value they believe is most important to the key parties they seek to satisfy.

    “Tetlock (1986) theorizes and empirically demonstrates that when people encounter value-based tradeoffs, they will prioritize the value they believe is most important to the key parties they seek to satisfy.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Tetlock (1986)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Managers typically prioritize or are knowledgeable of employees' skillsets and workload capacities when making assignment decisions.

    “Underrepresented employees' propensity to receive assignments that do not align with their training or workload capacity is puzzling, since managers typically prioritize (or at least, are knowledgeable of) employee's skillsets (Kristof-Brown 2000, Buckingham 2005, Sekiguchi and Huber 2011) and workload capacities (Boucher et al. 2024) when making assignment decisions.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Kristof-Brown 2000

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Buckingham 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Sekiguchi and Huber 2011

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Study 3 was pre-registered on OSF.

    “The study was pre-registered on OSF (https://osf.io/8gs9v/?view_only=5a94cba4dc6d4e568d18727a4c991a52).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The studies were approved under IRB Protocol UP-18-00638.

    “Our studies were approved under IRB Protocol: UP-18-00638.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Underrepresented employees would likely find visible assignments beneficial because they grant access to positions with more status and influence.

    “Underrepresented employees, in turn, would likely find these visible assignments beneficial since they are being granted access to positions with more status and influence (e.g., Merton 1968, Ragins et al. 1998, Dato et al. 2021).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Merton 1968

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Ragins et al. 1998

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Dato et al. 2021

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Implementing representation goals signals to managers that increasing the prevalence of underrepresented people in visible positions is important to leadership and that managers should balance whether the employee's identity aligns with the representation goal.

    “In turn, implementing representation goals signals to managers within the organization that increasing the prevalence of underrepresented people in visible positions is important to leadership (Ostroff and Bowen 2000, Ashforth and Anand 2003, Bowen and Ostroff 2004, Barsky 2008, Nishii et al. 2017, Leslie 2019); and that managers should balance an additional consideration when making task-assign”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Ostroff and Bowen 2000

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Ashforth and Anand 2003

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Bowen and Ostroff 2004

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Research has documented negative consequences when an employee's training and the knowledge requisite for the task are misaligned, including lower performance evaluations and decreased job satisfaction.

    “Indeed, research has documented a wide range of negative consequences when the employee's training and the knowledge requisite for the task are misaligned (Kristof-Brown et al. 2005): outcomes include lower performance evaluations (Sekiguchi and Huber 2011) and decreased job satisfaction (Kim et al. 2020).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Kristof-Brown et al. 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Sekiguchi and Huber 2011

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Kim et al. 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    In Study 2, 132 participants who failed one or more comprehension checks and/or the attention check were excluded, leaving a final sample of 347 participants.

    “As per our pre-registration, we filtered out 132 participants who failed one or more of our two comprehension checks and/or our single attention check.This left us with a final sample of 347 participants (53.6% female, 1.2% non-binary; Mage = 28.50,SDage= 9.93; Mwork experience = 8.63, SDwork experience = 8.94).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Decision-makers are likely to note that assigning someone to work for which they lack the requisite training puts the employee's success and the task's effective completion in jeopardy and could affect employees' perceptions of their manager's competence.

    “Moreover, decision-makers are also likely to note that assigning someone to work for which they lack the requisite training puts the employee's success and the task's effective completion in jeopardy (Cable andJudge 1996, Kim et al. 2020) and could have implications for employees' perceptions of their manager's competence.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Cable andJudge 1996

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Kim et al. 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Person-job fit is important not only for employees' job satisfaction but also for their overall performance and work engagement.

    “Abundant research demonstrates the importance of fit, not only for employees' job satisfaction but also for their overall performance and work engagement (Kristof-Brown et al. 2005).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kristof-Brown et al. 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Two of the most important task assignment parameters are workload concerns and person-job fit.

    “Two of the most important task assignment parameters are workload concerns (i.e., who has adequate time available to take on the additional assignment?) and person-job fit (i.e., who has the qualifications, skills, experience, or expertise to do the task effectively?Cable andJudge 1996, Cable andDeRue 2002).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Cable andJudge 1996

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Cable andDeRue 2002

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Interpretationown data

    Representation goals from organizational leadership increase the likelihood of assigning a task to a woman who does not have ideal person-job fit, but do not increase the likelihood of assigning a task to a man with low person-job fit.

    “Moreover, we found that, while top-down representation goals do increase the likelihood of assigning a task to a woman who does not have the ideal person-job fit (as predicted in Hypothesis 2), representation goals from organizational leadership do not increase the likelihood of assigning a task to a man with low person-job fit.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Underrepresented employees are less likely to have the requisite time or may not have the appropriate competency for the representation-goals position.

    “Further, these underrepresented employees are less likely to have the requisite time (e.g., see existing academic conversations about the”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Gewin 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Boykin et al. 2020

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Hamilton 2024

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Assigning a worker too many tasks can diminish the time and attention they give to their work and consequently harm their performance.

    “Workload considerations are important, as assigning a worker too many tasks can diminish the time and attention they give to their work and consequently harm their performance (Greenglass et al. 2001, Zeytinoglu et al. 2007, Tan and Netessine 2014, Fan and Smith 2017).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Greenglass et al. 2001

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Zeytinoglu et al. 2007

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Tan and Netessine 2014

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    In Study 1, 26 participants who failed pre-registered attention checks were excluded, leaving a final sample of 985 participants.

    “We filtered out 26 participants who failed our pre-registered attention checks, which left us with a final sample of 985 participants (38.4% female, 0.6% other/non-binary, Mage = 35.92,SDage = 9.77).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Study 1 was pre-registered on OSF.

    “The study was pre-registered on OSF (https://osf.io/2vjr4/?view_only=9704047423d349848313c86504c253dc).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Study 2 was pre-registered on OSF.

    “The study was pre-registered on OSF (https://osf.io/sda4c/?view_only=446a7d0421a2459da8783c2d2bd46bc4).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    These assignments often require training that employees do not have, or require them to take on additional tasks beyond those they were originally hired to complete.

    “Indeed, these assignments often require training that the employees do not have, or requires them to take on additional tasks to those which they were originally hired to complete (West 2018, Kern 2018, Jackson 2019, Hodgson 2019, Layne et al. 2023, Hill 2023).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites West 2018

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Kern 2018

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Jackson 2019

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Study design and conductnone given

    Study 3 was pre-registered.

    “Pre-registration is available here (https://aspredicted.org/WRR_QTP).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Study 3 results are ostensibly driven by participants' beliefs that they would not be adequately compensated for being assigned work that they do not have the capacity or qualifications to take on.

    “Ostensibly, our results in Study 3 are driven by participants' beliefs that they would not be adequately compensated for being assigned work that they do not have the capacity or qualifications to take on.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Hypotheses 1b and 2b were supported in both the full sample and in the subsample of only participants who commonly make these types of decisions.

    “Thus, Hypotheses 1b and 2b were supported in both the full sample and in the subsample of only participants who commonly make these types of decisions.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Study 3 used a pre-registered exclusion protocol and filtered out 236 participants who failed one or more of two comprehension checks and/or a single attention check.

    “As per our pre-registration, we filtered out 236 participants who failed one or more of our two comprehension checks and/or our single attention check.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    In Supplemental Study 1, 61 participants who failed one or more comprehension checks and/or the attention check were excluded, leaving a final sample of 140.

    “We filtered out 61 participants who failed one or more of our two comprehension checks and/or our single attention check, leaving us with a final sample of 140 (50.7%female, 3.6% non-binary; Mage = 32.09,SDage = 11.30).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationown data

    Women generally would prefer not to be assigned to tasks or committees that would overburden their workload, but if they are to be assigned additional work, they would prefer that which they have more of a vested interest in.

    “These findings suggest that, while women generally would prefer not to be assigned to tasks or committees that would overburden their workload, if they are to be assigned additional work, they would prefer that which they have more of vested interest in.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: These assignments are not desirable enough for women to voluntarily overload themselves with work to take on these assignments.

  • Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

    Study design and conductown data

    The final sample for Study 3 consisted of 737 women participants after exclusions.

    “This left us with a final sample of 737 participants (Mage = 35.85,SDage= 13.17; Mwork experience = 14.13,SDwork experience = 12.11).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S1 (asserted).

    Premisecitation

    Leslie (2019) predicts that representation goals might diminish intrinsic motivation to advance the cause of diversity.

    “Given the prediction in Leslie (2019) that representation goals might diminish intrinsic motivation to advance the cause of diversity, we wondered if our manipulation of representation goals might decrease participants' reported support for gender-diversity efforts.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Leslie (2019)

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    In Study 3, after applying pre-registered exclusions, 441 participants were retained.

    “After applying our pre-registered exclusions, we retained 441 participants (44.22% women; 5.89% Black, 71.21% White; Mage = 37.56, SDage = 9.68).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationnone given

    The studies focus on a specific situation: when a company is looking to improve the representation of underrepresented employees in visible positions, but these employees either do not have the requisite expertise for the positions being assigned, or accepting the position would ultimately overload their present workload capacity.

    “Our studies focus on a specific situation: when a company is looking to improve the representation of underrepresented employees in visible positions, but these employees either do not have the requisite expertise for the positions being assigned, or accepting the position would ultimately overload their present workload capacity.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    In Study 2, eight participants experienced a programming error as they proceeded through the study.

    “As indicated in the footnote to Study 2, eight participants experienced a programming error as they proceeded through the study.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Many diversity efforts fail to benefit underrepresented groups and can instead harm the intended beneficiaries or lead to other unintended consequences.

    “This is in support of prior research which has found that, despite the best intentions of organizational leaders, many diversity efforts fail to benefit underrepresented groups and can instead harm the intended beneficiaries or lead to other unintended consequences (Kalev et al. 2006, Plaut et al. 2011, Leslie et al. 2014, Dobbin and Kalev 2016, Cardador 2017, Leslie 2019).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Kalev et al. 2006

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Plaut et al. 2011

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Leslie et al. 2014

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Interpretationreasoning

    Assigning the woman candidate to the committee would lead her to have too many work hours, whereas the men could be assigned with hours to spare.

    “Consequently, it was clear that assigning the woman to the committee would lead her to have too many work hours (whereas the men could be assigned with hours to spare).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Study 4 was pre-registered on AsPredicted.

    “The study was pre-registered on AsPredicted (https://aspredicted.org/6FB_8SC).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The signal from a balanced endorsement of diversity goals enables managers to make more appropriate task assignments that sufficiently fulfill the tasks' need, grant fair and more desirable opportunities for underrepresented employees, and satisfy organizational leadership's top-down goals.

    “The resulting signal from a balanced endorsement of diversity goals enables managers to make more appropriate task assignments that sufficiently fulfill the tasks' need, grant fair and more desirable opportunities for underrepresented employees, and satisfy organizational leadership's topdown goals.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The final sample for Study 4 consisted of 529 participants after excluding those who did not pass the attention check.

    “After applying exclusions, we retained 529 participants (45.94% women; 3.02% Black, 77.88% White; Mage = 37.55, SDage = 10.36).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    In Study S2, 239 participants who failed comprehension checks were excluded, leaving a final sample of 611.

    “We filtered out 239 participants who failed our comprehension checks, which left us with a final sample of 611 participants (46.38% female, 1.0% other/non-binary, Mage = 38.98,SDage = 12.81).”
    Quote in the paperNot a substantial claim
41 of 41

  • The paper relies on this without support.

    Setting representation goals is an increasingly common organizational practice.

    “In recent years, an increasingly common practice amongst organizations has been setting representation goals, or organizational mandates intended to increase the prevalence of underrepresented employees in visible work assignments.”

  • The paper relies on this without support.

    Representation goals are frequently implemented to correct underrepresentation, so decision-makers may have few underrepresented employees in their candidate pools.

    “However, representation goals are frequently implemented to specifically help organizations correct underrepresentation: which means that decision-makers may not have many underrepresented employees in their candidate pools.”

  • The paper relies on this without support.

    Expectations regarding task assignment criteria other than representation goals, such as workload and fit, remain implicit.

    “while expectations regarding other task assignment criteria (i.e., workload and fit) are likely to remain implicit.”

  • The authors state this as an assumption.

    These assignments are not desirable enough for women to voluntarily overload themselves with work to take on these assignments.

    “This observation provides support for our underlying assumption that these assignments are not desirable enough for women to voluntarily overload themselves with work to take on these assignments.”

  • The authors state this as an assumption.

    Women view these outcomes negatively.

    “We also found evidence in support of our underlying assumption that women view these outcomes negatively, in that the effects of representation goals on task assignment were diminished when the only woman available was oneself.”
5 of 5

Statistics that don’t add up

Each reported p-value recomputed from its test statistic and degrees of freedom.

  • Inconsistent p-value across alpha = .05: t(344) = -0.24,p < .001

    Major

    The paper reports p < .001 here; recomputing from the test statistic and its degrees of freedom gives p = .810 (two-tailed). The two fall on opposite sides of alpha = .05. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p < .001 · recomputed p = .810

  • Inconsistent p-value: χ2 (2) =13.47, p < .001

    Minor

    The paper reports p < .001 here; recomputing from the test statistic and its degrees of freedom gives p = .001. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p < .001 · recomputed p = .001

  • p-value fits an unstated one-tailed test: t(25.33)= -2.48,p < .01

    Minor

    The paper reports p < .01 here; recomputing from the test statistic and its degrees of freedom gives p = .020 (two-tailed). The reported value is what a one-tailed test gives, but the paper does not say near this result that a one-tailed test was used.

    Reported p < .01 · recomputed p = .020

  • Inconsistent p-value: F(1, 605) = 8.14, p = .005

    Minor

    The paper reports p = .005 here; recomputing from the test statistic and its degrees of freedom gives p = .004. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p = .005 · recomputed p = .004

  • Inconsistent p-value: t(138) = 2.05, p = .038

    Minor

    The paper reports p = .038 here; recomputing from the test statistic and its degrees of freedom gives p = .042 (two-tailed). The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p = .038 · recomputed p = .042

  • Inconsistent p-value: χ2(2) = 11.30,p < .001

    Minor

    The paper reports p < .001 here; recomputing from the test statistic and its degrees of freedom gives p = .004. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p < .001 · recomputed p = .004

  • Inconsistent p-value: t(344) = 2.23, p < .001

    Minor

    The paper reports p < .001 here; recomputing from the test statistic and its degrees of freedom gives p = .026 (two-tailed). The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p < .001 · recomputed p = .026

  • Inconsistent p-value: χ2 (1) = 6.82, p = .008

    Minor

    The paper reports p = .008 here; recomputing from the test statistic and its degrees of freedom gives p = .009. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p = .008 · recomputed p = .009

8 of 8

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Doesn’t add upt(344) = -0.24reported p < .001 · recomputed p = .810The two fall on opposite sides of .05

    “t(344) = -0.24,p < .001”

  • Doesn’t add upχ²(2) = 13.47reported p < .001 · recomputed p = .001

    “χ2 (2) =13.47, p < .001”

  • Doesn’t add upχ²(1) = 6.82reported p = .008 · recomputed p = .009

    “χ2 (1) = 6.82, p = .008”

  • Doesn’t add upt(344) = 2.23reported p < .001 · recomputed p = .026

    “t(344) = 2.23, p < .001”

  • Doesn’t add upχ²(2) = 11.3reported p < .001 · recomputed p = .004

    “χ2(2) = 11.30,p < .001”

  • Doesn’t add upF(1, 605) = 8.14reported p = .005 · recomputed p = .004

    “F(1, 605) = 8.14, p = .005”

  • Doesn’t add upt(138) = 2.05reported p = .038 · recomputed p = .042

    “t(138) = 2.05, p = .038”

  • Doesn’t add upt(25.33) = -2.48reported p < .01 · recomputed p = .020

    “t(25.33)= -2.48,p < .01”

  • Not confirmedχ²(3) = 0.61reported p = .435 · recomputed p = .894

    “χ2 (3) =.61, p = .435”

  • Not confirmedt(734) = -1.92reported p = .054 · recomputed p = .055

    “t(734) = -1.92,p = .054”

  • Consistentχ²(1) = 37.24reported p < .001 · recomputed p < .001

    “χ2 (1) = 37.24, p < .001”

  • Consistentχ²(2) = 90.4reported p < .001 · recomputed p < .001

    “χ2 (2) = 90.40,p < .001”

  • Consistentχ²(1) = 33.41reported p < .001 · recomputed p < .001

    “χ2 (1) = 33.41,p < .001”

  • Consistentχ²(1) = 12.74reported p < .001 · recomputed p < .001

    “χ2 (1) = 12.74, p < .001”

  • Consistentχ²(2) = 114.88reported p < .001 · recomputed p < .001

    “χ2(2) = 114.88,p < .001”

  • Consistentχ²(2) = 36.2reported p < .001 · recomputed p < .001

    “χ2(2) = 36.20,p < .001”

  • Consistentχ²(3) = 18.94reported p < .001 · recomputed p < .001

    “χ2(3) = 18.94, p < .001”

  • Consistentχ²(1) = 57.68reported p < .001 · recomputed p < .001

    “χ2(1) = 57.68,p < .001”

  • Consistentχ²(1) = 23.72reported p < .001 · recomputed p < .001

    “χ2 (1) = 23.72,p < .001”

  • Consistentχ²(1) = 9.55reported p = .002 · recomputed p = .002

    “χ2(1) = 9.55, p = .002”

  • Consistentt(983) = 6.01reported p < .001 · recomputed p < .001

    “t(983) = 6.01, p < .001”

  • Consistentt(983) = -8.31reported p < .001 · recomputed p < .001

    “t(983) = -8.31,p < .001”

  • Consistentt(983) = 6.26reported p < .001 · recomputed p < .001

    “t(983) = 6.26, p < .001”

  • Consistentt(983) = -0.25reported p = .81 · recomputed p = .803

    “t(983) = -0.25,p = .81”

  • Consistentt(983) = 4.97reported p < .001 · recomputed p < .001

    “t(983) = 4.97, p < .001”

  • Consistentt(983) = 5.28reported p < .001 · recomputed p < .001

    “t(983) = 5.28, p < .001”

  • Consistentt(983) = -5.53reported p < .001 · recomputed p < .001

    “t(983) = -5.53,p < .001”

  • Consistentt(983) = -5.47reported p < .001 · recomputed p < .001

    “t(983) = -5.47,p < .001”

  • Consistentt(983) = -3.99reported p < .001 · recomputed p < .001

    “t(983) = -3.99,p < .001”

  • Consistentt(983) = -5.93reported p < .001 · recomputed p < .001

    “t(983) = -5.93,p < .001”

  • Consistentt(983) = 9.24reported p < .001 · recomputed p < .001

    “t(983) = 9.24, p < .001”

  • Consistentt(983) = 7.31reported p < .001 · recomputed p < .001

    “t(983) = 7.31, p < .001”

  • Consistentt(983) = 2.67reported p = .008 · recomputed p = .008

    “t(983) = 2.67, p = .008”

  • Consistentt(983) = 0.39reported p = .7 · recomputed p = .697

    “t(983) = 0.39, p = .70”

  • Consistentt(983) = 2.9reported p = .004 · recomputed p = .004

    “t(983) = 2.90, p = .004”

  • Consistentt(983) = 3.21reported p = .001 · recomputed p = .001

    “t(983) = 3.21, p = .001”

  • Consistentt(983) = -1.04reported p = .3 · recomputed p = .299

    “t(983) = -1.04,p = .30”

  • Consistentt(983) = 0.36reported p = .72 · recomputed p = .719

    “t(983) = 0.36, p = .72”

  • Consistentt(983) = 1.82reported p = .07 · recomputed p = .069

    “t(983) = 1.82, p = .07”

  • Consistentt(344) = 14.84reported p < .001 · recomputed p < .001

    “t(344) = 14.84, p < .001”

  • Consistentt(344) = 0.26reported p = .793 · recomputed p = .795

    “t(344) = .26,p = .793”

  • Consistentt(344) = 1.09reported p = .278 · recomputed p = .276

    “t(344) = 1.09, p = .278”

  • Consistentt(344) = -0.5reported p = .618 · recomputed p = .617

    “t(344) = -0.50,p = .618”

  • Consistentχ²(1) = 12.06reported p < .001 · recomputed p < .001

    “χ2(1) = 12.06, p < .001”

  • Consistentχ²(1) = 3.51reported p = .06 · recomputed p = .061

    “χ2(1) = 3.51, p = .06”

  • Consistentχ²(1) = 33.41reported p < .001 · recomputed p < .001

    “χ2 (1, N = 347) = 33.41,p < .001”

  • ConsistentF(1, 605) = 9.57reported p = .002 · recomputed p = .002

    “F(1, 605) = 9.57, p = .002”

  • ConsistentF(1, 605) = 133.83reported p < .001 · recomputed p < .001

    “F(1, 605) = 133.83,p < .001”

  • Consistentt(138) = 6.21reported p < .001 · recomputed p < .001

    “t(138) = 6.21, p < .001”

  • Consistentt(25.58) = -1.34reported p = .193 · recomputed p = .192

    “t(25.58)= -1.34,p = .193”

  • Consistentt(24.74) = 3.06reported p = .005 · recomputed p = .005

    “t(24.74)= 3.06, p = .005”

  • Consistentt(24.97) = -1.42reported p = .168 · recomputed p = .168

    “t(24.97)= -1.42,p = .168”

  • Consistentt(24.83) = 2.55reported p = .017 · recomputed p = .017

    “t (24.83) = 2.55, p = .017”

  • Consistentχ²(1) = 61.21reported p < .001 · recomputed p < .001

    “χ2(1) = 61.21,p < .001”

  • Consistentχ²(1) = 61.29reported p < .001 · recomputed p < .001

    “χ2(1) = 61.29,p < .001”

  • Consistentt(734) = 17.51reported p < .001 · recomputed p < .001

    “t(734) = 17.51, p < .001”

  • Consistentt(734) = 7.99reported p < .001 · recomputed p < .001

    “t(734) = 7.99, p < .001”

  • Consistentt(734) = 5.32reported p < .001 · recomputed p < .001

    “t(734) = 5.32, p < .001”

  • Consistentt(734) = -2.02reported p = .044 · recomputed p = .044

    “t(734) = -2.02,p = .044”

  • Consistentt(734) = -1.59reported p = .113 · recomputed p = .112

    “t(734) = -1.59,p = .113”

  • Consistentt(735) = 2.3reported p = .02 · recomputed p = .022

    “t(735) = 2.30, p = .02”

  • Consistentt(735) = 3.25reported p = .001 · recomputed p = .001

    “t(735) = 3.25, p = .001”

  • Consistentt(735) = 6.8reported p < .001 · recomputed p < .001

    “t(735) = 6.80, p < .001”

  • Consistentχ²(1) = 42.56reported p < .001 · recomputed p < .001

    “χ 2 (1, N = 611) = 42.56,p < .001”

  • Consistentt(306.47) = 8.18reported p < .001 · recomputed p < .001

    “t(306.47)= 8.18, p < .001”

  • Consistentt(350.31) = 7.92reported p < .001 · recomputed p < .001

    “t(350.31)= 7.92, p < .001”

  • Consistentt(359.5) = 6.69reported p < .001 · recomputed p < .001

    “t(359.50)= 6.69, p < .001”

  • Consistentt(429.26) = 0.49reported p = .626 · recomputed p = .624

    “t(429.26)= .49,p = .626”

68 of 68

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Study 4 diversity-consideration means differ between results and table

    Major

    The supplemental results give means of 4.10 and 3.19 for the strong and balanced representation-goals conditions, respectively; Table S7 gives 4.59 and 2.96 for those conditions.

  • Study 2 table reports impossible means for a binary outcome

    Minor

    The paper codes selecting a woman as either 0 or 1, but Table 2 reports means above 1 for that variable.

  • Study 4 supplemental analysis has more observations than the retained sample

    Minor

    The supplemental Study 4 analysis reports denominator degrees of freedom of 605, despite the Study 4 methods stating that only 529 participants were retained.

  • Study 4 binary-mediator index differs from its conditional indirect effects

    Minor

    For the reported first-stage moderated-mediation model, the index should equal the balanced-condition indirect effect minus the singular-condition indirect effect, as it does for the other two mediators. The binary-mediator effects give 0.01 − 0.230 = −0.220, not −0.08.

    Reported -0.08 · recomputed -0.22

  • Study 3 assigns the overwork prediction to the wrong hypothesis

    Minor

    The Study 3 description attributes the prediction about overworking a woman to Hypothesis 2a, whereas the stated Hypothesis 1a concerns workload and Hypothesis 2a concerns qualifications and fit.

  • Study 2 supplemental workload t statistic disagrees with coefficient and SE

    Minor

    Dividing the reported workload coefficient by its standard error gives −5.5, not the reported t statistic of −0.24.

    Reported -0.24 · recomputed -5.5

6 of 6

Cites

The publisher hasn’t deposited this paper’s reference list.

Research works printed but not linked: 26

  1. Ashforth BE, Anand V (2003) The normalization of corruption in organizations. Research in organizational behavior: An annual series of analytical essays and critical reviews, Vol 25 Research in organizational behavior. (Elsevier Science Ltd, Oxford, England), 1-52.
  2. Edwards J (1991) Person-job fit: A conceptual integration, literature review, and methodological critique.
  3. Cooper CL, Robertson IT, eds. International review of industrial and organizational psychology. (John Wiley & Sons, Ltd), 283-357.
  4. Heilman ME, Welle B (2006) Disadvantaged by Diversity? The Effects of Diversity Goals on Competence Perceptions1. Journal of Applied Social Psychology 36(5):1291-1319.
  5. Kelman HC, Hamilton VL (1989) Crimes of Obedience: Toward a Social Psychology of Authority and Responsibility (Yale University Press).
  6. Padilla AM (1994) Research news and Comment: Ethnic Minority Scholars; Research, and Mentoring: Current and Future Issues. Educational Researcher 23(4):24-27.
  7. Plaut VC, Garnett FG, Buffardi LE, Sanchez-Burks J (2011) "What about me?" Perceptions of exclusion and Whites' reactions to multiculturalism. Journal of Personality and Social Psychology 101(2):337-353.
  8. Ragins BR, Townsend B, Mattis M (1998) Gender gap in the executive suite: CEOs and female executives report on breaking the glass ceiling. AMP 12(1):28-42.
  9. Rokeach M (1973) The nature of human values (Free Press, New York, NY, US).
  10. Schneider B, Brief AP, Guzzo RA (1996) Creating a climate and culture for sustainable organizational change. Organizational Dynamics 24(4):7-19.
  11. Sekiguchi T (2004) Person-organization fit and person-job fit in employee selection: A review of the literature. Osaka keidai ronshu 54(6):179-196.
  12. Sekiguchi T, Huber VL (2011) The use of person-organization fit and person-job fit information in making selection decisions. Organizational Behavior and Human Decision Processes 116(2):203-216.
  13. Sherf EN, Tangirala S, Weber KC (2017) It Is Not My Place! Psychological Standing and Men's Voice and Participation in Gender-Parity Initiatives. Organization Science 28(2):193-210.
  14. Tan TF, Netessine S (2014) When does the devil make work? An empirical study of the impact of workload on worker productivity. Management Science 60(6):1574-1593.
  15. Tetlock PE (1986) A value pluralism model of ideological reasoning. JPSP 50(4):819-827.
  16. Thai M, Lizzio-Wilson M, Selvanathan HP (2021) Public perceptions of prejudice research: The double- edged sword faced by marginalized group researchers. Journal of Experimental Social Psychology 96:104181.
  17. Uhl-Bien M, Riggio RE, Lowe KB, Carsten MK (2014) Followership theory: A review and research agenda. The Leadership Quarterly 25(1):83-104.
  18. Arnaud A (2010) Conceptualizing and Measuring Ethical Work Climate: Development and Validation of the Ethical Climate Index. Bus. Soc. 49(2):345-358.
  19. Cantor N, Mischel W (1977) Traits as prototypes: Effects on recognition memory. J. Pers. Soc. Psychol. 35(1):38.
  20. Cantor N, Mischel W (1979) Prototypes in person perception. Adv. Exp. Soc. Psychol. (Elsevier), 3-52.
  21. Gruenfeld DH, Inesi ME, Magee JC, Galinsky AD (2008) Power and the objectification of social targets. J. Pers. Soc. Psychol. 95(1):111.
  22. Gutek BA, Cohen AG, Tsui A (1996) Reactions to Perceived Sex Discrimination. Hum. Relat. 49(6):791-813.
  23. Kristof-Brown AL, Zimmerman RD, Johnson EC (2005) Consequences of Individuals' Fit at Work: A Meta-Analysis of Person-Job, Person-Organization, Person-Group, and Person-Supervisor Fit. Pers. Psychol. 58(2):281-342.
  24. Leslie LM (2019) Diversity Initiative Effectiveness: A Typological Theory of Unintended Consequences. Acad. Manage. Rev. 44(3):538-563.
  25. Ostroff C, Zhan Y (2012) Person-Environment Fit in the Selection Process. Oxf. Handb. Pers. Assess. Sel. (Oxford University Press).
  26. Werbel JD, Gilliland SW (1999) Person-environment fit in the selection process. Res. Hum. Resour. Manag. Vol 17 (Elsevier Science/JAI Press, US), 209-243.

Cited by

Still being checked.

Related

Authors
Osborne, M. R., Roberson, J. W., & Tost, L.
Affiliations
  • Cornell University (US)
  • University of Southern California (US)
ORCID
Osborne, Merrick Robinson
Publisher
Center for Open Science
Published
2025-08-31
DOI
10.31234/osf.io/cse5r_v2
OpenAlex
W4413864927
Cited by
0
References
0
Type
posted content
Language
en
Open access
Licence
Topics
  • Social Sciences › Gender Studies › Gender Diversity and Inequality
  • Social Sciences › Business, Management and Accounting › Organizational Behavior and Human Resource Management › Job Satisfaction and Organizational Behavior
  • Social Sciences › Safety Research › Career Development and Diversity
Keywords
organizational mandate, underrepresented employees, task allocation, value pluralism, person-job fit, workload, goal communication, management decision-making

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-11-04.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-11-04.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review