Agency and Communion Predict Speaking Up in Acute Care Teams

Abstract

Speaking up with suggestions, problems, or doubts is important—especially in health care action teams where each team member’s input can be crucial for the treatment of a patient. Implementing a high-fidelity simulation study, we investigated individual predictors of speaking up in acute care teams (ACTs). Participants were 27 physicians and 27 nurses from a hospital who completed measures on self-perceived agency (i.e., assertiveness, persistence, independence) and communion (i.e., helpfulness, friendliness, sociability). In two-person teams, they managed simulated critical events that required speaking up. In line with our hypotheses, we found that agency positively and communion negatively predicted actual speaking up behavior. We discuss the differential effects of agency and communion on speaking up and thereby highlight theoretical and practical implications.

Keywords

speaking up agency communion action teams health care

In various interpersonal, team, and organizational contexts, people are faced with the decision of whether they should speak up or remain silent—and a vast body of research shows that far too often they choose the latter (Morrison, 2011). For health care professionals, it can be disastrous not to speak up with concerns as this might affect patient safety (e.g., Bromiley, 2008; Sexton, Thomas, & Helmreich, 2000). For example, speaking up can be required when a wrong diagnosis seems to be made, when another team member repeatedly fails to intubate the patient successfully, or when the team does not perform resuscitation despite a significant period of low cardiac output. By combining Morrison’s (2011) definition of voice and Edmondson’s (1999) definition of speaking up, we use both terms interchangeably and define speaking up as explicit communication of suggestions, problems, opinions, or doubts that challenge the status quo. Challenging the status quo in a group may involve interpersonal risk—especially in hierarchically structured groups such as health care action teams. For example, people fear that they might be punished or lose face when they would voice concerns to those in power and thus refrain from speaking up (e.g., Kobayashi et al., 2006).

Research has identified a variety of contextual factors that influence speaking up such as organizational structure (Morrison & Milliken, 2000), psychological safety (Edmondson, 1999), and supervisor behavior (e.g., Edmondson, 2003; Nembhard & Edmondson, 2006). In addition to these contextual factors, individual factors seem to affect speaking up as well. Even within the same context, people differ in their speaking up behavior (e.g., Premeaux & Bedeian, 2003). By stating that “speaking up is not an automatic result of removing barriers to expression,” Islam and Zyphur (2005, p. 101) point to the importance of studying interindividual differences as predictors of speaking up behavior in groups. This raises the question as to what kind of individual factors determine whether people speak up or remain silent. Grasping how individual factors contribute to speaking up is important not only for better understanding the mechanism of voice behavior but also for designing training programs and interventions targeted at enabling team members to speak up.

In this study, we build on and extend prior research in three ways: First, rather than analyzing self-reported speaking up, we look at actual speaking up behavior in health care teams and thereby extend the theoretical understanding of voice behavior to real-world action teams in a high-risk context. Second, we use the fundamental trait dimensions of agency (i.e., assertiveness, persistence, independence) and communion (i.e., helpfulness, friendliness, sociability) to identify traits that enable versus traits that inhibit speaking up. Third, based on our findings, we discuss implications for simulation-based team trainings and speaking up interventions. In the following, we (a) briefly describe the particular characteristics of acute care teams (ACTs), a special form of health care action teams, and the importance of speaking up in this context; (b) introduce agency and communion as two fundamental trait dimensions; and (c) subsequently derive two hypotheses to be tested in a simulation-based setting.

Speaking Up in ACTs

ACTs are composed of nurses and physicians who are put together ad hoc to perform various tasks. For example, they deliver anesthesia during an elective surgery, provide initial treatment and diagnoses to emergency patients, or are called to manage acute intra-hospital emergencies such as a cardiac arrest.¹ ACTs are a typical example of action teams as they engage in brief performance episodes to manage unpredictable events (Ishak & Ballard, 2012; Sundstrom, de Meuse, & Futrell, 1990). Moreover, ACTs are part of a highly diverse multi-team system composed of surgical teams, radiologists, technicians, or recovery room teams (cf. Mathieu, Marks, & Zaccaro, 2001). Engaging in explicit communication and speaking up with ideas can be key for team-based decision making and for finding the correct diagnosis (cf. Tschan et al., 2009). More importantly, speaking up in ACTs can also help to avoid medical mishaps and severe errors (Edmondson, 2012). Thus, it is crucial that people in ACTs speak up when they think that an alternative dose of medication seems indicated (i.e., suggestion), when a team colleague seems too tired to operate, when they feel that there are differential diagnoses, or when they simply question the appropriateness of a treatment or procedure.

Yet, the steep hierarchy prevalent in ACTs and inherent to the medical profession can pose a barrier to people’s speaking up behavior. Particularly, people with lower status such as residents and nurses may feel that challenging a person with higher status (e.g., an attending surgeon) could cause negative evaluations or even jeopardize their position (Bisel, Messersmith, & Kelley, 2012; Milliken, Morrison, & Hewlin, 2003). Residents indicated that they would not speak up to a surgeon who causes an intra-surgical complication as their fear penalties or negative repercussions (Kobayashi et al., 2006). However, even though hierarchy is a prevalent barrier to speaking up, there is evidence that some people do speak up. A recent study with anesthesia teams has revealed that nurses even spoke up more than physicians during a simulated critical event and that nurses’ amount of speaking up was positively associated with team performance (Kolbe et al., 2012). To better explain such counterintuitive findings, we suggest that it may be fruitful to consider interindividual differences as a predictor of voice behavior. Especially in a highly dynamic action team in which team members have to constantly adapt to changing circumstances, looking at individual differences can be helpful to make predictions across situational contexts and to better understand why people speak up or remain silent (cf. Baard, Rench, & Kozlowski, 2014).

Agency and Communion as Predictors of Speaking Up in ACTs

There are two fundamental trait dimensions that are relevant to the understanding of interpersonal behavior, namely agency—also called instrumentality or competence—and communion—also called expressiveness or warmth. (e.g., Abele, 2003; Bakan, 1966; Fiske, Cuddy, Glick, & Xu, 2002). Agency refers to individuals’ desire to master the environment, assert oneself, and experience competence, achievement, and power. Communion, however, refers to people’s desire to closely relate to others and is reflected in being kind, helpful, and nice. The two dimensions have been related to gender stereotypes with agency relating to prototypically masculine and communion relating to prototypically feminine traits (e.g., Bem, 1974; Spence & Helmreich, 1978). They are also conceptually related to the Big Five with agreeableness, conscientiousness, and emotional stability relating to communion, and extraversion and openness relating to agency (Digman, 1997). Thus, agency and communion can be viewed as the Big Two that reflect the duality of human existence: our striving for self-assertion and our desire for belongingness (Bakan, 1966). As voice has been conceptualized as a deliberate decision process in which individuals weigh the pros and cons of speaking up (Morrison, 2011), we propose that agency and communion may likewise affect this decision in ACTs (i.e., being assertive for the patient’s well-being vs. wanting to be liked by colleagues).

Although research looking at individual predictors of speaking up in teams is rare, some studies have used agency and communion or the Big Five to predict voice behavior or other forms of assertive communication in group contexts. One study has investigated verbal assertiveness in small groups and found that high-agency persons used more direct statements than low-agency persons, but that communion was not predictive of assertive behavior (Leaper, 1987). Looking at leadership behavior as a particular form of assertive communication, there is evidence that traits pertaining to the communion dimension can even have a negative effect. Streiff and colleagues (2011) conducted a study with medical students and simulated cardiac arrests. They found that people who scored high in extraversion (i.e., related to agency) and low in agreeableness (i.e., related to communion) were most likely to make leadership statements such as making decisions, voicing commands, and assigning tasks within the team. A similar pattern of results could be shown for the relationship between the Big Five and voice behavior assessed in laboratory study. LePine and Van Dyne (2001) found that extraversion related positively and agreeableness related negatively to constructive change-oriented communication. On the basis of these findings, we propose that agency may likewise have a positive effect on speaking up behavior in ACTs. Agentic traits such as being assertive, independent, and persistent may foster people’s willingness to voice suggestions, problems, opinions, or doubts. By contrast, communion seems to have a rather negative effect on assertive communication in groups. People for whom being highly communal (i.e., being kind, friendly, and nice) is central to their self-definition may be more likely to be silent. Thus, we hypothesize as follows:

Hypothesis 1: The more agentic team members perceive themselves the more likely they are to speak up during critical events.

Hypothesis 2: The more communal team members perceive themselves the less likely they are to speak up during critical events.

We tested these hypotheses in a high-fidelity simulation setting involving two-person anesthesia teams (i.e., one physician, one nurse) to best reflect the characteristics of ACTs. The teams were required to manage unexpected critical events (e.g., treatment of an unexpected difficult airway) during which they were confronted with multiple possibilities to speak up not only within their own team but also across disciplines (i.e., toward a surgeon). By investigating our hypotheses in a highly realistic simulated environment, we also follow recent calls to use simulation to adequately study processes in action teams (Ishak & Ballard, 2012).

Method

Participants

We implemented this study within 1-day simulation-based team-training sessions for anesthesia staff members from a university hospital. The entire study period involved 10 training days with up to six participants (e.g., three physicians and three nurses) participating each day. In all, 54 participants (33 females, 21 males, M_age = 34.07, SD = 7.61, M_{work experience} = 5.08, SD = 7.46) participated in the study. Of these 54 participants, 27 were anesthesia nurses and 27 were anesthesia physicians (26 residents, 1 attending). On each study day, the nurses and physicians were randomly put together to form a two-person anesthesia team consisting of one nurse and one physician. This resulted in 2 or 3 teams each day and in 27 anesthesia teams across the study period. Prior to the study, all participants were provided with information concerning the procedure and written consent was obtained. All participants participated for training purposes and received advanced education credits.

Simulator

The study was conducted at a simulation center utilizing a full-scale patient simulator (SimMan3G®, Laerdal, Norway) placed in a simulated operating room (OR). The patient simulator has highly realistic anatomic and physiological features and is able to breathe, blink, and talk (through a microphone). Chest movements can be observed and heart and lung sounds can be heard with a stethoscope; the pulse can be palpated. Vital parameters such as heart rate, pulse, respiratory rate, body temperature, oxygen saturation, and blood pressure were adjusted by attending anesthesiologists from a control room next door and directly corresponded with the participants’ actions. The simulated OR was equipped with an anesthetic machine, an intravenous pole, and a cart with medical supplies (e.g., drugs, laryngoscope), and provided a highly accurate reflection of participants’ natural work context. Three cameras were placed in the room to enable video recording from different perspectives.

Procedure and Scenarios

On arrival, all participants were introduced to the simulation facilities and received a comprehensive familiarization with the patient simulator. Following that, they completed a questionnaire including measures of demographics and work experience, as well as agency and communion. Subsequently, all participants were randomly assigned to two-person anesthesia teams involving one physician and one nurse. Depending on the number of teams we could form, we simulated up to three different medical scenarios on 1 training day. Importantly, each team performed only one scenario lasting approximately 20 min. The remaining participants scheduled for that day were able to observe the performing team through live video broadcasting in a room next door. All simulated scenarios were followed by an extensive instructor-led debriefing of which all participants were part. During this debriefing, the medical and teamwork challenges such as leadership, situational awareness, and communication were elaborated on, and potential opportunities for speaking up were discussed (Kolbe, Weiss, Grote et al., 2013).

The development of the scenarios was based on critical incidents reports and conducted by three attending anesthesiologists, one resident anesthesiologist, and two psychologists. Each scenario was designed to contain at least one speaking up opportunity—either within the anesthesia team or against a confederate surgeon—that was highly important for successful management of the critical events. Depending on the participants’ behavior, there could also be more speaking up opportunities if a team member’s behavior required unforeseen correction. As limitations in staff availability prevented us from performing each scenario on each study day, we wanted to ensure that the scenarios were evenly distributed with respect to opportunities to speak up within the team as well as across teams. Thus, Scenario C, which involved a confederate surgeon to create across-team speaking up opportunities, was performed 26 times over the whole study period. Scenarios A and B were performed 10 and 18 times, respectively. The usability of similar scenarios for investigating speaking up behavior in ACTs has been tested in prior studies (Kolbe et al., 2012; Kolbe, Weiss, Grote et al., 2013).

Scenario A: Unexpected difficult airway

This scenario was designed to simulate an unexpected difficult airway occurring during the induction phase of a general anesthetic for an elective surgery. In this phase, the patient receives a hypnotic (e.g., propofol), an analgetic (e.g., fentanyl), and a muscle relaxant (e.g., rocuronium) leading to unconsciousness and a total loss of muscle tone and ability to breathe autonomously. Thus, the patient needs to be ventilated with a facemask; as the induction phase progresses, intubation with a tube is required (i.e., ventilation tube inserted into the trachea). A difficult airway is defined as the clinical situation in which a conventionally trained anesthesiologist experiences difficulty with mask ventilation, difficulty with tracheal intubation, or both (Apfelbaum, Hagberg, & Caplan, 2013). In this case, the oxygen saturation decreases quickly, which leads to severe brain damage within minutes and eventually to death. The anesthesia team was required to follow an algorithm for unexpected difficult airway involving alternative intubation techniques (e.g., a fiber optic). Several opportunities to speak up emerged concerning the implementation of the difficult airway algorithm (see Table 1).

Table 1.

Examples for Opportunities to Speak Up and Respective Speaking Up Types Within Each Scenario.

Scenario	Opportunity to speak up	Speaking up type
A. Unexpected difficult airway	Omission AP starts to deliver the first drug without notifying AN who adjusts monitor settings. Harmful suggestion AP has failed twice to intubate the patient with the tracheal tube and does not proceed with the algorithm for difficult airway that requires alternative intubation techniques. AP: “I’m going to try [intubation with tracheal tube] again.” Harmful action AP is inspecting the difficult airway and carefully tries to place the tube into the trachea. AN pushes the tube in a different direction.	Suggestion-focused voice AN: “Hang on, I am not ready with the settings yet. I would wait until everything is clear with it.” No speaking up AN remains silent and assembles the instruments for a conventional intubation. Problem-focused voice (with explicit stopping) AP takes over control of the tube and says: “No, please don’t do anything!”
B. Difficult airway and cardiac arrest	Omission AP monitors the patient while AN is answering the phone. Meanwhile, the heart rate drops significantly (i.e., bradycardia) such that immediate resuscitation is necessary. Harmful suggestion AP and AN have decided to use a fiber optic for intubation as the patient chart describes the airway as difficult. They both know that conventional intubation with tube is not possible. AP: “Please hand me the laryngoscope and the tube!” Harmful action AP is ventilating the patient with a facemask. Although ventilation is rather insufficient (i.e., indicated by decreased thorax movements), AP suggests to starting the anesthetic by applying the first drug.	Problem-focused voice AP: “Please go away from the phone and prepare some Epinephrine [for resuscitation]!” Doubt-focused voice AN: “Don’t you want the fiber optic?” No speaking up AN observes the decreased thorax movements but does not voice this observation and prepares the medication.
C. Tension pneumothorax	Omission SRG: “Alright. Now turn off the light (i.e., to better view the monitor for the laparoscopy) and then we get going!” Harmful suggestion AP and AN are busy with the treatment of the pneumothorax when phone starts to ring. SRG to AP: “Can you answer the phone, it’s my secretary!” Harmful action The patient’s oxygen saturation decreases rapidly because SRG has injured the diaphragm and thereby caused a pneumothorax. SRG: “What is your problem over there? I need to finish the surgery here.”	Opinion-focused voice AP: “Actually, we have not finished the checklist (i.e., WHO surgical checklist to be performed by surgeons and anesthesiologists before incision) yet. What kind of surgery is this?” No speaking up Although AP is required to help AN, he does not voice concerns and picks up the phone for SRG. Problem-solving voice (explicit stopping) AN: “No, you need to wait. She [the patient] is getting cyanotic (i.e., blue skin color due to decreased oxygen saturation).”

Note. AN = Anesthesia Nurse; AP = Anesthesia Physician; SRG = (Confederate) Surgeon.

Scenario B: Difficult airway and cardiac arrest

This scenario was similar to Scenario A as it also involved the treatment of the difficult airway. However, in this scenario, the difficult airway was to be expected. The team was handed the patient chart that contained critical information about the airway, and thus the team could prepare for an alternative intubation and for implementing a difficult airway algorithm (e.g., by assembling special tools for difficult intubation). However, after the patient was successfully intubated with an alternative intubation technique, the heart rate first increased rapidly (i.e., tachycardia) and then decreased over a short period of time (i.e., bradycardia), eventually, leading to asystole (i.e., no cardiac electrical activity as evidenced by a flatline on the electrocardiogram). This situation requires immediate cardiopulmonary resuscitation through chest compressions and drugs (e.g., epinephrine). Several opportunities to speak up emerged concerning the procedure of the difficult airway algorithm as well concerning the initiation of the resuscitation (see Table 1).

Scenario C: Tension pneumothorax

This scenario started off as an elective surgical procedure in the OR. In addition to the anesthesia team, a surgeon was present to perform a laparoscopy. A male confederate endorsed the role of a surgeon. He was instructed to act unfriendly and ignorant toward the anesthesia team (e.g., complaining about the high room temperature). The task of the anesthesia team was to monitor the patient during surgery. After the team started to check the status of the patient, the surgeon injured the diaphragm with a needle and inflated the pleural cavity with carbon dioxide, which lead to a tension pneumothorax. A tension pneumothorax results from collection of air between the lungs and the chest wall that deflates the lung and immediately leads to impairment of respiration. In a very short period of time, this leads to a significant decrease of arterial oxygen saturation and cardiac output (because of missing cardiac volume preload). The pneumothorax can be a lethal condition within minutes if it remains untreated (Leigh-Smith & Harris, 2005). The task of the anesthesia team was to recognize the lowered oxygen saturation and treat the pneumothorax by releasing the air with a needle. Among other speaking up opportunities, the key speaking up challenge for the anesthesia team was to stop the surgeon from continuing with the laparoscopy (see Table 1).

Measures

Agency and communion

To assess agency and communion, participants completed a shortened eight-item version of the German Extended Personal Attributes Questionnaire (GEPAQ; Runge, Frey, Gollwitzer, Helmreich, & Spence, 1981). In line with previous research, we measured agency and communion, respectively, with the Positive Masculinity and the Positive Femininity subscales of the GEPAQ (cf. Abele, 2003). The GEPAQ is based on the Personal Attributes Questionnaire (PAQ; Spence, Helmreich, & Stapp, 1974) that was originally designed to measure a person’s orientation toward instrumentality/masculinity, expressiveness/femininity, and androgyny. The original GEPAQ consists of 40 items, assessing each of the three dimensions as well as differentiating between positive and negative masculine and feminine traits. The Positive Masculinity scale consists of eight items: (a) independent, (b) stand up well under pressure, (c) never give up easily, (d) self-confident, (e) can make decisions easily, (f) active, (g) competitive, and (h) superior. The Positive Femininity scale also consists of eight items: (a) helpful to others, (b) able to devote the self completely to others, (c) kind, (d) aware of others’ feelings, (e) understanding of others, (f) warm in relations with others, (g) gentle, and (h) emotional. As this study was part of a larger research project, we selected the first four items from each scale as a parsimonious measure for agency and communion. Selection of items was based on internal discussions among the research team about which attributes might be relevant in an acute care context. Participants were asked how they would generally describe themselves using a 5-point scale (1 = not at all, 5 = very much). Reliability of both measures was satisfactory (agency: α = .66; communion: α = .71) and similar to other studies using complete scales (e.g., Abele, 2003; Abele & Wojciszke, 2007). To provide evidence that agency and communion represented distinct constructs, we performed a confirmatory factor analysis using AMOS 22. The fit indices suggest that the two-factorial model provided an acceptable fit to the data (χ² = 33.74, df = 19, p = .02; goodness of fit index [GFI] = 0.88, root mean square error of approximation [RMSEA] = 0.12, standardized root mean square residual [SRMR] = 0.09).

Opportunities to speak up

To specify the number and kind of speaking up situations that emerged within each team, we used the video records of all simulated scenarios. An attending anesthesiologist and a psychologist jointly identified all situations in which speaking up was necessary. A speaking up opportunity was classified as an omission (e.g., anesthesia nurse does not perform chest compressions within 30 s after beginning of an asystole), a harmful suggestion (e.g., the anesthesia physician suggests to intubate conventionally although the airway is difficult), or a harmful action (e.g., the confederate surgeon harms the patient during surgery). Importantly, speaking up opportunities emerged both from planned, external manipulations (e.g., confederate surgeon) and unplanned, team-internal discrepancies during the treatment of a patient (e.g., how a difficult airway should be treated). We coded all speaking up opportunities on an individual level, thus taking into account whether the nurse, the physician, or both could speak up.

Speaking up

After having defined opportunities to speak up for each scenario, two further trained raters independently coded each team member’s utterances to detect whether people spoke up during the predefined speaking up opportunities. To this end, we applied a modified version of a coding manual developed for ACTs (Kolbe, Burtscher, & Manser, 2013) that differentiates between suggestion-, problem-, opinion-, and doubt-focused voice (cf. Morrison, 2011): Suggestion-focused voice was coded if a team member suggested a different course of action (e.g., “Should we apply another dose of propofol before we start the laryngoscopy?”). Problem-solving voice was coded when a team member voiced problems regarding the patient’s condition or regarding a certain course of action (e.g., “We have a problem, the oxygen saturation is decreasing!”). It was also coded if a team member voiced a problem and explicitly stopped another person from a detrimental course of action (problem-solving voice with explicit stop; for example, “Stop! We have a problem and we cannot continue with the surgery!”). Opinion-focused voice was coded if a team member expressed a different opinion (e.g., “No, I think we need to apply more muscle relaxant because she has a laryngospasm.”). Doubt-focused voice was coded if a team member questioned assumptions or actions of his or her teammate (e.g., “Could it be possible that you perforated something?”).

Using an event-sampling coding approach in Interact (Mangold International, Germany), all utterances were coded with respect to speaking up type (i.e., five types of speaking up vs. no speaking up), actor (i.e., nurse, physician), receiver (i.e., nurse, physician, surgeon), and time. We computed initial agreement among the coders in two steps. The first coder coded one video involving one nurse and one physician. The second coder received a data file with all set events specified by time but no codes and coded the same person. As each of the five speaking up types occurred very infrequently, we were not able to compute agreement scores for each of the five speaking up types. Thus, we merged all speaking up types into a dummy-coded speaking up variable and computed Cohen’s kappa subsequently. This resulted in a satisfactory agreement score (Cohen’s kappa = .76). After computing the initial agreement, both coders continued independently with the coding but met on a regular basis to compare their coding, compute Cohen’s kappa, and solve potential disagreements through discussions.

Control variables

We used age, sex, professional role (i.e., physician vs. nurse), and work experience (i.e., work experience in anesthesia since graduation from medical/nursing school) as control variables as prior research has demonstrated an effect on speaking up (Morrison, 2011). We also used opportunities to speak up as a control variable, as we found that Scenario C (M = 4.77, SD = 2.37) involved more speaking up opportunities than Scenario B (M = 1.33, SD = .59) and Scenario A (M = 1.30, SD = .95), F(2, 51) = 26.42, p < .001. In addition, we investigated whether controlling for the sequence of the simulated scenarios was necessary. As teams participating in the second or third scenario on a given training day were able to observe other teams during the simulation and also participated in their debriefing, we tested whether they might have benefitted and thus spoke up more. An ANOVA with simulation sequence (i.e., first, second, or third team) as between-subjects factor and speaking up as dependent variable revealed no significant differences, F(2, 51) = 1.03, p = .365.

Results

Before testing our hypotheses, we provide a descriptive overview on the behavioral data on speaking up. We found that even though all participants were confronted with at least one critical speaking up opportunity, 22 of all 54 participants (41%) did not speak up. To gain a more precise picture of the five different speaking up types, we compared physicians and nurses and the frequency of each speaking up type in the two groups. Across all simulated scenarios, 16 of all 27 physicians and 16 of all 27 nurses spoke up at least once (see Table 1). Although physicians (M = 1.63, SD = 1.99) had slightly more speaking up utterances than nurses (M = 1.11, SD = 1.22), we did not find a significant difference between the two professional groups regarding their overall speaking up frequency, t(52) = 1.16, p = .253. By comparing absolute frequencies of the five different speaking up types, we found that while nurses engaged mostly in opinion-focused voice, physicians engaged mostly in problem-solving voice with explicit stopping of an action (see Table 2).

Table 2.

Frequencies of Speaking Up Types by Nurses and Physicians.

Speaking up type	Frequencies (%) of speaking up types within professional group
Speaking up type	Nurses (n = 16)	Physicians (n = 16)
1. Suggestion-focused voice	4 (17%)	5 (16%)
2a. Problem-solving voice	5 (21%)	1 (3%)
2b. Problem-solving voice (explicit stopping)	5 (21%)	13 (42%)
3. Opinion-focused voice	8 (33%)	10 (32%)
4. Doubt-focused voice	2 (8%)	2 (7%)
Total	24 (100%)	31 (100%)

Note. Total number of frequencies is higher than the number of individuals within each group as people could speak up more than once using different types.

We also tested whether physicians and nurses differed with respect to whom they addressed their speaking up behavior. While physicians spoke up more often toward the confederate surgeon, nurses spoke up more often toward physicians within their own team. The proportion of nurses who spoke up toward the confederate surgeon was 0.18 whereas the proportion of physicians who spoke up toward the confederate surgeon was 0.82. This difference in proportions is significant, χ²(1, 25) = 5.31, p = .027.

Agency and Communion as Predictors of Speaking Up

Table 3 presents the means, standard deviations, and correlations between all assessed variables. To test both hypotheses, we performed a hierarchical regression analysis with speaking up frequency (absolute frequency) as dependent variable.² In the first step, we entered age, sex, professional status, work experience, and opportunities to speak up as control variables. In the second step, we entered agency and communion as main effects. Beta weights for each step as well as ΔR² for the two models are shown in Table 4. Neither age, gender, professional status, or work experience significantly predicted speaking up in the first step. However, we found a significant effect of opportunities to speak up (β = .62, p < .001). In line with our first hypothesis, we found that—above and beyond the effect of the number of speaking up opportunities—agency was a significant positive predictor of speaking up (β = .33, p = .011) whereas communion was a significant negative predictor of speaking up (β = −.29, p = .027). R² significantly changed when agency and communion were entered into the model, ΔR² = .09, F(2, 46) = 4.16, p = .022. The same direction of effects emerged when the regression analysis was repeated without covariates (for agency β = .33, p = .034 and for communion β = −.38, p = .014, respectively). We also tested for a possible interaction between agency and communion on speaking up but found no significant effect.

Table 3.

Means, Standard Deviations, and Correlations Between Variables.

Variable	M	SD	1	2	3	4	5	6	7
1. Age (in yrs)	34.07	7.61	—
2. Sex^a	0.39	0.49	−.16	—
3. Professional status^b	0.00	1.01	−.31*	.42**	—
4. Work experience (in yrs)	5.08	7.46	.81**	−.23	−.36**	—
5. Agency	3.76	0.37	−.06	.05	.04	−.03	—
6. Communion	4.12	0.53	−.17	−.06	−.05	−.14	.50**	—
7. Opportunities to speak up	2.98	2.44	−.12	.27*	.19	−.10	−.05	−.12	—
8. Speaking up	1.09	1.46	.07	.08	.09	.15	.18^†	−.22*	.60**

Note. N = 54, ^a0 = female, 1 = male, ^b0 = nurse, 1 = physician, ^†p < .10. *p < .05. **p < .01.

Table 4.

Hierarchical Multiple Regression Analysis Predicting Speaking Up Frequency.

Model	Speaking up
Model	B	SE B	β
Step 1
Age (in yrs)	0.03	0.04	.14
Sex^a	−0.42	0.43	−.13
Professional status^b	0.25	0.21	.15
Work experience (in yrs)	0.01	0.04	.05
Opportunities to speak up	0.42	0.08	.62***
ΔR²			.40***
Step 2
Agency	1.45	0.54	.33*
Communion	−0.89	0.39	−.29**
ΔR²			.09*

Note. N = 54, ^a0 = female, 1 = male, ^b0 = nurse, 1 = physician, *p < .05. **p < .01. ***p < .001.

Thus, on the basis of these results, we find empirical support for Hypotheses 1 and 2. The more agentic team members perceive themselves the more likely they are to speak up. In contrast, the more communal team members perceive themselves the less likely they are to speak up.

Discussion

The goal of this research was to identify individual antecedents of speaking up in a typical action team, namely, an ACT. We implemented a high-fidelity simulation study with anesthesia teams and studied the relationship between agency and communion and speaking up. In line with our predictions, we found that agency was a positive and communion was a negative predictor of speaking up. Notably, we found that a high proportion of people remained silent even though this caused a highly critical or even fatal condition for the “patient.” Importantly, whether team members chose to speak up or remain silent was not dependent on their age, their work experience, or their professional status but solely on how agentic and communal they perceived themselves to be. That is, people who feel that they are highly agentic and rather less communal were most likely to speak up. Conversely, the barriers to voice concerns might have weighed especially high for those individuals for whom being friendly and kind are more relevant to their self-definition than being assertive and independent.

Theoretical Implications

This research contributes both to the literature on voice as well as a better understanding of the behavioral effects of agency and communion. As has been discussed in the literature, agency is primarily directed at asserting the self and one’s interests, and communion is primarily directed at other people (Abele & Wojciszke, 2007). The present research shows that agency—but not communion—is also helpful when team members speak up with team- or patient-related concerns. Thus, agency can also be profitable from a team perspective. High levels of communion and low levels of agency, in contrast, hinder team members to voice their concerns as that this might lead people to focus on interpersonal consequences of speaking up. For example, these individuals might have remained silent because they feared to confront their teammates and to jeopardize future work relationships (Detert & Edmondson, 2011; Morrison, 2011). As Bienefeld and Grote (2012) have shown while studying speaking up in airline crews, especially captains and first officers—who are highly trained in crew resource management (CRM) skills—indicated that the most common reason for their silence is that they fear to damage relationships with their colleagues when they speak up. In light of the present findings, it seems that if people define themselves as highly communal and rather less agentic—regardless of professional status—they might believe that speaking up harms the work relationship (Detert & Edmondson, 2011). This corresponds to the finding that highly communal people are strongly motivated by social emotions such as anticipated guilt or gratitude (Grant & Wrzesniewski, 2010). Thus, highly communal team members may decide not to speak up because they fear being blamed as the “troublemaker.” As being friendly and agreeable is central to people with a highly communal orientation, they may refrain from speaking up as they might run at risk of being socially rejected. Moreover, as they are so strongly concerned about others’ feelings, they may also fear that the person they are speaking up to might lose face—especially if that person is a superordinate (Detert & Burris, 2007).

Team members who are highly agentic and less communal might be more willing to take a risk within the team and to challenge others’ viewpoints (cf. Edmondson & McLain Smith, 2006). This behavior might be due to the fact that these people have high self-efficacy beliefs and thus are confident to master a situation that requires speaking up (cf. Bandura, 2001). As these people perceive themselves as highly independent, they might be more willing to stand out from the group and defend their opinion. Moreover, although there might be barriers to speaking up, such as status and power differences (cf. Islam & Zyphur, 2005; Kobayashi et al., 2006), these individuals are less hesitant to express themselves because they might be less concerned about possible interpersonal consequences of speaking up and simply focus on achieving the task at hand. On a related note, we believe that the current results may also have implications for team performance. Although we did not explicitly assess team performance, a failure to speak up in the simulated scenarios implied a threat to patient safety and thus reflects lowered team performance. We therefore suggest that agency and communion are important characteristics to be considered with respect to team composition. The results suggest that when speaking up is required across teams (i.e., against the surgical team), it is least likely when an ACT is composed of team members who are all high in communion and low in agency. ACTs that are composed of members with differential levels of agency and communion can complement each other when across-team speaking up is required.

It is also important to highlight the fact that even though agency and communion are related to gender stereotypes (e.g., Abele, 2003), we did not find gender-related differences in agency and communion and subsequent speaking up behavior. Contrary to studies that demonstrated a link between lower professional status and silence (e.g., Morrison & Rothman, 2009), we did not find significant differences between physicians and nurses regarding their speaking up frequency and type of speaking up. However, we found differences between physicians and nurses in whom they addressed their speaking up. Nurses spoke up more often toward the anesthesia physician rather than toward the (confederate) surgeon. Physicians, on the contrary, addressed their speaking up more often toward the surgeon than to the nurses in their team. We suggest two explanations for this difference. One reason may be that physicians are responsible for voicing diagnoses and making decisions, and, therefore, it is more likely that a nurse questions a physician than vice versa. Another explanation may be the relative status of anesthesia physicians within the anesthesia team. As anesthesia physicians have a higher status within their own team than anesthesia nurses, they may feel—if they are highly agentic and rather less communal—that they are more powerful and thus speak up against the surgeon (cf. Schmid Mast, 2010). Conversely, nurses may expect their physician teammate to speak up against the surgeon. Drawing on the implications of implicit voice theories—beliefs about when and to whom speaking up is appropriate (Detert & Edmondson, 2011)—nurses might have assumed that the physician should challenge the surgeon as their lower status might not deem it appropriate to speak up against a high status physician. This corresponds to the findings of Bienefeld and Grote (2013) who showed that perceived status within one’s own team (i.e., cabin crew) predicted speaking up across teams (i.e., toward the cockpit crew).

Implications for Team Trainings and Interventions

The central practical implication of this research is to train and empower especially highly communal team members in speaking up. Current simulation-based team trainings such as CRM focus on training assertiveness (e.g., Pian-Smith et al., 2009; Salas & Cannon-Bowers, 2001; Smith-Jentsch, Salas, & Baker, 1996). This study suggests that, as part of these trainings, for highly communal individuals, it might be more effective to reduce their concerns regarding a potential negative interpersonal impact of speaking up. Thus, it might be fruitful to assess agency and communion within the context of CRM trainings. In this way, interventions and training modules can be designed and implemented according to participants’ self-perceptions. Especially highly communal individuals might need other strategies and interventions to engage in speaking up. For them, it might be important to establish a sense of psychological safety in the team (Edmondson, 1999, 2003), thus reducing their fear of negative consequences when they voice concerns. If highly communal team members feel that speaking up is valued, especially by their team leaders, they might be more encouraged to voice suggestions and concerns (cf. Grant & Gino, 2010; Nembhard & Edmondson, 2006). Moreover, highly communal team members could be supported in reflecting on and reframing their conceptions of speaking up. This could be done during the debriefing phase after a simulated scenario where it can be systematically explored what kind of consequences and perceived barriers might hinder highly communal team members from voicing concerns (Kolbe, Weiss, Grote et al., 2013).

Although the results point to a beneficial effect of being highly agentic and rather less communal, it is important to discuss potentially negative effects within teams. In the literature, this is referred to as unmitigated agency and refers to negative agentic attributes such as being aggressive or egoistic (Helgeson & Fritz, 1999). Such negative facets of agency might then again hinder other team members from speaking up and compromise effective team functioning. In a study of pilots, it was found that crews led by captains with a positive agentic–communal profile (i.e., operationalized through high achievement motivation and high interpersonal skill) were more effective than crews led by captains with a negative agentic profile (i.e., operationalized through high levels of competitiveness and irritability; Chidester, Kanki, Foushee, Dickinson, & Bowles, 1990). Therefore, it is crucial to provide team members with communication strategies that enable them to voice their concerns assertively but not aggressively (Minehart et al., 2012; Pian-Smith et al., 2009).

Limitations and Future Research

We investigated the relationship between agency and communion and speaking up in ACTs in a simulated setting. Thus, some limitations of this research need to be discussed. On one hand, simulation provides a useful tool as critical events can be created with social and task aspects being held constant. On the other hand, simulation is limited in terms of realism of the context and behavior of the participants (e.g., Gaba, 2007). People may behave differently during simulation than during real clinical performance. In our study, participants were aware of being observed and videotaped during the simulated critical events and might have acted in a socially desirable manner. Moreover, the lack of a “real patient” might have affected those team members with a more communal orientation. It might be plausible that in a real clinical setting, they would have spoken up as they would feel empathic toward the patient. To further increase external validity of our findings, we suggest that the relationships between agency and communion and speaking up may also be studied by implementing live observations in the OR.

Another restriction results from the sample. While the anesthesia nurses were representative with respect to work experience, almost all physicians (with the exception of one attending) were anesthesia residents who still undergo specialist training and are supervised by attending physicians. This circumstance and the fact that they might lack medical knowledge and work experience might also account for their relatively low rate of speaking up (cf. Minei & Bisel, 2013). Therefore, future research needs to investigate these results in larger and more diverse samples (e.g., health care professionals from all levels of work experience).

Moreover, the impact of agency and communion on speaking up should also be investigated in other action teams such as cockpit crews, firefighters, or air traffic control teams. In each of these action team contexts, task requirements may change dynamically and safe procedures need to be secured at all times. To voice the slightest doubt and overcome possible interpersonal hurdles is key to providing high reliability.

Conclusion

This study was conducted to investigate how self-perceptions influence speaking up behavior in ACTs. Our findings highlight that agency (i.e., assertiveness, persistence, independence) and communion (i.e., helpfulness, friendliness, sociability) as two fundamental dimensions of self-perception were significant predictors of actual speaking up behavior during simulated critical events. Specifically, agency positively predicted speaking up and communion negatively predicted speaking up. We conclude from our findings that ACT members need to be highly assertive, independent, and self-confident, and rather less concerned about their teammate’s reactions to speak up in critical situations.

Footnotes

Acknowledgements

We thank Carl Schick for his valuable comments on earlier versions of this article as well as Sonja Mötteli for her help with behavioral coding.

Authors’ Note

Ethics approval was obtained from the local ethics committee.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Swiss National Science Foundation (Grant Number 100014_138545/1).

Notes

Author Biographies

Mona Weiss is a doctoral candidate at ETH Zurich, Switzerland. Her main research interests include self- and social perception, intergroup relations, and voice in health care teams.

Michaela Kolbe, PhD, is a senior research associate and lecturer at ETH Zurich, Switzerland. Her research interests include speaking up, interaction processes in acute care teams, and debriefing techniques for simulation-based team trainings.

Gudela Grote, PhD, is a full professor of work and organizational psychology at ETH Zurich, Switzerland. Her research interests include leadership, teamwork, and standardization in high-risk systems, career development, and effects of new technologies on work processes.

Micha Dambach, MD, is a consultant anesthesiologist at the Institute of Anesthesiology of the University Hospital Zurich, Switzerland, and is specialized in emergency medicine and simulation.

Adrian Marty, MD, is a consultant anesthesiologist at the Institute of Anesthesiology and the director of the Simulation Center of the University Hospital Zurich, Switzerland.

Donat R. Spahn, MD, FRCA, is a professor and chairman of the Institute of Anesthesiology of the University Hospital Zurich, Switzerland. He completed his training at the University of Zurich and at the Duke University Medical Center in Durham, NC, USA.

Bastian Grande, MD, is a consultant anesthesiologist at the Institute of Anesthesiology of the University Hospital Zurich, Switzerland, where he developed a simulation-based team-training program with a focus on assertiveness interventions.

References

Abele

A. E.

(2003). The dynamics of masculine-agentic and feminine-communal traits: Findings from a prospective study. Journal of Personality and Social Psychology, 85, 768-776. doi:10.1037/0022-3514.85.4.768

Abele

A. E.

Wojciszke

(2007). Agency and communion from the perspective of self versus others. Journal of Personality and Social Psychology, 93, 751-763. doi:10.1037/0022-3514.93.5.751

Apfelbaum

Hagberg

Caplan

(2013). Practice guidelines for management of the difficult airway. Anesthesiology, 118, 251-270. doi:10.1097/ALN.0b013e31827773b2

Baard

S. K.

Rench

T. A.

Kozlowski

S. W. J.

(2014). Performance adaptation: A theoretical integration and review. Journal of Management, 40, 48-99. doi:10.1177/0149206313488210

Bakan

(1966). The duality of human existence: An essay on psychology and religion. Chicago, IL: Rand McNally.

Bandura

(2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1-26. doi:10.1146/annurev.psych.52.1.1

Bem

S. L.

(1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42, 155-162.

Bienefeld

Grote

(2012). Silence that may kill: When aircrew members don’t speak up and why. Aviation Psychology and Applied Human Factors, 2, 1-10. doi:10.1027/2192-0923/a000021

Bienefeld

Grote

(2013). Speaking up in ad hoc multiteam systems: Individual-level effects of psychological safety, status, and leadership within and across teams. European Journal of Work and Organizational Psychology. doi:10.1080/1359432X.2013.808398

10.

Bisel

R. S.

Messersmith

A. S.

Kelley

K. M.

(2012). Supervisor–subordinate communication: Hierarchical mum effect meets organizational learning. Journal of Business Communication, 49, 128-147. doi:10.1177/0021943612436972

11.

Bromiley

(2008). Have you ever made a mistake. Bulletin of the Royal College of Anaesthetists, 48, 2442-2445. doi:10.1016/j.ptsp.2011.03.003

12.

Chidester

T. R.

Kanki

B. G.

Foushee

H. C.

Dickinson

C. L.

Bowles

S. V.

(1990). Personality factors in flight operations: Volume I. Leader characteristics and crew performance in a full-mission air transport simulation. Mofett Field, CA: NASA, Ames Research Center.

13.

Detert

J. R.

Burris

E. R.

(2007). Leadership behavior and employee voice: Is the door really open? Academy of Management Journal, 50, 869-884. doi:10.5465/AMJ.2007.26279183

14.

Detert

J. R.

Edmondson

A. C.

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

15.

Digman

J. M.

(1997). Higher-order factors of the Big Five. Journal of Personality and Social Psychology, 73, 1246-1256. doi:10.1037/0022-3514.73.6.1246

16.

Edmondson

A. C.

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

17.

Edmondson

A. C.

(2003). Speaking up in the operating room: How team leaders promote learning in interdisciplinary action teams. Journal of Management Studies, 40, 1419-1452. doi:10.1111/1467-6486.00386

18.

Edmondson

A. C.

(2012). Teaming: How organizations learn, innovate, and compete in the knowledge economy. San Francisco, CA: Jossey-Bass.

19.

Edmondson

A. C.

McLain Smith

(2006). Too hot to handle? How to manage relationship conflict. California Management Review, 49, 6-31.

20.

Fiske

S. T.

Cuddy

A. J. C.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878-902. doi:10.1037//0022-3514.82.6.878

21.

Gaba

D. M.

(2007). The future vision of simulation in healthcare. Simulation in Healthcare, 2, 126-135. doi:10.1097/01.SIH.0000258411.38212.32

22.

Grant

A. M.

Gino

(2010). A little thanks goes a long way: Explaining why gratitude expressions motivate prosocial behavior. Journal of Personality and Social Psychology, 98, 946-955. doi:10.1037/a0017935

23.

Grant

A. M.

Wrzesniewski

(2010). I won’t let you down . . . or will I? Core self-evaluations, other-orientation, anticipated guilt and gratitude, and job performance. Journal of Applied Psychology, 95, 108-121. doi:10.1037/a0017974

24.

Helgeson

Fritz

(1999). Unmitigated agency and unmitigated communion: Distinctions from agency and communion. Journal of Research in Personality, 158, 131-158. doi:10.1006/jrpe.1999.2241

25.

Ishak

A. W.

Ballard

D. I.

(2012). Time to re-group: A typology and nested phase model for action teams. Small Group Research, 43, 3-29. doi:10.1177/1046496411425250

26.

Islam

Zyphur

M. J.

(2005). Power, voice, and hierarchy: Exploring the antecedents of speaking up in groups. Group Dynamics: Theory, Research, and Practice, 9, 93-103. doi:10.1037/1089-2699.9.2.93

27.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). Dyadic data analysis. New York, NY: Guilford Press.

28.

Kobayashi

Pian-Smith

Sato

Sawa

Takeshita

Raemer

(2006). A cross-cultural survey of residents’ perceived barriers in questioning/challenging authority. BMJ Quality & Safety, 15, 277-283. doi:10.1136/qshc.2005.017368

29.

Kolbe

Burtscher

M. J.

Manser

(2013). Co-ACT—A framework for observing coordination behavior in acute care teams. BMJ Quality & Safety, 22, 596-605. doi:10.1136/bmjqs-2012-001319

30.

Kolbe

Burtscher

M. J.

Wacker

Grande

Nohynkova

Manser

. . . Grote

(2012). Speaking-up is related to better team performance in simulated anesthesia inductions: An observational study. Anesthesia & Analgesia, 115, 1099-1108. doi:10.1213/ANE.0b013e318269cd3

31.

Kolbe

Weiss

Grote

Knauth

Dambach

Spahn

D. R.

Grande

(2013). TeamGAINS: A tool for structured debriefings for simulation-based team trainings. BMJ Quality & Safety, 22, 541-553. doi:10.1136/bmjqs-2012-000917

32.

Leaper

(1987). Agency, communion, and gender as predictors of communication style and being liked in adult male-female dyads. Sex Roles, 16, 137-149.

33.

Leigh-Smith

Harris

(2005). Tension pneumothorax—Time for a re-think? Emergency Medicine Journal, 22, 8-16. doi:10.1136/emj.2003.010421

34.

LePine

J. A.

Van Dyne

(2001). Voice and cooperative behavior as contrasting forms of contextual performance: Evidence of differential relationships with big five personality characteristics and cognitive ability. Journal of Applied Psychology, 86, 326-336. doi:10.1037/0021-9010.86.2.326

35.

Mathieu

J. E.

Marks

M. A.

Zaccaro

S. J.

(2001). Multi-team systems. In Anderson

Ones

D. S.

Sinangil

H. K.

Viswesvaran

(Eds.), Handbook of industrial, work and organizational psychology (Vol. 1, pp. 289-313). London, England: SAGE.

36.

Milliken

F. J.

Morrison

E. W.

Hewlin

P. F.

(2003). An exploratory study of employee silence: Issues that employees don’t communicate upward and why. Journal of Management Studies, 40, 1453-1476. doi:10.1111/1467-6486.00387

37.

Minehart

R. D.

Pian-Smith

M. C. M.

Walzer

T. B.

Gardner

Rudolph

J. W.

Simon

Raemer

D. B.

(2012). Speaking across the drapes: Communication strategies of anesthesiologists and obstetricians during a simulated maternal crisis. Simulation in Healthcare, 7, 166-170. doi:10.1097/SIH.0b013e31824e73fb

38.

Minei

Bisel

(2013). Negotiating the meaning of team expertise: A firefighter team’s epistemic denial. Small Group Research, 44, 7-32. doi:10.1177/1046496412467830

39.

Morrison

E. W.

(2011). Employee voice behavior: Integration and directions for future research. The Academy of Management Annals, 5, 373-412. doi:10.1080/19416520.2011.574506

40.

Morrison

E. W.

Milliken

F. J.

(2000). Organizational silence: A barrier to change and development in a pluralistic world. Academy of Management Review, 25, 706-725.

41.

Morrison

E. W.

Rothman

N. B.

(2009). Silence and the dynamics of power. In Greenberg

Edwards

(Eds.), Voice and silence in organizations (pp. 175-202). Bingley, UK: Emerald.

42.

Nembhard

I. M.

Edmondson

A. C.

(2006). Making it safe: The effects of leader inclusiveness and professional status on psychological safety and improvement efforts in health care teams. Journal of Organizational Behavior, 27, 941-966. doi:10.1002/job

43.

Pian-Smith

Simon

Minehart

Podraza

Rudolph

Walzer

Raemer

(2009). Teaching residents the two-challenge rule: A simulation-based approach to improve education and patient safety. Simulation in Healthcare, 4, 84-91.

44.

Premeaux

S. F.

Bedeian

A. G.

(2003). Breaking the silence: The moderating effects of self-monitoring in predicting speaking up in the workplace. Journal of Management Studies, 40, 1537-1562. doi:10.1111/1467-6486.00390

45.

Runge

T. E.

Frey

Gollwitzer

P. M.

Helmreich

R. L.

Spence

J. T.

(1981). Masculine (instrumental) and feminine (expressive) traits: A comparison between students in the United States and West Germany. Journal of Cross-Cultural Psychology, 12, 142-162. doi:10.1177/0022022181122002

46.

Salas

Cannon-Bowers

J. A.

(2001). The science of training: A decade of progress. Annual Review of Psychology, 52, 471-499.

47.

Schmid Mast

(2010). Interpersonal behaviour and social perception in a hierarchy: The interpersonal power and behaviour model. European Review of Social Psychology, 21, 1-33. doi:10.1080/10463283.2010.486942

48.

Sexton

Thomas

Helmreich

(2000). Error, stress, and teamwork in medicine and aviation: Cross sectional surveys. British Medical Journal, 320, 745-749. doi:10.1136/bmj.320.7237.745

49.

Smith-Jentsch

K. A.

Salas

Baker

D. P.

(1996). Training team performance-related assertiveness. Personnel Psychology, 49, 909-936. doi:10.1111/j.1744-6570.1996.tb02454.x

50.

Spence

J. T.

Helmreich

R. L.

(1978). Masculinity & femininity: Their psychological dimensions, correlates, and antecedents. Austin: University of Texas Press.

51.

Spence

J. T.

Helmreich

R. L.

Stapp

(1974). The Personal Attributes Questionnaire: A measure of sex-role stereotypes and masculinity–femininity. Catalog of Selected Documents in Psychology, 4, 43-44.

52.

Streiff

Tschan

Hunziker

Buehlmann

Semmer

N. K.

Hunziker

Marsch

(2011). Leadership in medical emergencies depends on gender and personality. Simulation in Healthcare, 6, 78-83. doi:10.1097/SIH.0b013e318209382b

53.

Sundstrom

de Meuse

K. P.

Futrell

(1990). Work teams: Applications and effectiveness. American Psychologist, 45, 120-133. doi:10.1037//0003-066X.45.2.120

54.

Tschan

Semmer

N. K.

Gurtner

Bizzari

Spychiger

Breuer

Marsch

S. U.

(2009). Explicit reasoning, confirmation bias, and illusory transactive memory: A simulation study of group medical decision making. Small Group Research, 40, 271-300. doi:10.1177/1046496409332928