The person in context: Using situational judgment to gain insight into social functioning

Abstract

How people interpret and respond to simulated interpersonal events may provide insight into their values, beliefs, and personality. In pursuit of this possibility, a three-study program of research (total N = 576) applied a situational judgment method to the domain of friendship behavior. Specifically, consensus scoring techniques were used to scale the effectiveness or ineffectiveness of different ways of responding to vignettes involving protagonists and their friends. These norms were then used to score a dimension of friendship competence that was defined in terms of endorsing ways of responding (to friendship scenarios) that tend to be effective. Study 1 established the reliability of the relevant individual differences, Study 2 showed that they were related to both agreeableness and extraversion, and Study 3 showed that they predicted friendship quality as well as prosocial and antisocial behavior. The research provides novel insights into key personality processes that underlie social behavior.

Keywords

Behavioral tendencies friendships personality situational judgment

Good, high-quality friendships contribute to both health and happiness. Along these lines, for example, Myers (2000) concluded that friendships, as well as social relationships more generally, are among the more important predictors of subjective well-being and high-quality friendships also play an important role in preserving and maintaining health, partly through social support processes (Feeney & Collins, 2015). Of relevance to the present studies, furthermore, there are pronounced individual differences in the number and quality of friendships that people have and such variations can be linked to constructs such as social competence (Crick & Dodge, 1994). Whether due to motivational factors, lack of skill, or both, some people simply cannot form and maintain high-quality friendships to the same degree as others (Fehr, 1996). The present research seeks to better understand such variations.

In the realm of social behavior, as in other realms in which individual differences figure prominently, our primary assessment approach relies on self-reports of personality traits. That is, people have been asked to report on their social qualities, as well as their success in social relationships, and scales of this type have been shown to possess value in predicting relevant outcomes (e.g., Harris & Vazire, 2016). Nonetheless, such reports are influenced by social desirability and it is also apparent that some people lack insight into how socially competent they actually are (Dunning, Heath, & Suls, 2004). Accordingly, there is the need for other approaches to social competence that do not rely on self-ascriptions of ability (Crick & Dodge, 1994). In the present research, we sought to determine whether a new approach to social competence—one based on situational judgments—could fill this important gap in the literature.

An ability-based conception of social competence

In his fascinating analysis of the self, William James (1890) contrasted the “I” with the “me.” The I is the online component of the self—it perceives the world and interacts with it, but cannot be directly observed. The me, by contrast, consists of the various concepts and beliefs that one has about oneself. These are generalizations and they are relatively fixed (James, 1890). In personality psychology, it could be argued, our assessment methods have primarily focused on the “me.” People are asked to reflect on themselves and make generalizations about their feelings, thoughts, or behaviors. The extent to which these generalizations capture online, more dynamic aspects of personality is either not certain or controversial (Pervin, 1994).

Indeed, there are some prominent aspects to the “I” self (James, 1890) that might be missed by trait-based measurement techniques. The I self is always embedded within a situational context rather than abstracted from it. And the I self is essentially an agent rather than a target of introspection. Assessments of this component of the self—the actor—would seem to require a method that could place people in a situational context (Shoda & Mischel, 2000) without asking them to ascribe personality traits to themselves (Pervin, 1994).

The situational judgment test (SJT) possesses relevant procedures. Beginning in the 1920s, people have been given written, vignette-based tests that (a) place them in a particular situational context and (b) ask them how they should or would respond in the described situation (Whetzel & McDaniel, 2009). The “would” instructions—which are sometimes called behavioral tendency (BT) instructions—are particularly interesting because they essentially model the person in action, from presenting situation to deliberation to behavioral response.

These sorts of tests are typically used to probe for job-related knowledge, skills, or abilities (KSAs). For example, an aspiring manager might be asked what he or she would do if a particular piece of equipment was malfunctioning or if an important customer was dissatisfied with services. Applicants or employees who provide more “correct” answers—as defined by subject-matter experts—are typically better employees, with predictive validities in the .3 range (Whetzel & McDaniel, 2009). Of further note, SJT/outcome relationships often remain significant when controlling for other variables such as personality traits, cognitive ability, or job experience (Corstjens, Lievens, & Krumm, 2017).

Although the SJT method has primarily been applied to workplace functioning, it could lend itself to other, broader foci as well. In support of this possibility, a content analysis of extant SJTs revealed that many of them seemed to target interpersonal skills as well as, or instead of, technical features of the job (Christian, Edwards, & Bradley, 2010). In addition, people’s ideas about effective responding overlap with their ideas about personality—for example, more conscientious ways of responding are often deemed more effective (Corstjens et al., 2017). Finally, there is some evidence that one can target particular sorts of personality tendencies—such as self-consciousness or compliance (Mussel, Gatzka, & Hewig, 2018)—through the use of carefully constructed SJT tests (Motowidlo & Beier, 2010).

In the present investigation, we sought to target friendship competence (FC), a type of social competence. The developmental literature has long proposed that individuals differ in their abilities to make friends, keep friends, and respond to potential conflicts with skillful means (Rose-Krasnor, 1997). Moreover, many of these differences are apparent in social cognitive processes such as situation interpretation and response selection. Children who are prone to reactive aggression, for example, tend to interpret ambiguous provocations in hostility-related terms, and this hostile attribution bias contributes to their relationship problems with other children (Crick & Dodge, 1994). If enough of these information processing mechanisms are packaged, they can begin to explain the divergent pathways of children who do, versus do not, become successful social beings (Dodge & Crick, 1990).

Very little of this information processing tradition has infused the adulthood literature on friendship formation and maintenance, which is also fairly sparse. When researchers have considered the topic, they have typically created self-report scales of social competence or they have inferred it from existing self-report scales (Gurtman, 1999). This approach may be problematic because people are motivated to think of themselves in socially competent terms, which influences what they say about their social skills (Dunning et al., 2004). Furthermore, social feedback is often ambiguous and people who lack social competence may not realize just how deficient they are (Dunning, Johnson, Ehrlinger, & Kruger, 2003). In either case, it makes sense to explore other approaches to social competence assessment, such as that which may be possible using an SJT method.

Following this line of thinking, we drafted numerous scenarios involving realistic, but nontrivial, interactions between protagonists and their friends. Participants in the initial stages of research were asked to rate how effective it would be to respond to each situation (e.g., a friend losing a job) in four different ways (e.g., loaning rent money, helping the friend find another job, etc.). By examining the convergence and divergence of these answers, we could develop a scoring key that scaled particular ways of responding in terms of how effective they are, as defined by the “wisdom of the crowd” or consensus-based scoring norms (Legree, Psotka, Tremble, & Bourne, 2005). Briefly, although particular answers tend to be error-prone, aggregated answers from a larger number of respondents tend to converge on correct answers, as determined by experts or objective criteria (Barchard, Hensley, & Anderson, 2013). This approach seemed to be particularly suited to the friendship realm, which has implicit rules or guidelines that can be identified with enough raters (Argyle & Henderson, 1984).

To study the tendencies of the social self, we asked our primary participants how they would respond to each scenario, essentially placing the self in a variety of social situations (Shoda & Mischel, 2000). These answers were then scored using the effectiveness norms that had been collected earlier. Essentially, we could quantify a version of social competence that defined it in terms of whether the self’s responses to friendship-related scenarios tended to be aligned with how effective those actions would be (high social competence) or not (low social competence). The relevant processes involve simulation rather than reality, behavioral intentions rather than behavior, and consensus norms rather than expert opinion. Nonetheless, the resultant individual differences are likely to be consequential. We conducted a three-study, five-sample program of research to investigate such processes.

Study 1 had two phases. We first developed the measurement approach and then scored a new set of participants in terms of their FC levels, using an SJT-inspired method. The domain of friendships is one in which prosocial attitudes, beliefs, and behaviors tend to be rewarded (Rose-Krasnor, 1997). We therefore expected high-scoring individuals, relative to low-scoring individuals, to have more favorable attitudes about helping, and we hypothesized that these individuals would experience greater social support. Study 2 then focused on the personality profile of FC. Among other predictions, we expected our newly developed test to align itself with the warm-dominant vector of the interpersonal circumplex (Gurtman, 1999). Study 3, finally, had many purposes but a central one was to better understand the behavioral correlates of FC. FC, it seemed to us, should correlate positively with prosocial behaviors and negatively with antisocial behaviors (Crick & Dodge, 1994). Altogether, the program of research concerned itself with a new approach to FC in particular and personality assessment more generally.

Study 1: Method

Designing scenarios and obtaining effectiveness norms

The first step of the research program involved writing scenarios and obtaining effective norms. Five social scientists—two of whom have PhDs and two of whom have MBAs—wrote 65 scenarios involving a protagonist deciding how to respond to the actions or inactions of a generically named “friend.” The relevant interaction episodes were realistic as well as potentially consequential to the relationship and they centered on themes such as conflict, uncertainty, and/or opportunities for intimacy development (Fehr, 1996). We limited scenario descriptions to one to two sentences so that they were easy to comprehend while avoiding unnecessary detail (Corstjens et al., 2017) and paired them with four potential ways of responding. Collectively, these ways of responding were designed to cover the full range of things a person might do in the context of a relationship challenge, threat, or ambiguity (Fehr, 1996). Accordingly, the options often involved direct actions, but sometimes involved avoidance, information-seeking, or emotion-regulation attempts, and so on (Crick & Dodge, 1994). The ways of responding were diverse and plausible, yet varied in their likely effectiveness. With respect to the diversity criterion, we avoided any specific model in generating the ways of responding, but did consult the circumplex (Gurtman, 2009) and five-factor model (John & Srivastava, 1999) as reference points.

We then administered the measure to 102 undergraduate students (53.92% female; 92.18% Caucasian; M age = 19.32) from a Midwestern University, who received course credit. Participants were told that they would read multiple situations involving named protagonists (e.g., Jack, Jade) and their friends. For each way of responding to each of the 65 scenarios, participants were asked to rate how effective it would be (1 = Not at all effective; 5 = Very effective) to respond in the indicated manner (Corstjens et al., 2017). These ratings averaged 3.09 and also varied across the ways of responding in the desired manner (SD = 0.95).

Most of the ways of responding were neither “right” nor “wrong” in their entirety, as is true for social behaviors more generally (Dirks, Treat, & Weersing, 2007). To develop a scoring key with the desired degree of nuance, we borrowed procedures used in the emotional intelligence literature (e.g., Mayer, Salovey, Caruso, & Sitarenios, 2003). Specifically, for each way of responding to each situation, we computed the percentage of respondents who gave a 1, 2, 3, 4, or 5 rating along the 5-point effectiveness scale. These percentages will bunch toward the lower side of the scale for less effective ways of responding and toward the higher side for more effective ways (Legree et al., 2005; Mayer et al., 2003).

For item selection purposes, each rating was then scored in terms of the extent to which it matched consensus norms (e.g., if a given rating agreed with 28.72% of the sample as a whole, the person would receive a score of .2872 for that rating; Mayer et al., 2003). These percentage-based agreement scores were then averaged across ways of responding for a given scenario and then across scenarios to constitute a total score. Importantly, the scenarios appeared to tap a common set of abilities in that all item-total correlations were positive (M = 0.38; SD = .13). Nonetheless, some scenarios would almost necessarily be better than others and we wanted to select these particularly good ones. We therefore selected the 10 scenarios that had the highest item-total correlations (ranging from .48 to .61) for the next stage of the research. Please note that the 10 scenarios corresponded with 40 effectiveness ratings, which seemed a good basis for scaling the abilities of interest. The relevant scenarios involved responding to a betrayal (Tracie), a sick friend (Victor), a long-lost friend (Darcy), a touching comment (Wilma), a competition (Jack), an imposition (Scott), being ignored (Randy), a blameworthy action (Lesley), an inappropriate behavior (Roy), and troubles with a group project (Jade).

Behavioral tendencies toward FC

We sought to understand the tendencies of the social self, but in a way that would speak to its competence. To meet these goals, we developed a new procedure. Participants would be asked how they, personally, would respond to the friendship situations that were developed. These responses, though, would be scored in terms of whether the relevant ways of responding tend to be effective or not, as captured by norms. For this phase of the project, we recruited 64 undergraduates (66.67% female; 85.71% Caucasian; M age = 19.97) from the same university that had produced the effectiveness norms, which gave us adequate (.80) power to detect correlations in the medium range (.3).

Participants were asked to place themselves in each protagonist’s situation and then rate how likely it would be (1 = Not at all likely; 5 = Extremely likely) that they would respond in each of the 4 indicated ways. Through these person-in-situation ratings, we sought to capture the strategies and tendencies of the social self, distinctly so for each participant. These self-likelihood ratings were then scored using the effectiveness norms from phase 1 of the study. Table 1 gives a detailed description of how, concretely, this scoring system works. In response to a friend who seems to be neglecting his/her component of a group project, our hypothetical participant indicates that she would be very likely (five rating) to explain the importance of the project. Explaining the importance of the project was also deemed to be a very effective way of responding by the initial norming sample and the participant would receive a score of .5686 for this rating because 56.86% of the original sample gave a five rating for effectiveness.

Table 1.
Example scenario, effectiveness norms, self-likelihood ratings, and their scoring.

Scenario: Jade will get a poor grade on a joint project if her friend, who is also involved in the project, does not start doing something.

Ways of responding: (i) prioritize the friendship over the project, (ii) prod the friend until he or she does something, (iii) explain the importance of the project, and (iv) hope the friend gets started soon

Question: If you were in the situation, how likely would it be that YOU would do the following? 1 = Not at all likely, 5 = Very likely

Effectiveness norms (%)

Way 1 (%) 2 (%) 3 (%) 4 (%) 5 (%) Self-likelihood Score

(i) Prioritize 33 37 18 08 04 3 .18

(ii) Prod 06 12 32 32 18 2 .12

(iii) Explain 00 01 08 34 57 5 .57

(iv) Hope 25 31 12 13 19 2 .31

Note. FC = friendship competence. The hypothetical participant made self-likelihood ratings of 3, 2, 5, and 2 and would receive a scenario-specific FC score of .2950 (the average of .18, .12, .57, and .31).

On the other hand, the hypothetical participant would under-prod and displays a bit too much of a tendency to prioritize the friendship over the project, which is not very effective in this situation (see Table 1). The participant’s total score for the scenario was therefore somewhat average (.2950). In sum, the scoring system awards participants whose high self-likelihood ratings correspond to effective actions and whose low self-likelihood ratings correspond to ineffective actions, as established by norms. By contrast, a low score would be obtained to the extent that the self’s actions would tend to be ineffectual or problematic.

Nomological network and outcomes

In Study 1, Phase 2, we also assessed a number of outcomes that seemed likely to vary by FC levels. First, establishing high-quality relationships requires that one be able to trust other people (Fehr, 1996). Accordingly, we thought it likely that participants receiving higher FC-BT scores would express higher levels of generalized trust, which was assessed with the altruism subscale of the Philosophy of Human Nature Scale (Wrightsman, 1964; M = 4.11; SD = 0.49; α = .79). In more personal terms, too, we expected FC scores to vary positively with prosocial attitudes and behaviors. Attitudinal processes were assessed with the Attitudes Toward Helping Scale (Nickell, 1998; M = 3.63; SD = 0.61; α = .89) and behavioral tendencies were assessed with the Prosocialness Scale for Adults (Caprara, Steca, Zelli, & Capanna, 2005; M = 3.79; SD = 0.72; α = .93).

If one is able to develop close relationships with others, one should enjoy higher levels of social support as a result (Feeney & Collins, 2015). On the basis of this reasoning, we expected positive relationships between the FC-BT and the levels of social support from friends and family. Such potential relationships were probed with an abbreviated form of the Interpersonal Support Evaluation List (Payne et al., 2012), with separable subscales for appraisal (M = 3.34; SD = 0.53; α = .77), belonging (M = 3.14; SD = 0.63; α = .81), and tangible support (M = 3.25; SD = 0.46; α = .58). In addition, a total score was computed (M = 3.24; SD = 0.48; α = .89).

Although our primary interests involved prosocial tendencies and social support, we thought it possible that individuals with higher FC levels would, because of their greater successes in the interpersonal realm, develop self-perceptions consistent with social competence. To examine potential correlates of this type, we administered the Tromso Social Intelligence Scale (Silvera, Martinussen, & Dahl, 2001), which has separable subscales for awareness (M = 4.71; SD = 1.10; α = .83), social information processing (M = 5.40; SD = 0.92; α = .85), and social skills (M = 4.66; SD = 1.05; α = .79). It seemed to us that participants would have the greatest introspective access concerning their social skills, which are associated with social feedback (Farmer & Chapman, 2016), and we therefore emphasize this scale.

Results

The FC-BT

A person with no systematic knowledge about friendships could expect a score of .2000, matching 20% of the norming sample for each rating. A person with maximal levels of knowledge could expect a score of .4223, which would occur if one’s self-likelihood ratings always matched the greatest percentage of the norming sample. The FC-BT mean was intermediate between these possibilities at .2975. Individual differences were also apparent, as reflected in a standard deviation of .0393. Of further importance, these individual differences were fairly reliable across the 40 items of the test (α = .67).

Correlates and outcomes

We used simple regression procedures to examine possible relationships between the FC-BT and the Study 1 outcomes. As shown in Table 2, higher levels of FC were linked to higher levels of prosociality. Specifically, participants receiving higher FC-BT scores had more benevolent attitudes about others, had more positive attitudes toward helping, and were more frequently prosocial in their everyday lives (Caprara et al., 2005). The posited relationships with social support were similarly robust in that people with higher FC-BT scores felt more positively about their social support networks—for example, as manifest in perceptions that others would provide tangible support when needed (Payne et al., 2012). Finally, the performance-based measure of FC was positively linked to self-perceptions of social skills, though not to self-perceptions of social awareness or information processing.

Table 2.
Relationships between the FC-BT and the Study 1 outcomes (simple regressions).

Category

Measure t p β

Prosocial tendencies

Altruistic philosophy 2.99 .004 .39

Helping attitudes 3.54 <.001 .44

Prosocial behavior 3.96 <.001 .48

Social support

Appraisal 3.71 <.001 .46

Belonging 2.09 .042 .28

Tangible support 2.15 .036 .29

Total 2.97 .005 .38

Social intelligence

Social awareness 1.55 .128 .21

Information processing 1.22 .268 .15

Social skills 2.40 .020 .32

Total 2.22 .031 .30

Note. FC = friendship competence; BT = behavioral tendency.

Sex differences

Women often invest more in their friendships and there are plausible sex differences in the requisite skills (Gurtman, 2009). Consistent with this possibility, a one-way analysis of variance revealed that women (M = 0.3094) tended to receive higher FC-BT scores than men (M = 0.2748), F(1, 62) = 13.39, p < .001, η_p ²= .18. Additional analyses focused on possible gender by FC interactions, using the multiple regression procedures of Aiken and West (1991). Directionally, FC/outcome relationships tended to be somewhat larger in magnitude among men than women (average interactive β = −.11). However, only the interaction involving social awareness was significant, t = −2.15, p = .036, β = −.32. We therefore emphasize similarities by sex.

Discussion

The reliability of the FC-BT compares favorably with other SJT tests, likely because of its consistent focus on friendship-related events and scenarios. Study 1 also began to examine questions about the relevant individual difference continuum. Participants who received higher FC-BT scores had more favorable attitudes about others, about helping, and they engaged in prosocial behaviors more frequently. Perhaps in part because of this helpfulness, participants with higher FC-BT scores enjoyed more social support from others, including in the realms of belonging and instrumental or tangible support. The skills assessed by the FC-BT, though, only partially overlapped with beliefs about social intelligence, which could be erroneous among some people (Dunning et al., 2003). Even so, the FC-BT should exhibit predictable relationships with several components of personality and we focused on this possibility in Study 2.

Study 2

The personality trait of agreeableness is a robust predictor of friendship skills and achievements (Jensen-Campbell, Knack, & Gomez, 2010). Because this is true, we hypothesized that agreeable, relative to disagreeable, participants would obtain FC-BT scores. Competence in the friendship realm also, though, requires some degree of assertiveness (Farmer & Chapman, 2016), and such considerations suggest a possible correlation with the personality trait of extraversion. From a complementary perspective, social competence seems to have features of both warmth and dominance (Gurtman, 1999) and trends of this type were examined with respect to interpersonal circumplex measures. Finally, it is reasonable to think that the KSAs of the FC-BT would not be compatible with trait-related tendencies toward psychopathy (Jones & Paulhus, 2011) and we expected a negative correlation along such lines.

Method

Participants

Participants were 136 undergraduate students (61.61% female; 92.63% Caucasian; M age = 19.16) who received course credit. This sample size gave us adequate power (.80) to detect correlations in the .21 range and more than adequate power (.97) to detect medium effect sizes.

Friendship competence

Participants indicated the likelihood that they would respond to the 10 friendship scenarios of Study 1 in each of 4 ways. Such BT responses were scored using the effectiveness norms of Phase 1 of Study 1, as detailed in Table 1, such that higher numbers were given to participants whose interpersonal behavior should tend to be more effective. The internal consistency of the measure was good (M = 0.2861; SD = .0460; α = .74).

Personality assessments

Extraversion and agreeableness

Extraversion and agreeableness are the two traits of the Big 5 that are particularly interpersonal (Gurtman, 2009). We assessed both of these traits (extraversion: M = 3.37; SD = 0.87; α = .91; agreeableness: M = 4.03; SD = 0.75; α = .89) using the relevant broad-bandwidth scales of Goldberg (1999), which correlate highly with their counterparts from other Big 5 scales (John & Srivastava, 1999).

Interpersonal circumplex

The FC-BT is likely to have prominent projections within the interpersonal circumplex space. To examine possible relationships of this type, we administered the well-validated Revised Interpersonal Adjectives Scales (Wiggins, Trapnell, & Phillips, 1988), which asks participants to rate the extent to which 64 interpersonal adjectives describe them. These ratings allowed us to score the four dimensions of warmth (e.g., “kind”; M = 4.72; SD = 0.82; α = .92), dominance (e.g., “assertive”; M = 4.20; SD = 0.69; α = .81), extraversion (e.g., “jovial”; M = 4.70; SD = 0.77; α = .92), and arrogance (e.g., “boastful”; M = 3.65; SD = 0.69; α = .81).

Dark triad traits

Among the dark triad, psychopathy is particularly hostile (Jones & Paulhus, 2011) and it made sense to posit an inverse relationship between the psychopathy and the FC-BT. We assessed psychopathy, as well as the other two dark triad members, with the Short Dark Triad (Jones & Paulhus, 2014) instrument (narcissism: M = 3.04; SD = 0.50; α = .67; Machiavellianism: M = 3.29; SD = 0.49; α = .60; psychopathy: M = 2.26; SD = 0.55; α = .73).

Results

Simple regression results

Zero-order relationships between the FC-BT and the personality measures were examined in a series of simple regressions. We expected, and found, a fairly pronounced correlation between the FC-BT and the personality trait of agreeableness, t = 5.63, p < .001, β = .44. In addition, the FC-BT varied positively, though to a lesser extent, with extraversion, t = 2.01, p = .046, β = .17. Similar correlates were found with respect to the dimensions of the circumplex. Specifically, participants who obtained higher FC-BT scores were interpersonally warmer, t = 7.24, p < .001, β = .54, and more extraverted, t = 5.65, p < .001, β = .45. They also tended toward dominance rather than submission, t = 2.28, p = .024, β = .20, but the dimension of arrogance was unrelated to FC-BT scores, t = −0.33, p = .740, β = −.03.

Relationships involving the dark triad traits were similarly informative. There was a fairly strong inverse relationship between the FC-BT and the psychopathy, t = −4.27, p < .001, β = −.35. By contrast, the relationship with Machiavellianism was more modest, t = −2.11, p = .037, β = −.18, and there was no relationship with narcissism, t = −0.19, p = .849, β = −.02. These results suggest that impulsive forms of hostility, which are most pronounced in psychopathy (Furnham, Richards, & Paulhus, 2013), are particularly problematic in the FC-BT realm.

Circumplex location

To gain a further understanding of the FC-BT, we sought to map its position within interpersonal space (Gurtman, 2009). Following procedures outlined by Wiggins and Broughton (1991), we first standardized octant scores for the IAS-R. We then calculated factor scores for the warmth and dominance axes and correlated the FC-BT with both sets of factor scores. These computations allowed us to position the FC-BT within the interpersonal space while also speaking to the interpersonal-ness (Gurtman, 2009) of our measure.

The results revealed that the FC-BT can be located in the friendly dominant quadrant of the circumplex (see Figure 1). Consistent with this idea, the FC-BT had an angular displacement of 51.2°, placing it in contact with interpersonal traits such as “friendly,” “jovial,” and “neighborly” (Wiggins et al., 1988). In addition, the relevant vector length was .58, which indicates a good degree of interpersonal saturation (Wiggins & Broughton, 1991).

Figure 1.
Mapping of the FC-BT within the interpersonal circumplex, Study 2. FC = friendship competence; BT = behavioral tendency.

Sex differences

In a replication of Study 1, women (M = 0.2960) received higher FC-BT scores than men (M = 0.2716), F(1, 134) = 9.81, p = .002, η_p ²= .07. Nonetheless, the correlates of the FC-BT were practically identical among men and women, in that gender by FC interactions was not evident within regressions using gender as a moderator (Aiken & West, 1991), all ps > .20.

Discussion

Agreeable (versus disagreeable) people care about their friendships to a greater extent (Jensen-Campbell et al., 2010) and the present results suggest that they are also more competent in responding to situational descriptions of friendship conflict or uncertainty. That the FC-BT also correlated positively with extraversion is consistent with the idea that there are proactive or assertive features to FC as well (Rose-Krasnor, 1997). These personality trends were confirmed in the circumplex correlates for FC: Participants who received high FC-BT scores were warmer, friendlier, and more dominant than participants who received low FC-BT scores. As a final point, the FC-BT seemed to track the “darkness” of the dark triad such that psychopathy was particularly problematic with respect to FC levels. Generally speaking, the results of Study 2 support the idea that FC has an agreeable signature to it, which makes sense (Jensen-Campbell et al., 2010).

Study 3

The results of Studies 1 and 2 suggest that the FC-BT may have broad implications for understanding prosocial and antisocial functioning. Study 3 sought to further develop this case, particularly with respect to behaviorally defined outcomes. In this connection, for example, we hypothesized that FC would correlate positively with prosocial behavior in the dictator game (Engel, 2011). Also, it seemed likely to us that FC would support higher levels of friendship quality, potentially even after controlling for self-reports of agreeableness (Motowidlo & Beier, 2010). Of final interest, Study 3 sought to better understand what sorts of actions tend to be considered effective within the friendship scenarios of the FC-BT. In part based on the findings of Study 2, but also consistent with other lines of data (Crick & Dodge, 1994; Gurtman, 1999), we hypothesized that effective ways of responding would be characterized (by an independent group of participants) as both helpful and practical.

Method

Participants and procedures

The primary participants were 140 undergraduates (65.47% female; 87.86% Caucasian; M age = 18.80) who received course credit for the relevant laboratory experiment. This sample size gave us .98 power to detect medium-sized (.3) correlations.

Friendship competence

Participants read about each friendship situation and rated the likelihood (1 = Not at all likely; 5 = Very likely) that they would respond in 40 different ways. We scored these ratings using the effectiveness norms of Study 1, such that higher FC-BT scores were given to participants whose behavioral tendencies had been deemed to be effective by the initial norming sample (M = 0.3062; SD = .0487; α = .77).

We were also interested in more particular perceptions of the ways of responding. Pursuant of this aim, we recruited an additional 134 undergraduates from the same participant pool as the primary study. These additional participants were asked to rate how practical (1 = Not at all; 5 = Very much so) and helpful (1 = Not at all; 5 = Very much so) the 40 ways of responding were. In general, these ratings suggested that ruminative responses (e.g., “mope about the situation”) were neither practical nor helpful, whereas proactive responses (e.g., “help the friend find another job”) often possessed both qualities.

Friendship quality

Participants with higher levels of FC should have better, higher quality friendships. To examine this possibility, we administered the multidimensional Friendship Quality Scale (Bukowski, Hoza, & Boivin, 1994), which asks people to identify their “best friend” and then rate that friendship according to five subscales. We were primarily interested in the help (M = 4.56; SD = 0.62; α = .62), security (M = 4.41; SD = 0.55; α = .71), and closeness (M = 4.63; SD = 0.52; α = .52) subscales because the companionship subscale (M = 4.01; SD = 0.73; α = .76) suggests elements of co-dependency and the conflict subscale (M = 3.02; SD = 1.03; α = .82) could suggest problems in the relationship.

Prosocial behavior

We assessed broader tendencies toward prosocial behavior in two ways. Participants completed the Rushton altruism scale (Rushton, Chrisjohn, & Fekken, 1981), which focuses on how frequently the person performs concrete acts that are altruistic in nature (M = 2.65; SD = 0.55; α = .84). A second scale (Williams & Anderson, 1991) focused on prosociality in the workplace, in the form of organizational citizenship behaviors (M = 5.37; SD = 0.68; α = .78).

Antisocial behavior

There are reasons for thinking that social competence should correlate negatively with aggression as well as deviance (Crick & Dodge, 1994) and we sought to determine whether such correlates characterize the FC-BT. Participants completed an aggression scale (Orpinas & Frankowski, 2001) that asked them how frequently they had engaged in concrete acts of aggression during the previous week (M = 1.45; SD = 0.72; α = .88). Deviance, which reflects more general tendencies toward antisocial behavior, was assessed with the Normative Deviance Scale (Vazsonyi, Pickering, Junger, & Hessing, 2001). This scale, like the aggression scale, focused on concrete acts (e.g., running stop sign, skipping payments) and the frequency with which the person had done them (M = 1.68; SD = 0.52; α = .76).

Laboratory behavior

Participants completed two laboratory behavior or decision-making measures that sought to capture prosocial tendencies in action. The first focused on charitable giving (Cheung & Chan, 2000) and asked participants to imagine that they had US$20,000 per year that they could spend on whatever they wanted, including themselves. Participants were asked how much of the $20,000 they would donate to nine charitable targets including the arts, community outreach programs, and environmental causes. For this measure, we had to delete the data of 10 participants who did not follow directions or typed in nonnumeric answers. After doing so, the measure had good properties (M = $6,786.44; SD = $5,521.80; α = .74). Participants also played five rounds of the dictator game (Engel, 2011), which asked them how much of US$10 they would split with another person who was dependent on their choices (M = $4.63; SD = $2.29; α = .91).

Responses to provocation

We wanted to model affective and behavioral responses to provocation in a relatively nuanced, contextual way. We did this by presenting five provocation scenarios (e.g., a friend calling them “stupid”) from the Anger Response Inventory (ARI; Tangney et al., 1996). Following a description of each event, participants first indicated how angry they would feel (M = 3.19; SD = 0.72; α = .73). They then reported on the extent to which they would be motivated to get back at the provoker (SD = 0.76; α = .74) and the likelihood that they would engage in two aggressive behaviors (e.g., shove the friend against a wall), which were averaged into a single aggression index (M = 1.74; SD = 0.59; α = .78).

Discriminant validity

Given our focus on prosocial and antisocial outcomes, and given the results of Study 2, it made sense to assess the personality trait of agreeableness as a potential control variable. Accordingly, participants were asked to complete Johnson’s (2014) 24-item International Personality Item Pool (IPIP) agreeableness scale, which was modeled on the IPIP-NEO (Goldberg, 1999; M = 3.91; SD = 0.48; α = .87).

Results

Simple regression results

As shown in Table 3, the FC-BT predicted the Study 3 outcomes in expected ways. Participants with higher FC-BT scores had higher quality friendships, which were marked by greater helping, security, and closeness along with lesser conflict. More broadly, such individuals were more likely to engage in prosocial behaviors such as altruism and they were less likely to enact antisocial behaviors such as aggression or deviance.

Table 3.
Relationships between the FC-BT and the Study 3 outcomes (simple regressions).

Category

Measure t p β

Friendship quality

Helping 2.76 .007 .23

Security 4.16 <.001 .34

Closeness 4.07 <.001 .33

Companionship 1.81 .072 .15

Conflict −2.14 .030 −.18

Prosocial behavior

Altruism 2.80 .006 .24

OCB 4.81 <.001 .38

Antisocial behavior

Aggression −3.77 <.001 −.31

Normative deviance −4.25 <.001 −.35

Laboratory behavior

Charity donations 3.74 <.001 .32

Dictator 3.51 .001 .29

Responses to provocation

Anger −1.72 .089 −.15

Revenge −5.00 <.001 −.39

Aggression −4.76 <.001 −.38

Note: FC = friendship competence; BT = behavioral tendency; OCB = organizational citizenship behavior.

Considering behavior in the laboratory, the FC-BT was a positive predictor of donations to charity and it was a positive predictor of contributions within the dictator game. Similarly, data from the ARI indicated that the FC-BT was inversely related to revenge intentions and aggression in response to simulated provocations. In total, these results highlight the prosocial nature of the FC continuum.

Further analyses of ARI responses

When provocations are unjustified, people may have more control over how they respond to anger than to anger’s occurrence (Tangney et al., 1996). Consistent with this line of thinking, Table 3 shows that the FC-BT had a more inverse relationship with ARI revenge motivation and aggression than it did with anger. In within-subject terms, too, the FC-BT might moderate the relationship between anger and its behavioral expression. To examine this possibility, we ran a multilevel analysis examining reactions to provocation as a function of the within-subject, cross-scenario predictor of anger in concert with the level 2 predictor of FC-BT scores. The anger predictor was person-centered and the FC-BT predictor was z-scored. The dependent variable averaged the revenge and aggression measures, which exhibited similar patterns.

This analysis resulted in main effects for anger, b = .19, t = 6.94, p < .001, and the FC-BT, b = −.23, t = −5.47, p < .001, as well as a cross-level interaction, b = −.06, t = −2.22, p = .027. Figure 2 presents estimated means for the interaction at +1 and −1 SD along each of the predictor dimensions (Aiken & West, 1991). As shown there, reactivity to anger was more pronounced when FC-BT levels were low, b = .26, t = 6.40, p < .001, relative to high, b = .13, t = 3.33, p < .001. Or, expressed in other terms, the skills assessed by the FC-BT seemed to play some role in the self-regulation of anger-motivated behaviors, at least within the ARI platform.

Figure 2.
Reactivity to anger as a function of FC-BT scores, Study 3 (estimated means). FC = friendship competence; BT = behavioral tendency.

Controlling for agreeableness

Study 2 showed that FC was positively correlated with the personality trait of agreeableness and this relationship was replicated in Study 3, r = .37, p < .001. Nevertheless, when controlling for agreeableness, most of the FC/outcome relationships remained significant, as shown in Table 4. For example, people with higher FC-BT scores still perceived their best friendships to be closer and more secure. Also, FC remained a significant predictor of laboratory donations as well as behavioral altruism. The skills and knowledge assessed by the FC-BT are therefore partially independent of agreeableness.

Table 4.
Relationships between the FC-BT and the Study 3 outcomes after controlling for agreeableness (multiple regressions).

Category

Measure t p β

Friendship quality

Helping 2.56 .012 .23

Security 3.10 .002 .26

Closeness 3.77 <.001 .32

Companionship 1.12 .265 .10

Conflict 1.25 .213 .11

Prosocial behavior

Altruism 2.49 .014 .23

OCB 3.45 .001 .28

Antisocial behavior

Aggression −1.69 .094 −.14

Normative deviance −2.98 .003 −.24

Laboratory behavior

Charity donation 3.81 <.001 .34

Dictator 2.34 .021 .20

Responses to provocation

Anger −1.46 .146 −.13

Revenge −3.40 .001 −.27

Aggression −2.98 .003 −.23

Note: FC = friendship competence; BT = behavioral tendency; OCB = organizational citizenship behavior.

Sex differences

Replicating prior studies, women (M = 0.3170) received higher FC-BT scores than men (.2880), F(1, 136) = 12.16, p < .001, η_p ² = .08. In subsequent analyses, we tested for possible gender (−1 = male; +1 = female) by FC interactions using multiple regression procedures (Aiken & West, 1991). Two of the 15 interactions were significant. In these cases, which involved friendship conflict and behavioral aggression, the FC-BT variable was a stronger predictor among males than among females. Even so, 13 of the 15 analyses did not result in significant interactions and the average interaction β was −.03. Overall, then, FC appears to assess similar skills and tendencies among males and females.

What is friendship effectiveness?

People receive high FC-BT scores when their self-likelihood ratings are closely aligned with effective norms—that is, when their likely actions would be effective in the friendship situations. What sorts of actions are effective, though? To answer this question, we created a data set in which the 40 ways of responding were the units of analysis. Each way of responding had an effectiveness mean from Phase 1 of Study 1 as well as helpfulness and practicality means from Phase 2 of Study 3. As hypothesized, effective ways of responding were both practical, r = .91, p < .001, and helpful, r = .90, p < .001, and the same was true in a multiple regression (practical: t = 4.26, p < .001, β = .54; helpful: t = 3.32, p = .002, β = .42).

As an additional way of probing this data set, we sorted the ways of responding by each dimension, which quickly revealed relationships among them. For example, to “mope about the situation” when passed up for a leadership role was deemed unhelpful, not practical, and ineffective. However, there were instances in which the dimensions diverged from each other. In response to a friend who lost his job, loaning rent money was viewed as helpful (in the short-term), but not very effective (in the long-term). Participants with high FC-BT scores would not be particularly attracted to such ways of responding.

Discussion

Study 3 further established the prosocial nature of FC. Participants who received higher FC-BT scores engaged in more frequent acts of altruism in their daily lives and were more generous in their behaviors in the laboratory. By contrast, FC exhibited inverse relationships with aggressive behavior and deviance. The findings, then, are somewhat dramatic in contrasting individuals who receive high (more prosocial) versus low (more antisocial) FC scores. As one additional point, FC correlated positively with the trait of agreeableness, replicating Study 2 findings. Nonetheless, FC/outcome relationships tended to remain significant when controlling for agreeableness. These findings contribute to the idea that the FC measurement approach may capture novel components of personality, components that may be more linked to the “I” than the “me” (James, 1890).

General discussion

We have described a new approach to personality and its assessment. Rather than asking people to ascribe traits to themselves, we presented situations and asked participants how they would respond to them. Rather than scoring these responses in terms of personality traits, we scored them in terms of their competence. That is, the approach focuses on the actions of the self and whether they would tend to be effective or not. Because all of the situations involve friendship-related occurrences and how to handle them, the resulting measure—the FC-BT (which is shorthand for “Friendship Competence-Behavioral Tendency”)—captures a collection of processes that may underlie variations in social competence, at least in the friendship domain.

Somewhat inherent to the SJT method is the idea that there is some degree of situational specificity to the relevant knowledge sources (Whetzel & McDaniel, 2009). In this respect, the assumptions of the method overlap with the social cognitive approach to personality, which similarly advocates for specificity in behavioral profiles (Shoda & Mischel, 2000). As a consequence of situational specificity, among other factors, the internal reliability of SJT tests can be low, typically falling well below values found for self-report scales (Corstjens et al., 2017). With respect to such issues, we note that the FC-BT had alphas of .67, .74, and .77 in Studies 1, 2, and 3, with an average value of .73. These coefficients indicate that there was an important degree of cross-situational consistency to the skills that were modeled in the current studies, justifying reference to a general construct of FC.

Nonetheless, it might be of interest to attempt to parse FC into some component parts. The situations of the FC-BT, for example, could be classified in terms of their degree of ambiguity or in terms of the extent to which they involve unwarranted provocations or not. Some sources of incompetence—such as tendencies toward reactive aggression (Crick & Dodge, 1994)—might be particularly pronounced in the context of provocation. Relatedly, it could be useful to distinguish errors of commission (i.e., endorsing ineffective responses) from errors of omission (i.e., failing to endorse effective responses). Pertinent to this discussion, though, is the fact that ratings of helpfulness and practicality correlated at r = .88 in Study 3. The magnitude of this correlation suggests that forms of competence may typically form a multidimensional whole when assessed in the manner that was pioneered.

Women often invest in their friendships more than men do and may learn the rules of friendship (Argyle & Henderson, 1984) to a greater extent. Consistent with this possibility, there were reliable sex differences on the FC-BT, average η_p ² = .11. This difference, though, exceeded effectiveness knowledge per se, as sex differences in knowledge-based scores (Ployhart & Ehrhart, 2003) were less pronounced in Phase 1 of Study 1. Thus, there are some particular self- or personality-related processes implicated here. Regardless, the correlates of FC were remarkably similar for men and women, as defined by the lack of interactions between FC and sex. We therefore conclude that the processes linking FC to outcomes display generalizability across this subject characteristic.

Prosocial attitudes and behaviors seem to be effective in the friendship domain (Jensen-Campbell et al., 2010) and high FC scorers possessed these characteristics. They had more benevolent ideas about human nature and more positive attitudes toward helping (Study 1). Perhaps as a result of these attitudinal processes, they engaged in prosocial behavior to a greater extent, both in the laboratory and in real life (Study 3), and they had personalities that favored prosocial action (Study 2). In understanding such prosocial tendencies, it is important to note that they occurred within the context of activity and assertiveness rather than passivity. High FC scorers scored higher in extraversion as well as agreeableness and the interpersonal vector for FC was one that blended dominance and warmth (Study 2). Thus, FC (at least as we have investigated it) seems to combine elements of agency and communion. This profile converges with Gurtman’s (1999) theoretical analysis.

The profile for low FC scorers was quite different. These people felt that they lacked interpersonal skills (Study 1) and they were more antagonistic (Study 2). Of particular note, such individuals scored higher in psychopathy, which combines elements of hostility and impulsivity (Furnham et al., 2013). Consistent with this profile, low FC scorers were more likely to enact deviant behaviors and they were more aggressive (Study 3). In this context, too, Study 3 suggested that low FC individuals may be particularly prone to reactive aggression, which is triggered by provocations and the experience of anger (Crick & Dodge, 1994). Processes of this type will tend to be problematic in the friendship domain (Rose-Krasnor, 1997) and low FC scorers reported that their friendships were of lower quality (Study 3) and their social support networks were more meager (Study 1). In sum, we have learned something important about social functioning by probing the manner in which individuals respond to social scenarios.

Questions, limitations, and future directions

One version of the SJT asks people to choose effective ways of responding under the assumption that these choices tell us something about what a person would do in the relevant situations (Corstjens et al., 2017). We amplified the self-relevance of these ratings by asking people what they themselves would do when faced with different dilemmas and challenges. There are reasons to think that the latter instructions are likely to increase predictive validity (Ployhart & Ehrhart, 2003), in part because they place the “I” in the relevant simulations. Nonetheless, there should be moderators of FC/outcome relationships, perhaps related to emotional impulsivity or to the nature of the relationship. We encourage such directions of research as they could provide a better understanding of how social cognitions about situations and behaviors get translated into real-world outcomes (Funder, 2009).

Another relevant SJT distinction contrasts tests that have been designed to tap a specific “construct” or not (Guenole, Chernyshenko, & Weekley, 2017; Lievens, 2017). From one perspective, we think that FC is a construct—one associated with effective warmth or agentic communion (Gurtman, 1999). From another perspective, though, we should admit that our procedures for developing the FC-BT did not favor any particular taxonomy of responding and we did not design the FC-BT to measure only one trait. Doing so would essentially require scaling all response options for one particular trait—such as agreeableness—while also scoring the test with respect to that one particular trait (Guenole et al., 2017). Given our multidimensional view of FC, we would regard such a test as too narrow for our purposes, but there is also value to this line of inquiry (e.g., Mussel et al., 2018).

Nonetheless, there are limitations to our work that should be acknowledged. The consistent focus on one particular population (i.e., college students from the U.S.) allowed us to develop norms and apply them in an integrated manner. Still, this consistent focus meant that we did not explore the relevant processes among other age groups or cultures. In this context, it is worth mentioning that the scenarios of the FC-BT were written in a straightforward manner and in such a way that they should target somewhat universal themes such as trust and uncertainty. Still, it would be interesting to compare FC answers across cultures, as there are likely to be at least subtle differences involved (Argyle & Henderson, 1984).

In concluding the article, we point to some broader contributions and implications. Investigators typically measure social-behavioral tendencies through the use of self-report tests, but such tests clearly have limitations (Dunning et al., 2004). By approaching a similar territory from a different perspective, one that appreciates that there are probably skills and sources of knowledge involved (Fehr, 1996), we were able to show that it is possible to scale people along an FC dimension that has broad implications for friendship functioning, social support, and interpersonal behavior. Through mechanisms of this type, furthermore, FC could relate to important life outcomes such as health and well-being, a topic for future research. Finally, given that the FC-BT focuses on thinking and decision-making within concrete situations, there is the very real possibility that tests like the FC-BT could be used to both assess and remedy certain classes of social deficiency (Crick & Dodge, 1994).

Scenario: Jade will get a poor grade on a joint project if her friend, who is also involved in the project, does not start doing something.
(i) Prioritize	33	37	18	08	04	3	.18
(ii) Prod	06	12	32	32	18	2	.12
(iii) Explain	00	01	08	34	57	5	.57
(iv) Hope	25	31	12	13	19	2	.31

Category
Prosocial tendencies
Altruistic philosophy	2.99	.004	.39
Helping attitudes	3.54	<.001	.44
Prosocial behavior	3.96	<.001	.48
Social support
Appraisal	3.71	<.001	.46
Belonging	2.09	.042	.28
Tangible support	2.15	.036	.29
Total	2.97	.005	.38
Social intelligence
Social awareness	1.55	.128	.21
Information processing	1.22	.268	.15
Social skills	2.40	.020	.32
Total	2.22	.031	.30

Category
Friendship quality
Helping	2.76	.007	.23
Security	4.16	<.001	.34
Closeness	4.07	<.001	.33
Companionship	1.81	.072	.15
Conflict	−2.14	.030	−.18
Prosocial behavior
Altruism	2.80	.006	.24
OCB	4.81	<.001	.38
Antisocial behavior
Aggression	−3.77	<.001	−.31
Normative deviance	−4.25	<.001	−.35
Laboratory behavior
Charity donations	3.74	<.001	.32
Dictator	3.51	.001	.29
Responses to provocation
Anger	−1.72	.089	−.15
Revenge	−5.00	<.001	−.39
Aggression	−4.76	<.001	−.38

Category
Friendship quality
Helping	2.56	.012	.23
Security	3.10	.002	.26
Closeness	3.77	<.001	.32
Companionship	1.12	.265	.10
Conflict	1.25	.213	.11
Prosocial behavior
Altruism	2.49	.014	.23
OCB	3.45	.001	.28
Antisocial behavior
Aggression	−1.69	.094	−.14
Normative deviance	−2.98	.003	−.24
Laboratory behavior
Charity donation	3.81	<.001	.34
Dictator	2.34	.021	.20
Responses to provocation
Anger	−1.46	.146	−.13
Revenge	−3.40	.001	−.27
Aggression	−2.98	.003	−.23

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Michelle R. Persich

Open research statement

As part of IARR’s encouragement of open research practices, the authors have provided the following information: This research was not pre-registered. The data used in the research are available. The data can be obtained by emailing Michelle Persich at michelle.persich@ndsu.edu.

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: SAGE.

Argyle

Henderson

(1984). The rules of friendship. Journal of Social and Personal Relationships, 1, 211–237.

Barchard

K. A.

Hensley

Anderson

(2013). When proportion consensus scoring works. Personality and Individual Differences, 55, 14–18.

Bukowski

W. M.

Hoza

Boivin

(1994). Measuring friendship quality during pre- and early adolescence: The development and psychometric properties of the Friendship Qualities Scale. Journal of Social and Personality Relationships, 11, 471–484.

Caprara

G. V.

Steca

Zelli

Capanna

(2005). A new scale for measuring adults’ prosocialness. European Journal of Psychological Assessment, 21, 77–89.

Cheung

Chan

(2000). Social-cognitive factors of donating money to charity, with special attention to international relief organization. Evaluation and Program Planning, 23, 241–253.

Christian

M. S.

Edwards

B. D.

Bradley

J. C.

(2010). Situational judgment tests: Constructs assessed and a meta-analysis of their criterion-related validities. Personnel Psychology, 63, 83–117.

Corstjens

Lievens

Krumm

(2017). Situational judgement tests for selection. In Goldstein

H. W.

Pulakos

E. D.

Passmore

Semedo

(Eds.), The Wiley Blackwell handbook of the psychology of recruitment, selection and employee retention (pp. 226–246). Hoboken, NJ: Wiley Blackwell.

Crick

N. R.

Dodge

K. A.

(1994). A review and reformulation of social information-processing mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74–101.

10.

Dirks

M. A.

Treat

T. A.

Weersing

V. R.

(2007). Integrating theoretical, measurement, and intervention models of youth social competence. Clinical Psychology Review, 27, 327–347.

11.

Dodge

K. A.

Crick

N. R.

(1990). Social information-processing bases of aggressive behavior in children. Personality and Social Psychology Bulletin, 16, 8–22.

12.

Dunning

Heath

Suls

J. M.

(2004). Flawed self-assessment: Implications for health, education, and the workplace. Psychological Science in the Public Interest, 5, 69–106.

13.

Dunning

Johnson

Ehrlinger

Kruger

(2003). Why people fail to recognize their own incompetence. Current Directions in Psychological Science, 12, 83–87.

14.

Engel

(2011). Dictator games: A meta study. Experimental Economics, 14, 583–610.

15.

Farmer

R. F.

Chapman

A. L.

(2016). Behavioral interventions in cognitive behavior therapy: Practical guidance for putting theory into action (2nd ed.). Washington, DC: American Psychological Association.

16.

Feeney

B. C.

Collins

N. L.

(2015). A new look at social support: A theoretical perspective on thriving through relationships. Personality and Social Psychology Review, 19, 113–147.

17.

Fehr

(1996). Friendship processes. Newbury Park, CA: SAGE.

18.

Funder

D. C.

(2009). Persons, behaviors and situations: An agenda for personality psychology in the postwar era. Journal of Research in Personality, 43, 120–126.

19.

Furnham

Richards

S. C.

Paulhus

D. L.

(2013). The dark triad of personality: A 10 year review. Social and Personality Psychology Compass, 7, 199–216.

20.

Goldberg

L. R.

(1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. In Mervielde

Deary

De Fruyt

Ostendorf

(Eds.), Personality psychology in Europe (Vol. 7, pp. 7–28). Tilburg, The Netherlands: Tilburg University Press.

21.

Guenole

Chernyshenko

O. S.

Weekley

(2017). On designing construct driven situational judgment tests: Some preliminary recommendations. International Journal of Testing, 17, 234–252.

22.

Gurtman

M. B.

(1999). Social competence: An interpersonal analysis and reformulation. European Journal of Psychological Assessment, 15, 233–245.

23.

Gurtman

M. B.

(2009). Exploring personality with the interpersonal circumplex. Social and Personality Psychology Compass, 3, 601–619.

24.

Harris

Vazire

(2016). On friendship development and the Big Five personality traits. Social and Personality Psychology Compass, 10, 647–667.

25.

James

(1890). The principles of psychology. New York, NY: Henry Holt and Company.

26.

Jensen-Campbell

L. A.

Knack

J. M.

Gomez

H. L.

(2010). The psychology of nice people. Social and Personality Psychology Compass, 4, 1042–1056.

27.

John

O. P.

Srivastava

(1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In Pervin

L. A.

John

O. P.

(Eds.), Handbook of personality: Theory and research (pp. 102–138). New York, NY: Guilford Press.

28.

Johnson

J. A.

(2014). Measuring thirty facets of the five factor model with a 120-item public domain inventory: Development of the IPIP-NEO-120. Journal of Research in Personality, 51, 78–89.

29.

Jones

D. N.

Paulhus

D. L.

(2011). Differentiating the Dark Triad within the interpersonal circumplex. In Horowitz

L. M.

Strack

(Eds.), Handbook of interpersonal psychology: Theory, research, assessment, and therapeutic interventions (pp. 249–267). Hoboken, NJ: John Wiley & Sons.

30.

Jones

D. N.

Paulhus

D. L.

(2014). Introducing the Short Dark Triad (SD3): A brief measure of dark personality traits. Assessment, 21, 28–41.

31.

Legree

P. J.

Psotka

Tremble

Bourne

D. R.

(2005). Using consensus based measurement to assess emotional intelligence. In Schulze

Roberts

R. D.

(Eds.), Emotional intelligence: An international handbook (pp. 155–179). Ashland, OH: Hogrefe & Huber.

32.

Lievens

(2017). Construct-driven SJTs: Toward an agenda for future research. International Journal of Testing, 17, 269–276.

33.

Mayer

J. D.

Salovey

Caruso

D. R.

Sitarenios

(2003). Measuring emotional intelligence with the MSCEIT V2.0. Emotion, 3, 97–105.

34.

Motowidlo

S. J.

Beier

M. E.

(2010). Differentiating specific job knowledge from implicit trait policies in procedural knowledge measured by a situational judgment test. Journal of Applied Psychology, 95, 321–333.

35.

Mussel

Gatzka

Hewig

(2018). Situational judgment tests as an alternative measure for personality assessment. European Journal of Psychological Assessment, 34, 328–335.

36.

Myers

D. G.

(2000). The funds, friends, and faith of happy people. American Psychologist, 55, 56–67.

37.

Nickell

(1998). The helping attitudes scale. Paper presented at the 106th annual convention of the American Psychological Association, San Francisco, CA.

38.

Orpinas

Frankowski

(2001). The Aggression Scale: A self-report measure of aggressive behavior for young adolescents. The Journal of Early Adolescence, 21, 50–67.

39.

Payne

T. J.

Andrew

Butler

K. R.

Wyatt

S. B.

Dubbert

P. M.

Mosley

T. H.

(2012). Psychometric evaluation of the interpersonal support evaluation list-short form in the ARIC study cohort. SAGE Open, 2, 1–8.

40.

Pervin

L. A.

(1994). A critical analysis of current trait theory. Psychological Inquiry, 5, 103–113.

41.

Ployhart

R. E.

Ehrhart

M. G.

(2003). Be careful what you ask for: Effects of response instructions on the construct validity and reliability of situational judgment tests. International Journal of Selection and Assessment, 11, 1–16.

42.

Rose-Krasnor

(1997). The nature of social competence: A theoretical review. Social Development, 6, 111–135.

43.

Rushton

J. P.

Chrisjohn

R. D.

Fekken

G. C.

(1981). The altruistic personality and the self-report altruism scale. Personality and Individual Differences, 2, 293–302.

44.

Shoda

Mischel

(2000). Reconciling contextualism with the core assumptions of personality psychology. European Journal of Personality, 14, 407–428.

45.

Silvera

D. H.

Martinussen

Dahl

T. I.

(2001). The Tromso social intelligence scale, a self-report measure of social intelligence. Scandinavian Journal of Psychology, 42, 313–319.

46.

Tangney

J. P.

Hill-Barlow

Wagner

P. E.

Marschall

D. E.

Borenstein

J. K.

Sanftner

… Gramzow

(1996). Assessing individual differences in constructive versus destructive responses to anger across the lifespan. Journal of Personality and Social Psychology, 70, 780–796.

47.

Vazsonyi

A. T.

Pickering

L. E.

Junger

Hessing

(2001). An empirical test of a general theory of crime: A four-nation comparative study of self-control and the prediction of deviance. Journal of Research in Crime and Delinquency, 38, 91–131.

48.

Whetzel

D. L.

McDaniel

M. A.

(2009). Situational judgment tests: An overview of current research. Human Research Management, 19, 188–202.

49.

Wiggins

J. S.

Broughton

(1991). A geometric taxonomy of personality scales. European Journal of Personality, 5, 343–365.

50.

Wiggins

J. S.

Trapnell

Phillips

(1988). Psychometric and geometric characteristics of the Revised Interpersonal Adjective Scales (IAS-R). Multivariate Behavioral Research, 23, 517–530.

51.

Williams

L. J.

Anderson

S. E.

(1991). Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 17, 601–617.

52.

Wrightsman

L. S.

(1964). Measurement of philosophies of human nature. Psychological Reports, 14, 743–751.

Scenario: Jade will get a poor grade on a joint project if her friend, who is also involved in the project, does not start doing something.
Ways of responding: (i) prioritize the friendship over the project, (ii) prod the friend until he or she does something, (iii) explain the importance of the project, and (iv) hope the friend gets started soon
Question: If you were in the situation, how likely would it be that YOU would do the following? 1 = Not at all likely, 5 = Very likely
	Effectiveness norms (%)
Way	1 (%)	2 (%)	3 (%)	4 (%)	5 (%)	Self-likelihood	Score
(i) Prioritize	33	37	18	08	04	3	.18
(ii) Prod	06	12	32	32	18	2	.12
(iii) Explain	00	01	08	34	57	5	.57
(iv) Hope	25	31	12	13	19	2	.31

Category
Measure	t	p	β
Prosocial tendencies
Altruistic philosophy	2.99	.004	.39
Helping attitudes	3.54	<.001	.44
Prosocial behavior	3.96	<.001	.48
Social support
Appraisal	3.71	<.001	.46
Belonging	2.09	.042	.28
Tangible support	2.15	.036	.29
Total	2.97	.005	.38
Social intelligence
Social awareness	1.55	.128	.21
Information processing	1.22	.268	.15
Social skills	2.40	.020	.32
Total	2.22	.031	.30