People overestimate the self-presentation costs of deadline extension requests

Across five studies (N = 4151), we investigate a novel barrier that prevents people from making personally beneficial requests: the overestimation of self-presentation costs. Even when deadlines are easily adjustable, people are less likely to request an extension and submit lower quality work when perceived self-presentation costs are higher—such as when the request is visible to a supervisor (Study 1a). Specifically, people are less likely to request an extension when they are concerned with appearing incompetent (Study 2). Yet, other people do not negatively respond to deadline extension requests (Study 1b). Attesting to the importance of selfpresentation concerns in shaping extension request behaviors, formal policies that reduce self-presentation concerns increase requests in both online (Study 3) and in-person (Study 4) settings. These findings highlight a novel psychological barrier that prevents people from requesting resources that could increase their performance and more effectively manage their deadlines. Deadlines can help people make progress on complex, multistage projects by improving productivity and time-management (McGrath, 1991). Deadlines are frequently assigned by other people. Indeed, most work deadlines are assigned by a direct supervisor to motivate employees to overcome procrastination and assist teams with coordinating their tasks (Aeon & Aguinis, 2017; Ariely & Wertenbroch, 2002; Gevers, van Eerde, & Rutte, 2009; Zhu, Yang, & Hsee, 2018). Yet deadline estimates are often overly optimistic. On tasks ranging from mundane lab tasks (Buehler, Griffin, & Ross, 1994) to public infrastructure work (Flyvbjerg, Holm, & Buhl, 2002; Hall, 1982), both experts and novices underestimate how long tasks will take to complete. Thus, assigned deadlines can be stressfully tight, compromising output quality (Amabile, DeJong, & Lepper, 1976; Karau & Kelly, 1992) and creating negative work experiences due to feelings of time-pressure (Perlow, 1999). One potential way of maximizing the benefits of an assigned deadline while minimizing the costs of miscalibration is to set a deadline and adjust it as needed by proactively requesting an extension. 1. Deadlines and deadline adjustment Deadlines vary in their urgency. Some deadlines require a fast turnaround which can cause feelings of time-pressure; other deadlines may be less urgent (Ballard, Vancouver, & Neal, 2018). Deadlines also vary in whether they can be adjusted. Some assigned deadlines are strict. In these cases, once an original deadline has passed, taking any action related to the task is impossible or costly. However, many everyday work tasks are subordinate tasks that involve smaller actions that must be completed to achieve a larger goal (Cropanzano, James, & Citera, 1993). Such tasks are less likely to incur adjustment costs. A supervisor may ask an employee to submit an initial draft of a proposal for an event happening next month by the end of this week, to provide time to edit and revise the proposal before the final deadline. Proactively adjusting the deadline for this initial draft is less costly than extending a deadline for a final version of the proposal for an event happening later in the week. In an initial pilot study, working adults reported that more than half of their task deadlines (53.3%) were adjustable (Supplemental Study A). Thus, proactively requesting an extension on an adjustable deadline could be a useful strategy to gain time, reduce stress, and improve work quality for people facing tight yet adjustable deadlines. In the current research, we isolate a novel psychological barrier—self-presentation concerns—that prevents people from requesting extensions on adjustable deadlines. In doing so, we shed new light on the psychology behind ☆ This paper has been recommended for acceptance by Dr Vanessa Bohns. * Corresponding author. E-mail address: awhillans@hbs.edu (A.V. Whillans).

Deadlines can help people make progress on complex, multistage projects by improving productivity and time-management (McGrath, 1991). Deadlines are frequently assigned by other people. Indeed, most work deadlines are assigned by a direct supervisor to motivate employees to overcome procrastination and assist teams with coordinating their tasks (Aeon & Aguinis, 2017;Ariely & Wertenbroch, 2002;Gevers, van Eerde, & Rutte, 2009;Zhu, Yang, & Hsee, 2018). Yet deadline estimates are often overly optimistic. On tasks ranging from mundane lab tasks (Buehler, Griffin, & Ross, 1994) to public infrastructure work (Flyvbjerg, Holm, & Buhl, 2002;Hall, 1982), both experts and novices underestimate how long tasks will take to complete. Thus, assigned deadlines can be stressfully tight, compromising output quality (Amabile, DeJong, & Lepper, 1976;Karau & Kelly, 1992) and creating negative work experiences due to feelings of time-pressure (Perlow, 1999). One potential way of maximizing the benefits of an assigned deadline while minimizing the costs of miscalibration is to set a deadline and adjust it as needed by proactively requesting an extension.

Deadlines and deadline adjustment
Deadlines vary in their urgency. Some deadlines require a fast turnaround which can cause feelings of time-pressure; other deadlines may be less urgent (Ballard, Vancouver, & Neal, 2018). Deadlines also vary in whether they can be adjusted. Some assigned deadlines are strict. In these cases, once an original deadline has passed, taking any action related to the task is impossible or costly. However, many everyday work tasks are subordinate tasks that involve smaller actions that must be completed to achieve a larger goal (Cropanzano, James, & Citera, 1993). Such tasks are less likely to incur adjustment costs. A supervisor may ask an employee to submit an initial draft of a proposal for an event happening next month by the end of this week, to provide time to edit and revise the proposal before the final deadline. Proactively adjusting the deadline for this initial draft is less costly than extending a deadline for a final version of the proposal for an event happening later in the week. In an initial pilot study, working adults reported that more than half of their task deadlines (53.3%) were adjustable (Supplemental Study A). Thus, proactively requesting an extension on an adjustable deadline could be a useful strategy to gain time, reduce stress, and improve work quality for people facing tight yet adjustable deadlines. In the current research, we isolate a novel psychological barrier-self-presentation concerns-that prevents people from requesting extensions on adjustable deadlines. In doing so, we shed new light on the psychology behind ☆ This paper has been recommended for acceptance by Dr Vanessa Bohns. why people often fail to request personally useful resources.

Request making and self-presentation concerns
People often find it difficult to make personally beneficial requests. Across diverse contexts ranging from asking for directions (Flynn & Lake, 2008) to asking for a raise (Small, Gelfand, Babcock, & Gettman, 2007), people frequently fail to make requests for resources that they want and need. People avoid making requests because they fear that their requests will be a burden (Bohns & Flynn, 2010) or that their requests will be rejected (Bohns, 2016). Relatedly, researchers have speculated-but have not formally tested-the idea that people fail to request help because of the desire to appear competent (Bohns & Flynn, 2015;Lee, 1997). To date, little is known about how competence concerns shape people's willingness to request personally beneficial resources.
Prior research has focused on understanding when people are likely to make requests for help or information-two requests that involve the direct effort of the requestee. Unlike these requests, when people consider requesting an extension for an adjustable deadline on an independent task, they should not expect to rely on the direct effort of the requestee. Concerns about burdening other people with the request should be less relevant to decision-making. Thus, by focusing on adjustable deadline extension requests, we can identify the unique role of self-presentation concerns in shaping people's willingness to make personally beneficial requests.
Research suggests that people closely attend to how their actions impact other people's impressions (Goffman, 1959). The desire to maintain a positive impression can both encourage and deter people from taking certain actions. People boast about their success as opposed to being humble because they want to appear competent (Sezer, Gino, & Norton, 2018). They also go out of their way to help other people because they want to appear likable (Hui, Lam, & Law, 2000). The desire to appear competent may be a particularly salient goal when requesting deadline extensions at work. This is because we generally believe that other people have a high degree of control over how they use their time each day (Donnelly, Wilson, Whillans, & Norton, 2021). Furthermore, the efficient use of time is highly valued as a signal of competence, ability, and social status in the workplace (Keinan, Bellezza, & Paharia, 2019;Whillans, Giurge, & West, 2020) and time management skills are recognized as one of the single most important skills of top performers (Maxfield, Grenny, Hale, & Hoffman, 2017). Performance speed is also considered a critical component of workplace competence (Paul & Anantharaman, 2003). Following from this research, we examine whether self-presentation concerns-and concerns about competence in particular-prevent employees from requesting extensions on adjustable deadlines that would likely help their performance. 1 While we test various perceptions that might prevent people from requesting extensions on adjustable deadlines at work-including concerns about burdening managers and team members-we predict that competence concerns will be a central predictor of whether employees feel comfortable requesting extensions on adjustable deadlines.

Hypothesis 1. (H1):
Employees will avoid requesting a deadline extension, even when they recognize that the deadline is adjustable, and even when an extension could improve the quality of their work, out of the belief that the request will make others see them as incompetent.

Overestimating the self-presentational costs of requesting an extension
Why will employees overestimate the competence costs of asking for more time on adjustable deadlines? People often inaccurately predict how observers will judge them for their actions. In conversations, people underestimate how much their partners will like them (Boothby, Cooney, Sandstrom, & Clark, 2018). People also anticipate that observers will judge them as more incompetent for gaffes, mistakes, and personal shortcomings than they actually do (Savitsky, Epley, & Gilovich, 2001).
People frequently overestimate the negative impact of their actions because they fail to consider other factors that might affect observers' impressions. Observers may be distracted or overwhelmed by cues other than an actor's blunder (Savitsky et al., 2001). They are also overly focused on their own thoughts and actions (Bohns & Flynn, 2015;Gilovich, Medvec, & Savitsky, 2000;Savitsky et al., 2001). Observers' own experiences may further prompt them to empathize with the actor's perspective, yet actors often fail to recognize this possibility (Epley, Savitsky, & Gilovich, 2002). Building on this research, we propose that employees are likely to overestimate how harshly they will be judged for requesting an extension from their supervisors on an adjustable deadline at work.

Hypothesis 2. (H2):
People who request a deadline extension will overestimate how incompetent they will appear to the requestee. We will also test whether employees' overestimation is partially explained by people underestimating how much their manager cares about the quality of their task performance vs. the speed of their work. This investigation follows from research showing that when people face resource scarcity, they demonstrate greater recall for information that is relevant to the focal task and neglect other features of the task that could help them perform the task more efficiently (Zhao & Tomm, 2017). Applying this logic to the context of deadlines, when people are working under a tight deadline, they are likely to focus more on deadline relevant information ("How fast am I performing this task?") as opposed to other information that managers might care about more such as task quality. This heightened concern with speed while working under deadlines could lead employees to worry too much about their managers' perceptions of task speed and not enough about their managers' perceptions of the quality of their work. This misperception could therefore also predict employees' reluctance to ask for more time on adjustable deadlines.

Reducing ambiguity to mitigate competence concerns
Hypothesis 3. (H3): Introducing formal policies will mitigate selfpresentation concerns. When formal policies exist that normalize deadline extension requests, people will experience lower selfpresentation concerns and be more likely to request a deadline extension.
When there are no clear norms to guide our behaviors, we turn to the immediate environment to know how to act (Bandura, 1986;Blanchard, Crandall, Brigham, & Vaughn, 1994;Cialdini & Trost, 1998). In the absence of formal policies, people are likely to feel uncertain about whether they can ask for more time on adjustable deadlines and avoid doing so-especially because extension requests lower employees' perceived competence. Providing further evidence for our mechanism of interest, we propose that the introduction of formal policies should lower perceived competence concerns and increase deadline extension requests.
Taken together, we propose the following conceptual model to explain why people fail to request extensions on adjustable and relatively costless deadlines in work contexts (Fig. 1). 1 In this paper, we focus primarily on the employee and manager relationship because people are especially attuned to impression management concerns in the context of workplace interactions that have the potential to influence their standing or financial compensation. Indeed, in an influential review paper, Jones and Pittman (1982) outlined several impressions that employees strive to maintain-among the top of this list included competence impressions.

Overview of the current research
Across five studies (N = 4151) we examine why people who find themselves working under challenging yet adjustable deadlines under employ deadline extension requests. We examine the psychological barriers of making such requests in a context where people commonly experience the need to request an extension: working under a demanding task deadline that a supervisor or superior has assigned. We focus primarily on workplace interactions, given the importance of competence concerns in organizational settings. We focus on adjustable deadlines, a context where other well-studied factors that deter people from making requests-such as the fear of rejection or objective costs for the manager or team-are mitigated. In doing so, our research identifies the central role of self-presentation concerns in preventing people from asking for resources that could increase task performance and reduce stress. We also experimentally manipulate competence concerns through the provision of a formal policy. In doing so, we provide direct evidence that reducing self-presentation concerns can increase extension requests.
In Study 1a, we provide an initial test of the role of self-presentation concerns by examining whether employees are less likely to request an extension when the request is visible to a supervisor. In Study 1b, we examine whether employees' concerns are well-founded by testing whether supervisors judge employees who request a deadline extension as less competent. In Study 2, we extend these findings in a study of actual managers and employees and show that competence concerns and not other concerns like perceptions of burden undermine extension requests. We also explore the role of speed vs. quality concerns in predicting employees' willingness to make extension requests. In Study 3, we provide employees with a formal policy that is designed to reduce self-presentation concerns and increase intentions to make a deadline extension. In Study 4, we replicate these findings in a consequential field study.

Studies 1a & 1b: self-presentation concerns and deadline extension requests
In Studies 1a and 1b, we test our conceptual model by examining whether self-presentation (i.e. competence) concerns prevent employees from requesting extensions on explicitly adjustable deadlines. In Study 1a, employees face a difficult task and receive an explicit chance to request an extension. In Study 1b, supervisors rate employees from Study 1a.

Study 1a: self-presentation concern and avoidance of deadline extension requests
In Study 1a, we tested whether employees were less likely to proactively request an extension when self-presentation concerns were high-i.e., when the request was visible to their supervisor-as compared to when extension requests were not visible and selfpresentation concerns were low. We purposefully created a work task that required high output in a short period of time, such that a deadline extension request could improve employees' objective performance. This study was preregistered (http://aspredicted.org/blind.php? x=dc5he4). 2

Participants and procedure
We recruited 901 adults through MTurk to complete a work task for a fictional publishing company in exchange for $0.50 and the chance to earn a $0.30 performance bonus. Seventy-six participants did not meet our preregistered inclusion criteria, leaving a final sample size of 825 adults (54.7% female; M age = 36.51, SD age = 12.36). This final sample allowed us to detect a minimum difference of 0.55 in a logistic regression with 80% power (see Supplementary Material, Section A for a table summarizing sample characteristics across studies).
As part of this task, participants were told to imagine that they worked for a fictional publishing company. Participants were asked to describe a complex image that depicted various events for two minutes, with the goal of describing as many events in as much detail as possible. During the task, on the top of the screen, a 2-min timer counted down and informed participants how much time they had remaining to complete the task. We pre-tested this task to ensure that it induced timepressure among participants, such that an extension was desirable.
All participants were then provided with the opportunity to request a deadline extension. At any time during the two-minute task, participants could click a button labeled "I need more time." If the participant clicked this button, they were immediately granted an additional minute to work on the task (i.e., an additional minute was added to the timer). After the deadline had passed, using the identical extension request procedure, participants were once again asked whether they wanted an additional minute to complete the task. 3 To isolate self-presentation concerns, and to eliminate previously studied barriers to requesting extensions such as the fear of non-compliance (Bohns, 2016), we designed the request process to be simple and frictionless.
All participants were told that another participant would act as their supervisor and would assess their competence based on their performance. To ensure that all participants were motivated to make a positive impression on their supervisor, as they would in an actual workplace, participants were told that their final payment would be determined by their supervisor's evaluations. Employees were told that if managers rated them as a 6 or 7 on a 7-point scale on perceived competence, they would receive a $0.30 bonus payment. In reality, all participants received a $0.30 bonus payment regardless of their performance.
In the high self-presentation concern condition, participants were told that if they had requested a deadline extension that their supervisor would be notified of the request. The employee was informed that their manager would not see the exact wording of this request-instead the manager would view a note on top of the evaluation form that read 'Note: the employee asked for more time to work on the task.' Thus, both the employee and manager received the identical extension request information. See Appendix A. In the low self-presentation concern condition, participants were told that their supervisor would not be notified of their decision to request a deadline extension.
To evaluate whether receiving a deadline extension increased performance, we measured the total character length of the description as well as the total number of events described, as counted by two research assistants who were blind to the experimental conditions.

Deadline extension request
Providing evidence that the extension was desirable, across conditions, most people asked for an extension before or after the deadline had passed (60.2%). Consistent with our pre-registration and conceptual model, in Study 1a & 1b we focused our analyses on people who asked for more time before the deadline had passed (27.6%).
Consistent with H1, when the extension request was visible and selfpresentation concerns were high, participants were less likely to ask for an extension (22.6%) than when the request was invisible and selfpresentation costs were low (

Mediation analysis
Next, we conducted mediation analyses (Preacher & Hayes, 2008 with 5000 bootstrapping resamples) to test whether requesting a deadline extension explained the effect of condition assignment (1 = high self-presentation concern) on the two objective performance measures: description length and number of topics reported. Consistent with our conceptual model, requesting a deadline extension explained the effect of condition assignment on description length, B = 1.94, SE = 0.64, 95% CI = [0.74, 3.24] as well as the effect of condition assignment on the number of topics reported, B = 0.13, SE = 0.05, 95% CI = [0.06, 0.24].

Discussion
Study 1a provides initial evidence that self-presentation concerns prevent employees from requesting a helpful extension, even under a tight deadline that is explicitly adjustable. Although requesting an extension was desirable-overall 60.2% of participants requested more time to work on the task-simply changing the visibility of the request to a supervisor resulted in a significant decrease in the number of people who proactively made extension requests. However, participants who requested a deadline extension submitted longer and more detailed descriptions. This study supports our proposition that even when employees desire an extension while working under tight and adjustable deadlines, they are hesitant to make this request to their superior and consequently forgo the chance to improve their task performance. In Study 1b, we examined whether employees' self-presentation concerns were well-founded by assessing how supervisors evaluate employees as a function of whether they requested a deadline extension.

Study 1b: supervisors do not judge employees who request deadline extensions as less competent
In Study 1b, we tested whether supervisors' judgments of employee competence were negatively impacted by the employee's pre-emptive extension request.

Participants and procedure
We recruited 1415 adults (54.42% female; M age = 37.62, SD = 11.70) through MTurk. Participants imagined that they worked as a supervisor at the fictional company described in Study 1a. Supervisors were told that they had recently assigned their employees to describe as many events in an image with as much detail as possible in a 2-min period (i.e., the task that was assigned to participants in Study 1a). After reading example descriptions to ensure that they were familiar with task expectations, each supervisor evaluated the work of one employee from Study 1a. The employee that the supervisor reviewed was selected through random assignment.
The supervisor was provided with (a) the one image description task submitted by the randomly selected employee, (b) information about whether the employee requested a deadline extension before the deadline had passed, and (c) a third-party rating by a coder blind to the research hypotheses of how detailed the image description was (Fig. 2). All supervisors were provided with detailed performance information to prevent supervisors from assuming that any employee who requested a deadline extension should have performed better due to the additional time, which could have positively biased their evaluation (Chinander & Schweitzer, 2003). Supervisors were exposed to the image description task from one participant in Study 1a who either requested a deadline extension (extension request made) or did not (extension request not made). Regardless of what the participant did in Study 1a, we varied whether the supervisor was told that the target employee asked for more time (told extension requested); or did not receive any information about requesting a deadline extension (not told extension requested). Thus, this study followed a 2 (supervisor was told that the employee requested a deadline extension: yes vs. no) x 2 (employee requested a deadline extension: yes vs. no) between-subjects design where the average quality of the descriptions was held constant across conditions. This design allowed us to detect a minimum effect of η 2 = 0.005 in a two-way ANOVA with 80% power.

Supervisor evaluation
Supervisors evaluated the employee's competence by responding to the question "Compared to the average employee, how competent do you think the individual who created the image description above is?" using a scale ranging from 1 (much less than the average employee) to 7 (much more than the average employee).

Competence
Consistent with our conceptual model, the data from this study suggest that employees overestimated the negative consequences of extension requests on perceptions of competence. Supervisors rated employees as similarly competent regardless of whether they learned that the employee had requested an extension (M told = 4.21, SD = 1.50; M not told = 4.21, SD = 1.49), F(1, 1414) = 0.000004, p = 1.00, η 2 = 0.00001. Consistent with the results of Study 1a, which showed that extension requests increased task performance, supervisors rated employees who requested an extension as significantly more competent compared to those who did not (M request made = 4.68, SD = 1.31; M request not made = 4.11, SD = 1.51), F(1, 1410) = 31.52, p < .001, η 2 = 0.022.
There was no interaction between whether supervisors were told that employees asked for an extension and whether employees actually asked for an extension in predicting managers' perceptions of employees'

Discussion
Study 1b provides evidence that supervisor evaluations of competence are not negatively impacted by extension requests on adjustable deadlines. In fact, employees who requested an extension in Study 1a (who created longer and richer descriptions) were rated as more competent by their supervisors, highlighting the benefits of requesting extensions on adjustable deadlines.
While Studies 1a & 1b offer evidence that employees overestimate supervisors' negative judgements, these studies are not without limitations. Critically for our conceptual model, we did not explicitly test the role of self-presentation concerns. In Study 1b, we evaluated manager predictions of competence; however, we did not provide direct evidence that visible requests reduced employees' rates of making extension A.V. Whillans et al. requests through self-presentation concerns. Study 1a & 1b also relied on tightly controlled online lab paradigms, which do not clearly reflect the real-time experience of asking for more time on adjustable deadlines in workplace settings.
To provide direct evidence for the proposed self-presentation mechanism, we conducted three additional studies. In Study 2, we isolated the unique role of competence concerns in a sample of actual employees and managers. In Study 3, we tested self-presentation concerns as the mechanism of interest by attempting to lower these concerns via the provision of a formal policy that allowed for deadline extension requests. We tested whether this policy increased proactive deadline extension requests and explored the moderating role of individual differences in self-presentation concerns. In Study 4, we replicated the effectiveness of formal policies in a real-world classroom setting where students had to request an extension from their instructor.

Study 2: employees under-request deadline extensions
To isolate the role of self-presentation-and in particular, competence concerns-we conducted an ecologically valid scenario-based study with a sample of actual supervisors and employees.

Participants and procedure
We targeted 200 supervisors and 200 employees who worked at least 21 h per week at a part-time or full-time job outside the home. We recruited 200 employees and 203 supervisors (47.0% female; M age = 37.58, SD = 11.06; Employees = 45.9% female, M age = 36.89, SD = 10.59; Supervisors = 48.0% female, M age = 38.28, SD = 11.50). This study was pre-registered https://aspredicted.org/blind.php? x=/LTB_QKP. Supervisors had to work with at least three direct reports in their primary jobs; employees had to have at least one direct manager. To increase the ecological validity of our design, participants imagined having an interaction with a direct report (supervisor) that they currently worked with. Supervisors imagined assigning one of their direct reports (who was randomly selected among the three direct reports that they had reported on) to complete a task that was due the next day, and the direct report requested a deadline extension. Employees imagined that one of their supervisors assigned them to complete a task that was due the next day, and that they had requested a deadline extension.
All participants then completed six statements about perceived competence. Employees answered questions such as "When I requested an extension, my manager judged me as less competent" on a scale ranging from − 3 (Strongly Disagree) to +3 (Strongly Agree); (α = 0.95). Supervisors answered the identical six questions from their perspective such as "When this employee requested an extension, I judged them as less competent" (α = 0.95).
Employees and supervisors then "imagined that [they] were in this scenario in [their] actual life" and completed four statements about their intentions to ask for-or provide-a deadline extension. Employees answered questions such as "In this scenario, I would plan to ask my supervisor for an extension" (α = 0.94) and supervisors answered questions such as "In this scenario, I would plan to give my employee an extension (α = 0.96) on a scale ranging from 1 (Strongly Disagree) to 7 (Strongly Agree). These items were adapted from Grant (2012).
To isolate competence concerns as the conceptual mechanism underpinning extension request avoidance among employees, we also tested whether employees hesitated to ask for more time because they overestimated the burden of the request (see Whillans, Lee-Yoon, & Dunn, 2020). Specifically, employees reported the extent to which they felt that the request would 1) let down their team, 2) let down their manager, 3) burden their team and 4) burden their manager (α = 0.92). Supervisors answered the identical questions from their perspective (α = 0.94).
Finally, we explored whether employees hesitated to ask for more time because they underestimated how much their managers cared about quality vs. work speed. Employees answered the following two statements: "In this scenario, my direct manager would care more that I completed my work quickly" and "In this scenario, my direct manager would care more that I completed high quality work." (α = 0.74). Supervisors rated the identical items from their perspective: "As a manager in this scenario, I would care more that my direct report completed their work quickly" and "As a manager in this scenario, I would care more that my direct report completed high quality work" from − 3 (Strongly Disagree) to 3 (Strongly Agree) (α = 0.69). 4 Lastly, both employees and supervisors answered two items about whether quality vs. speed objectively mattered more for the task to rule out the alternative explanation that any observed perceptual differences were driven by role-specific beliefs about speed vs. quality.

Competence concerns
Following our pre-registered analysis plan, we analyzed each perception item separately. Consistent with our a priori predictions, employees reported greater concerns that extension requests signaled incompetence as compared to supervisors (M Employee = 0.16, SD = 1.63 vs. M Supervisor = − 0.25, SD = 1.63), t(400) = 2.51, p = .013, 95% CI [0.09, 0.73], d = 0.25. Consistent with our conceptual model, using Bonferroni corrections to adjust for multiple comparisons, concerns about perceived competence and capabilities were the only statistically significant between-condition differences. See Table 1 for condition differences across measures using an adjusted alpha level of p = .006 to account for multiple comparisons.

Extension request intentions
Consistent with our pre-registered predictions, employees reported lower intentions to ask for more time as compared to managers' intentions to provide an extension in this scenario (M Employee = 4.49, SD = 1.73 vs. M Supervisor = 5.04, SD = 1.50), t(400) = 3.41, p = .001, 95% CI [− 0.86, − 0.23], d = 0.33. These results provide further evidence that employees underestimated managers' willingness to grant more time on adjustable deadlines. We used Bonferroni correction which resulted in an adjusted alpha of p = .006.

Competence predicting intentions
Next, we examined whether competence concerns predicted employees' intentions to ask for more time in this scenario. Consistent with our conceptual model and pre-registration plan, path analysis using bootstrap estimation revealed a significant partial mediation. Employees' competence concerns partially explained their lower intentions to ask for more time relative to managers' willingness to grant employees more time from β = 0.17, p = .001 to β = 0.13, p = .006, IE = − 0.13 (0.06), 95% CI IDE [− 0.25, − 0.02].

Quality vs. speed
Employees overestimated the extent to which managers cared that their direct reports completed the proposed work quickly (M Employee

Competence, quality & speed predicting intentions
To examine the predictive nature of competence concerns, we entered competence, speed, and quality concerns into a mediation model to predict employees' intentions to ask for more time to work on the task if faced with this scenario in their daily lives. In this model, employees' competence concerns still partially explained employees' lower intentions to ask for more

Study 2 Discussion
Study 2 provides evidence that employees request deadline extensions less often than would be acceptable by managers, and that this tendency is associated with the belief that extension requests signal incompetence. These findings also provide confirmatory evidence that employees overestimate the degree to which supervisors perceive extension requests as a signal of incompetence which-together with an underestimation of managers' quality concerns-undermines employees' willingness to ask for more time on adjustable deadlines at work. In Study 3, we directly manipulate competence concerns by providing a formal policy for granting extensions and test whether these policies increase the willingness to request an extension.

Study 3: Removing Self-Presentation Concerns to Increase Extension Requests
In Study 3, we attempted to lower self-presentation concerns through the provision of a formal policy and studied the role of reduced competence concerns on the willingness to ask for more time. We used the identical procedure from Study 1a to ensure that the task that employees were assigned to complete required high output in a short period of time. To directly test the conceptual link between formal policies, impression management, and deadline extension request behavior, we measured employees' general proclivity to worry about other people's impressions of their behavior. This study was preregistered (https://asp redicted.org/r3cu9.pdf).

Participants and procedure
We recruited 1419 working adults through MTurk to complete the identical work task from Study 1a for $0.50 and the chance to earn a $0.30 bonus based on performance (59.0% female; M age = 39.22, SD age = 12.41). We recruited more people than our pre-registered stopping point of 1350 due to uncertainty over the number of possible exclusions. Our final sample allowed us to detect a small main effect in a logistic regression (OR of 1.50) with 80% power.
The protocol of this study was identical to Study 1a. All participants were given the chance to request a deadline extension by clicking a button labeled "I need more time." In the high self-presentation concern condition, participants were told that if they requested a deadline extension their supervisor would be notified of this request with the following note on the top of their evaluation form: "Note: the employee asked for more time to work on the task." In the low self-presentation concern condition, participants were told that their supervisor would not be notified of their decision to request an extension. In the policy condition, participants were told that if they requested a deadline extension their supervisor would receive the following note on top of their evaluation form: "Note: The formal policy of this assignment allowed people to request more time to work on the task. The employee asked for more time to work on the task." After the task, participants completed the Brief Negative Evaluation Scale (BFNE), a 12-item measure of self-presentation concerns that is widely used and well-validated (Leary, 1983;α = 0.95). This measure allowed us to test the relationship between impression concerns and extension requests. We tested whether formal policies reduced the proposed link between self-presentation concerns and extension requests when the request was visible to a supervisor.

Deadline extension request
There was a main effect of condition to predict whether participants pre-emptively asked for more time, F(2, 1418) = 5.01, p = .007, η 2 = 0.01. Replicating Study 1a, when the extension request was visible and self-presentation costs were high, participants were significantly less likely to pre-emptively ask for an extension (21.2%) as compared to when the request was not visible and self-presentation costs were low (29.2%), B = − 0.43, SE = 0.15, p = .004, Exp(B) = 0.65, 95% CI for Exp (B) [0.49, 0.88], Wald = 8.09, OR = 1.54 [1.14, 2.06]. As per our preregistered predictions, having a formal policy increased extension requests: In the policy condition, participants asked for an extension at a similar rate (28.8%) as participants who were randomly assigned to the low self-presentation concern condition (29.2%), B = − 0.02, SE = 0.14, p = .880, Exp(B) = 0.98, 95% CI for Exp(B) [0.74, 1.30], Wald = 0.02, OR = 1.02 [0.77, 1.35]. Running a Bayesian chi-square analysis with an uninformative prior resulted in a Bayes Factor of B 01 = 0.08, revealing moderate support for the null hypothesis (Baath, 2014). In these analyses H 0 (no differences across the low self-presentation concern and policy conditions) was 13.37 times more likely than H 1 (differences across the low self-presentation concern and policy conditions) to have generated the data.

Fear of negative evaluation
We then tested the moderating role of self-presentation concerns on the decision to request an extension. Within the low self-presentation concern condition, we observed a positive association between participants' self-presentation concerns and the decision to ask for more time, r (476) = 0.08, p = .080. These results suggest that participants with stronger impression concerns were more likely to ask for time when impression management concerns were low-likely to ensure that they did their best on the task.
Consistent with our conceptual model, in the high self-presentation concern condition, there was a significant negative association between impression management concerns and asking for more time, r (477) = − 0.17, p < .001. This result provides evidence that impression management concerns directly influenced participants' willingness to ask for more time. Most critically, in the policy condition, there was no association between impression management concerns and asking for more time, r(466) = − 0.04, p = .414. A Fisher R to Z transformation showed that the correlation coefficients in the high self-presentation concern and policy conditions were significantly different from one another, Z = 2.02, p = .022.
Next, we entered participants' BFNE scores, condition assignment (1 = policy, 0 = high self-presentation concern) and an interaction term between a centered BFNE variable and condition assignment to predict deadline extension requests. This analysis yielded a significant interaction between condition assignment and BFNE scores, B = − 0.33, SE = 0.15, p = .032, Exp(B) = 0.72, 95% CI for Exp(B) [0.54, 0.97], Wald = 4.61. Plotting this interaction further supported our conceptual model (see Fig. 3). The provision of a formal policy reduced the association between self-presentation concerns and deadline extension requests. Comparing the policy and high self-presentation concern conditions, even when these requests were visible to their supervisor in the policy condition, self-presentation concerns no longer prevented employees from asking a direct supervisor for more time to work on an adjustable deadline.

Discussion
In Study 3, we provide direct evidence for the relationship between self-presentation concerns and deadline extension requests. When the request was visible to supervisors, self-presentation concerns prevented employees from asking for more time on explicitly adjustable deadlines that improved their performance. When formal policies were put into place, employees' self-presentation concerns no longer prevented employees from asking for more time. Instead, employees-especially those with higher self-presentation concerns-asked for extensions at a similar rate as employees whose requests were not visible to their supervisors.
In Study 4, we aimed to replicate this result showing the importance of impression management concerns in an ecologically valid, real-world context. We recruited college students completing an assignment in an undergraduate business course. In one section of the course, as verbally described to students and outlined in the syllabus, students could ask for more time by asking the instructor for an extension before the assignment was due. In another class, students could ask for more time as outlined in a formal class policy verbally described in class and outlined in the syllabus. Following from Study 3, we tested whether a formal classroom policy led students to ask for extensions more often and to demonstrate lower competence concerns.

Participants and procedures
We recruited 90 college students who were enrolled in an undergraduate business course at a university located in the Midwest USA (56.7% female). Students were given one week to submit a discussion paper worth 20% of their grade. Treatment was randomized by section: to ensure similarly, both sections were taught by the same instructor and involved the same content.
In one section of the course (N = 50) students were told that there was an official class policy regarding a deadline extension request for the assignment. If students needed more time to complete the essay, they could email the instructor to request an extension without penalty. In the other section (N = 40) students were not informed of an official class policy. However, these students did receive the same instructions. If students needed more time to complete the essay, they were told that could email the instructor to request an extension without penalty. The same information was also reiterated on the syllabus (Appendix A). Importantly, students across both classes were provided with the identical information and opportunity to request an extension, yet the extension request was only framed as an official class policy in one of the class sections. This study therefore allowed us to replicate the results of Study 3 in an ecologically valid context. The sample size allowed us to detect a small effect of w = 0.23 in a chi-square test with 80% power.

Deadline extension request
Consistent with the results of Study 3, students were significantly more likely to request a deadline extension request from their instructor when the extension request was framed as a formal policy as compared to when it was not (42.0% vs. 22.5%), X 2 (1, 90) = 3.80, p = .051, 95%CI [0.98, 6.33], OR = 0.40, 95%CI [0.16, 1.02].

Self-presentation concerns
We used available data from this field experiment to assess selfpresentational concerns. Because the process of requesting an extension involved sending an email to the professor to ask for an extension, we evaluated whether there were differences in the content of the students' requests.
First, we computed a character count of the extension request. In the policy condition, students who requested an extension wrote fewer characters than students in the informal policy condition M policy = 252.95, SD = 138.92 vs. M informal policy = 448.78, SD = 268.98; t(28) = 2.65, p = .013, 95%CI [44.34,347.31], d = 1.05, suggesting that these students were less concerned with making a positive impression on their instructor when making the request.
Second, we captured the time stamp of when students made the request and computed how long the request was made before the deadline. Students in the policy condition made their request significantly closer to the deadline than students in the informal policy condition (M policy = 22:16, SD = 30.52 vs. M informal policy = 79:19, SD = 65:00; t (28) = 3.30, p = . 003,d = 1.31,95%CI [21:34,92:30]. These results suggest that students were less worried about their instructor's impression because they felt more comfortable asking for an extension closer to the deadline.

General discussion
Requesting an extension on a tight yet objectively adjustable deadline has the potential to reduce employee stress and improve task performance. Across five studies (three preregistered), employees avoided A.V. Whillans et al. requesting extensions on adjustable deadlines because they overestimated the self-presentational costs (Studies 1a & 3). However, in contrast to these beliefs, supervisors did not judge employees more negatively for requesting an extension (Studies 1b & 2). In fact, requesting an extension on an adjustable deadline led to the completion of higher quality work resulting in greater perceptions of competence (Study 1b) and higher grades on a consequential class assignment (Study 4). In Study 2, consistent with our conceptual model, and underscoring the importance of impression management concerns, we provided evidence that competence concerns were driving extension request avoidance as compared to other mechanisms such as concerns about burdening other people or concerns about seeming unmotivated or uncommitted. Furthermore, by reducing self-presentation concerns through the provision of a formal policy, people were more likely to request a deadline extension (Studies 3 and 4). These studies suggest that the overestimation of self-presentational concerns prevente people from taking advantage of extension requests, even when deadline adjustments are possible and beneficial.
In addition to formal policies, what other factors might reduce employees' self-presentation concerns? In two additional studies reported in the Supplemental Material (Supplemental Study B and C), increased relationship closeness between employees and supervisors reduced employees' self-presentation concerns as well as the perceived negative impact of an extension request on supervisors' evaluations. These findings are consistent with our theorizing about the important role of selfpresentation concerns in predicting employees' willingness to request an extension on adjustable deadlines at work. In particular, when a relationship between an employee and supervisor is more established, employees should be less worried about self-presentation concerns because the relationship provides a longer history that the evaluator can draw on when forming competence judgements. However, across studies, relational closeness did not close the gap between employees' predictions and supervisors' evaluations. Employees might be particularly concerned with performing well when they feel close to a supervisor, accounting for the persistent gap observed in perceived competence. More research should explore when and how relationship closeness facilitates extension requests.

Conceptual contributions
Our findings make several contributions to the request making and time management literatures. First, by evaluating request making in the context of adjustable deadlines, we extend past research that examines people's underestimation of request compliance (Bohns, 2016). We document a novel psychological barrier that hinders people from making helpful requests: self-presentation concerns. Across experiments, the probability of receiving an extension was explicitly high: In Study 1a and 3 participants received an extension if they pressed a button and in Study 4 all students received an extension who requested one-thus the cost of adjusting deadlines in these studies was relatively low. If employees based their extension requests solely on how beneficial and likely receiving an extension would be, most participants should have requested an extension. Yet, our data suggest that people often refrain from requesting an extension even in contexts where they recognize that an extension request would be helpful and unlikely to be rejected due to high perceived self-presentation costs.
Additionally, we shed light on a type of help-seeking where people's self-presentation concerns may be overestimated and unrealized: making deadline extension requests. In the domain of advice-seeking, prior research has found that advice-seekers often underestimate how competent they will appear, because they do not consider that their request for advice might boost the advice-giver's self-esteem, which can cause advice-givers to evaluate them more favorably (Brooks, Gino, & Schweitzer, 2015). In contrast to this work, our studies offer insight into the self-presentational effects of help-requests that do not boost the selfesteem of the requestee.
Finally, our work provides evidence that challenges the common notion that delays are interpersonally costly (e.g. Sheldon, Thomas-Hunt, & Proell, 2006). Research suggests that delays in organizations (e.g. delays in flight schedules or being late to meetings) often evoke anger (Weiner, 1985), impatience (Blount & Janicik, 2001), and frustration (Amsel, 1992;Spector, 1978). Requests that delay tasks can come at the cost of the perception of the requester's competence, particularly among low-status actors (Sheldon et al., 2006). However, across our experiments, supervisors did not always react negatively to lower status employees' requests for more time. Across studies, requesting an extension allowed the requesters to improve task performance, which led to a boost in perceived competence from their supervisor and enhanced assignment evaluations.
One critical difference between our studies and prior research is that the employees we studied preemptively requested a deadline extension before the deadline had passed, whereas in prior studies observers' expectations were violated without warning (e.g., an employee submitting the task late without notice; Sheldon et al., 2006). Requesting an extension before missing a deadline may help employees demonstrate an awareness of their own capabilities, protecting perceived competence. Our results suggest that a proactive extension request is an effective strategy to minimize interpersonal costs when task delays are imminent. Future work should directly compare reactions to extension requests made both before and after a deadline.

Future directions
In this paper, we focused on requests for extensions on adjustable deadlines to provide a conservative test of our conceptual framework. If participants were unwilling to ask for more time on relatively costless, effortless, and adjustable deadlines at work, they should also be less likely to ask for more time on costlier, more effortful, and less adjustable deadlines. Of course, in real-life situations deadlines extensions may require effort and could have negative consequences. Thus, to move beyond the tightly controlled vignette scenarios that we used in Studies 1a-3, we studied extension requests in a real-world situation where college students could request a deadline extension from their instructor (Study 4). While this request was not exceptionally costly to the instructor-who noted that students could ask for an extension-it did involve significant effort-students had to communicate with the instructor to make the request. In this study, the instructor was a longterm interaction partner with whom students were presumably interested in making and maintaining a positive impression. This study thereby provided a more ecologically valid context by which to observe the link between competence concerns and deadline extension requests.
Because our studies purposefully focused on contexts where deadlines were explicitly adjustable and relatively costless to adjust, these research findings may not generalize to deadlines that are perceived as costly-to-adjust by both parties. Consistent with this idea, both supervisors and employees believed that asking for an extension on a costlyto-adjust deadline would signal a greater lack of employee competence (Supplemental Study D). Yet, employees still overestimated these competence costs as compared to managers' judgements. This study suggests that employees still overestimate managers' judgements, but are somewhat more well-calibrated to the consequences of requesting extensions on costly-to-adjust deadlines. Future research should explore whether employees' impression management concerns are better calibrated in the context of repeated requests or in the context of requests that create noticeable delays. Future research should also explore how the length of the extension request (such as asking for one day vs. one month) shapes both employees' perceptions and managers' reactions.
In our studies, we evaluated one-time (versus repeated) deadline extension requests, where employees asked for a deadline extension at a single time point. It is possible that asking for an extension request for the first time might heighten employees' concerns with appearing incompetent, while requesting extensions repeatedly might desensitize employees to possible interpersonal concerns. It is also possible that employees who frequently request extensions at work are employed in contexts that explicitly allow for these requests, which-following from our data-should facilitate greater comfort. In line with this possibility, employees who more frequently requested extensions were less likely to believe that doing so would decrease their perceived competence (Supplemental Study C). As this evidence is correlational, more experimental research is needed to expand on these findings such as by showing that existing policies and social norms shape requests for and perceptions of repeated extension requests.
Requesting a deadline extension has the potential to both benefit and hurt perceived competence because deadline extensions can improve performance quality at the expense of speed. Consistent with prior research (Brewer & Ridgway, 1998), our data show that when evaluation criteria are ambiguous and quality is subjectively measured (as was the case with all of the task stimuli used across our studies), employees focus more on impressing supervisors by optimizing speed over quality. In fact, in Study 2, employees underestimated the importance of quality, undermining extension requests. Employees' focus on performance speed is consistent with research showing that people tend to maximize easily measured mediums, such as points and time (Hsee, Yu, Zhang, & Zhang, 2003). Yet, in some contexts-such as when the importance of quality is explicitly stated-quality might be prioritized over speed. In these contexts, employees may be less concerned with self-presentation when requesting an extension and supervisors may react more favorably. Future research would benefit from examining how explicitly communicating the organizations' focus on task speed or task quality influences employees' willingness to request deadline extensions and supervisors' reactions to employees' extension requests.

Conclusion
While deadlines are useful for productivity and coordination, people may benefit from pushing back tight yet adjustable deadlines. However, people often avoid making extension requests, overstretching themselves to meet deadlines by engaging in suboptimal strategies like multitasking and working unnecessary overtime (Waller, Giambatista, & Zellmer-Bruhn, 1999). Across five studies, we highlight an underexplored factor that contributes to the unwillingness to request help for tasks like requesting an extension for workplace deadlines: people overestimate how negatively others will judge them for their extension request. Our research expands the literature on help-requests and selfpresentation, highlighting a promising point of intervention for time stress at work: reducing the perceived self-presentational costs of asking for more time.

Appendix B
Study 4 syllabus information across conditions.

Syllabus Assignment Information
Super Bowl Ad Reflection: The purpose of this reflection is to provide you with the opportunity to apply your knowledge of [name of course] and marketing communications to Super Bowl advertisements. This is an individual assignment where you will be expected to pick a Super Bowl ad from 2020. In this 2-page reflection (single spaced, 12-point font), you will be asked to review an advertisement that speaks to socio-ecological problems, the SHIFT model and social marketing. The Super Bowl will take place on Sunday, February 7 at 6:30 pm EST. This paper is due by midnight on Monday, February 8th as we will discuss Super Bowl ads in class on Tuesday, February 9th. This class has a policy on deadline extensions. If you need an extension for this assignment, please email me [email of professor] to request one.

Super Bowl Ad Reflection:
The purpose of this reflection is to provide you with the opportunity to apply your knowledge of [name of course] and marketing communications to Super Bowl advertisements. This is an individual assignment where you will be expected to pick a Super Bowl ad from 2020. In this 2-page reflection (single spaced, 12-point font), you will be asked to review an advertisement that speaks to socio-ecological problems, the SHIFT model and social marketing. The Super Bowl will take place on Sunday, February 7 at 6:30 pm EST. This paper is due by midnight on Monday, February 8th as we will discuss Super Bowl ads in class on Tuesday, February 9th. If you need an extension for this assignment, please email me [email of professor] to request one.

Deadline Extension Policy:
For all individual assignments, the is a formal deadline extension policy. If you would like an extension for any of these assignments, you will need to email me [email of professor] to request one. All requests should be made prior to the deadline. For example, if the assignment is due by 5 pm on February 1, you will need to make your request by 5 pm on February 1.

Not applicable.
There was no policy about deadline extensions listed in the syllabus.
To maintain confidentiality, we have only presented the key information from the syllabus. The full versions of the syllabi for these classes are available by request from the authors.