Personality Feedback as an Intervention to Encourage Positive Changes on Moral Traits

ABSTRACT

Individuals have a pervasive bias to see themselves as more moral than justified based on their actions, which may undermine their desire to change for the better. To overcome this lack of self-knowledge, a useful method may be to provide individuals with personality feedback. No research has systematically investigated how individuals receive and perceive personality feedback, especially when it is negative and deals with a key aspect of identity, such as morality. The present mixed-methods study aimed to tackle this issue. Participants (N = 17) completed a self-report measure of trait morality (covering honesty, humility, fairness, compassion, and gratitude) and invited 3–5 well-acquainted others to provide corresponding informant reports. Participants then received detailed feedback about their trait morality based on these reports and completed a semi-structured interview with a trained psychologist to assess their perception of the feedback. Results showed that participants generally found the feedback helpful, even when it was negative. As expected, feedback – especially when participants were negatively surprised – was associated with an increase in goals to change in socially desirable ways. “Surprise” and “explanations” were two main recurring themes in the interviews, illuminating the process behind receiving (negative) feedback and developing a motivation to change.

KEYWORDS:

• Trait morality
• personality feedback
• identity relevance
• self-knowledge

Moral behavior (e.g., cooperating with others, treating others in a fair and sincere manner) is crucial for maintaining healthy interpersonal relationships and for the functioning of societies at large (Carnes et al., 2022; Tomasello, 2016). Finding ways to promote moral behavior is therefore critical. A promising approach in this regard is to encourage individuals to behave in (more) moral ways. The current investigation pursues such a “person-centered” approach, which draws on individual-level changes instead of common situation-based approaches where external incentives (e.g., rewards or sanctions) or nudges are implemented to encourage individual behavior change. Specifically, we report on an intervention that leverages the high identity-relevance of morality to motivate people to volitionally change in socially desirable ways.

Morality is an integral part of individuals’ identity: People consider thinking, feeling, and behaving according to moral principles as highly relevant to their self-concept (Mazar et al., 2008; Strohminger & Nichols, 2014; Thielmann et al., 2020) and strive to be seen as moral by others (e.g., Sperber & Baumard, 2012; Vonasch et al., 2018). Indeed, it has been proposed that “moral traits – more than any other mental faculty – are considered the most essential part of identity, the self, and the soul” (Strohminger & Nichols, 2014, p. 159). However, particularly in the moral domain, individuals have a pervasive bias to see themselves in more desirable (e.g., honest, compassionate) ways than their actions warrant (Brown, 2012; Paulhus & John, 1998; Tappin & McKay, 2017). As an example, around 70% of adult participants in Thielmann and de Vries (2021) rated themselves as scoring above average on Honesty-Humility – a morality-related basic trait from the HEXACO personality model (Ashton & Lee, 2007) –, and this relative self-assessment converged only weakly with their trait score based on aggregating across multiple self-report items (r = .16, vs. r ≥ .44 for the remaining HEXACO dimensions). This lack of self-knowledge, known as moral self-enhancement (Alicke & Sedikides, 2009), can arguably undermine corresponding desires to change for the better. Indeed, for morality-related traits in particular, the prevalence of change goals is lower than for traits that are less evaluative (Sun & Goodwin, 2020; Thielmann & de Vries, 2021). Considering that change goals seem to be a necessary condition for volitional longitudinal changes in personality (Hudson et al., 2020), finding ways to promote desires to change represents a first, fundamental step for realizing long-lasting behavioral change.

Based on this reasoning and evidence, increasing people’s self-knowledge may be an effective means to instill a desire to change. But how can one’s self-knowledge be increased? Prior research suggests that personality feedback, which provides people with information about their actual standing on certain personality traits in relation to others, may fulfill this purpose (e.g., Bollich et al., 2011). Applied to morality-related traits, personality feedback will likely hint individuals at a discrepancy between their (overly positive) self-view and their actual trait level. Due to the high identity-relevance of morality, this should then trigger a motivation to change on these traits. Indeed, once receiving feedback on their relative standing on Honesty-Humility, participants reported a stronger desire to change in a socially desirable way on that trait than without feedback (Thielmann & de Vries, 2021).

Although this finding is promising, no research has systematically examined how personality feedback in general, and about different moral traits in particular, is perceived and received by targets. Instead, research has taken for granted that personality feedback constitutes a useful approach by providing individuals with “new, true information about oneself that could not have been accessed through introspection alone” (Bollich et al., 2011, p. 3). However, since the information that individuals may gather about themselves through personality feedback may threaten their positive self-image – which, as outlined above, is particularly likely when referring to morality-related traits – (some) individuals may evaluate the feedback negatively or even discard its accuracy completely. In any case, evidence is mute on the thought processes that personality feedback may trigger in individuals, thereby preventing any insights into how and why feedback may be useful. The present study aimed to offer initial responses to these questions using a mixed-methods approach. Specifically, participants received detailed feedback about their trait morality based on their self-reports and informant reports from three to five well-acquainted others, and we examined their reactions to the feedback in an interview study combining quantitative and qualitative methods.

Methods

Open science and ethics

We preregistered the study (https://doi.org/10.17605/OSF.IO/QGZ3K) and provide all materials and (quantitative) data on the Open Science Framework (OSF; https://osf.io/d947q). Ethical approval was provided by the ethics committee of the Max Planck Society (#2023_03).

Materials and procedure

Participants were recruited at a German university using mailing lists and flyers. They were informed that the study contained two parts: an online pre-assessment and a one-hour interview in the lab. The main language of the study was German, while the interview was conducted in English; thus, being fluent in both German and English was a requirement for participation. Moreover, as per the ethics approval, participants had to be 21 years or older because receiving feedback about one’s morality may be psychologically straining. Participants also knew that in order to be invited to the lab, at least three well-acquainted others (“informants”) had to complete an online survey for them. This ensured that it was impossible for participants to trace back any feedback they received to individual informants, thus warranting informants’ anonymity. Participants were compensated with a base rate of 25€ and could receive up to 30€ (i.e., 2.50€ per additional informant who completed the informant survey, with a maximum number of five informants in total); informants could participate in a raffle of 30 gift vouchers worth 10€ each.

Online pre-assessment

After providing informed consent and basic demographics, participants completed a self-report measure of trait morality containing 51 items from validated scales (see Table S1) in German. Items referred to five dimensions, i.e., honesty (sincerity and incorruptibility), humility (modesty and greed avoidance), fairness (equality and respect), compassion (empathy and helpfulness), and gratitude (give thanks and recognize one’s blessings). Responses were collected on a scale from 1 = strongly disagree to 5 = strongly agree. Next, participants provided names and e-mail addresses of three to five acquaintances who could serve as informants. Upon clicking “Next,” e-mails containing an invitation to complete a survey on behalf of the participant were automatically sent to informants through the survey software and their contact details were deleted immediately. Thus, we experimenters never had access to informants’ personal data.

Next, participants received detailed descriptions of the traits in question (i.e., overall morality and five morality dimensions), one at a time, and were asked to report for each trait (1) how high they scored relative to others on a stanine scale from 1 = very low to 9 = very high and (2) how, if at all, they would like to change, using a scale from 1 = much lower on that trait to 5 = much higher on that trait, with 3 = neither lower nor higher.

Lab session

Around one week after the online pre-assessment, participants were invited to an individual lab session. Upon arrival, they received a feedback sheet (see Figure S1) containing detailed information on their overall morality and the five morality dimensions. The feedback included participants’ raw scores (average trait level, from 1 to 5), ranks (relative position in comparison to a diverse German sample,¹ from 1% to 100%), and stanine scores (from “very low” to “very high”), displayed separately based on their self-reports and the aggregated informant reports. Note that we intentionally refrained from providing information on the range of informant scores to prevent participants from guessing the specific scores their informants provided, thus protecting informants’ anonymity even further. After reading the feedback, participants discussed it in a semi-structured interview with an experienced coach and psychologist (initials blinded). Interview questions referred to participants’ perception and reception of the feedback, to what extent the feedback matched their self-image, how participants explained deviations between the feedback and their self-perception, the importance of the different morality dimensions for their identity, and any behavioral intentions following the feedback (including change goals). Interviews were recorded and transcribed by a professional transcriber.

Finally, participants filled in a post-interview questionnaire. The first block of questions referred to reactions to the feedback (i.e., feelings and surprise) and perceptions of the feedback (i.e., agreement, clarity, depth, and helpfulness). Participants answered these questions separately for the feedback overall, the feedback based on their self-report, and the feedback based on the aggregated informant report. The second block of questions referred to the identity-relevance of morality and its dimensions, change goals, and how relevant participants perceived the five morality dimensions for overall morality.

Informant survey

Informants received a link via e-mail to access the online survey. They first provided informed consent, basic demographics, and information on the kind and duration of their relationship with the target. Then, they reported on the targets’ (participants’) morality using the 51-item trait morality questionnaire (Table S1) in observer-report form.

Participants

As pre-registered, we aimed to recruit 30 participants. However, due to difficulties in recruitment and time constraints² we ended up with 20 participants completing the online pre-assessment. Of these, one participant was not fluent in German and two recruited less than three informants. Thus, these three participants were not invited to the lab, resulting in a total of N = 17 (11 females, 6 males) aged 21 to 30 years (M = 24.7, SD = 2.6). All participants were students (16 university students, 1 doctoral student) from various disciplines (7 psychology, 1 other behavioral science, 1 mathematics, 1 engineering, 1 medicine, 3 other disciplines, 3 unknown). Due to the small sample size and resulting low statistical power, we refrained from running any statistical tests but focus on descriptive statistics.

Participants recruited a total of 76 informants (44 females, 30 males, 2 diverse, 2 unknown) aged 18 to 66 years (M = 29.1, SD = 12.5), with a median of 4 informants per participant. Informants described the targets to be friends (57.9%), relatives (26.3%), close acquaintances (7.9%), romantic partners (5.3%), or distant acquaintances (2.6%). Informants knew the participants for at least 6 months and up to 32 years (M = 9.8, SD = 9.5, Md = 6).

Results

Preliminary analyses

Our approach rests on the notion that morality-related traits are relevant for people’s identity. Thus, we first investigated the identity-relevance of the traits under scrutiny (Table S2). Participants indeed considered morality and its dimensions to all be highly important for their identity. All means exceeded 4 (“important”), ranging from M = 4.12 (SD = 0.70) for humility to M = 4.53 (SD = 0.51) for compassion and fairness (SD = 0.62).

In addition to identity-relevance, we asked participants to assess the traits’ relevance for morality. Descriptively, all five dimensions were considered relevant for morality (Table S2): All means exceeded the scale’s midpoint of 3 (“neither relevant nor irrelevant”), ranging from M = 3.65 (SD = 0.93) for gratitude to M = 4.76 (SD = 0.44) for fairness. In terms of standardized mean-level differences (Cohen’s d), fairness and compassion were considered more relevant for morality than humility (d = 1.69 and d = 1.19, respectively) and gratitude (d = 1.54 and d = 1.13, respectively), yielding large effects.

Lastly, considering that people tend to self-enhance their moral traits (e.g., Tappin & McKay, 2017), we examined how participants perceived themselves relative to others. Indeed, participants ascribed themselves higher than average trait levels on all morality dimensions, except humility: 88.2% considered themselves as more compassionate, 70.6% as fairer, 64.7% as more honest, 64.7% as more moral overall, and 58.8% as more grateful than the average person, whereas only 41.2% considered themselves as humbler.

Reactions to the feedback

Figure 1 summarizes participants’ reactions to and perceptions of the feedback. The feedback was considered rather detailed and easy to understand. Means referring to participants’ feelings were also all above the scale’s midpoint (“neutral”), indicating rather positive reactions to the feedback. Feedback based on informant reports was received more positively than feedback based on self-reports, reflecting that informant reports were higher than self-reports for all dimensions except compassion and gratitude (see Table S3).

Figure 1. Reactions to and perceptions of the feedback (post-interview questionnaire) separated for the different feedback source (i.e., aggregated self- and informant report, self-report, and informant report).

Agreement with the feedback varied quite substantially, showing that some participants disagreed while others agreed with the feedback. Concerning surprise, all means again exceeded the scale’s midpoint (“neither unsurprised, nor surprised”), indicating that participants were indeed somewhat surprised about the information they received.

Most importantly, participants found the feedback generally helpful. Here, means were all above 4 (“helpful”), and no one found the feedback completely useless. Both feedback sources (self and informants) were rated similarly helpful, with a slight tendency for the combination of the two to be rated most helpful.

Change goals

Figure 2 summarizes the results for change goals before and after receiving the feedback. Participants showed noteworthy increases in change goals after receiving the feedback for honesty (d = 0.59) and fairness (d = 0.58), and small increases for humility (d = 0.35) and compassion (d = 0.28). Change goals on overall morality showed only a very small increase (d = 0.11) and change goals for gratitude remained essentially the same.

Figure 2. Change goals before and after receiving the feedback.

The response scale ranged from 1 = much lower to 5 = much higher, with 3 = neither lower nor higher

Qualitative analysis

The qualitative interview data were analyzed by means of coding reliability thematic analysis (Braun & Clarke, 2023). First, the first and second author (NC and CM) familiarized themselves with the interviews by listening to the audio recordings, reading the corresponding transcripts several times, and revisiting the audio tapes whenever necessary. Second, CM identified two main themes relevant to the research topic, based on the number of people whose coded responses clustered around that theme. The first theme was called “being surprised about a score.” Twelve participants had expressed surprise about at least one of their scores, in particular when a score deviated from that person’s expectations. The second theme was called “explaining a score,” referring to a variety of ways participants tried to explain how their scores came about. This theme was present in all interviews. Third, NC and CM searched for quotes related to change goals as this was another focus of the study. Although change goals were not identified as a major theme across interviews, it was referred to by several participants. In selecting quotes (also regarding the two key themes), the authors specifically focused on participants who received discrepant scores as these participants were usually most surprised and motivated to explain the discrepancies while at the same time ensuring that the quotes were clear, illustrative, and not identifiable.

Many participants expressed surprise with regard to one, and occasionally two, of their self-scores. This mainly concerned lower self-scores than expected on honesty and/or humility. For example, one person with a “below average” self-score on honesty stated “Personally looking into myself I see myself as an honest person” (Participant A). With regard to honesty in particular, participants often referred to instances of white lies to explain their surprisingly low self-scores, as in “In the way that I act and what I say, (…) that I don’t say something that would come across in a negative way. That’s why I don’t think it’s a bad thing, but I don’t think it’s a very honest thing to do” (Participant F). Surprise about a low honesty self-score was only once explained in relation to breaking legal rules: “I’m not trying to corrupt or trick people but sometimes I’m not following the rules completely and strictly. (…) If there is a chance sometimes to do a small thing, not a big thing, but something small, where you can get in somewhere for free, I would do this maybe. However, I’m not doing big criminal things” (Participant N).

Participants were even more surprised about lower than expected self-scores on humility. This is apparent in comments such as “I don’t think that I regard myself as better than others” (Participant H), “I would also put myself more to the average” (Participant E), and “I do think that I am pretty humble” (Participant F). Only two participants came up with explanations for their surprisingly low self-scores. One attempt resulted in a very specific example, that is, a desire to buy “an expensive sportscar … because I am into cars” (Participant E) without having a general desire for luxury. The other attempt was “I think sometimes it’s hard for me to accept things as they are and to be content with that. That’s something I want to work on” (Participant H).

Not all participants with a low self-score on a certain dimension were necessarily surprised, but these individuals also tried to explain their low scores. For example, low self-scores on fairness were justified by it being difficult to treat everyone exactly the same, as in “I would say in the bubble, probably I’m very fair, but being in a bubble isn’t like treating everyone equally” (Participant M). Regarding honesty, clarifications from the only two participants that had received low self-scores but had not expressed any surprise downplayed the unethicality of their actions. For example, one person stated “Yes, I know that there were questions like if you would be extra nice to someone to get profits, for example, at work. I would say I’d do that but not in an extreme way” (Participant C). For humility, behavioral examples to explain low scores referred to exaggerating stories, posting pictures of exotic holidays on social media, and talking or thinking too much about themselves.

In reaction to informant reports, several participants expressed a relief that informants perceived them more positively than their self-scores implied. However, participants were also aware that other people may not necessarily see all aspects of their personality: “They just see the one personality that I show” (Participant F). In the few cases that below average informant scores were lower than participants’ self-scores, participants specifically tried to explain the differences between scores. These explanations mostly referred to informants having a different perspective on participants’ personality than they themselves have. Examples are: “Most people who filled out the form for me don’t live here in Germany. They don’t actually live in my day-to-day life” (Participant K) and “I assume they don’t know me as well as me” (Participant G). In some cases, people could not find a reasonable explanation or concluded that their intentions did not come across as they had hoped, which made them feel somewhat sad: “Well, for me, it’s really important what other people think of me. If they think that I’m not really that fair, it hurts” (Participant L).

Change goals

Besides providing explanations for low scores, several participants reported goals to change in reaction to the feedback. This is well reflected in “That’s actually good feedback. I know I have to work on it. I am working on it” (Participant B) and “If that is how I presented myself and how other people perceive me even less than that, I want to do better” (Participant D). Goals to change on specific morality dimensions provided further insights into how exactly participants would like to change and potential barriers to change. For example, one person expressed their change goal for honesty as follows: “The only thing is, I want to learn how to be more real and that this at the same time doesn’t create too much conflict” (Participant N). Some participants wanted to become humbler, as in “I feel like I tend to talk about myself a little bit too much or about my accomplishments. I think it’s something that I’m still working on” (Participant P) or “ … not taking myself too seriously or my life and everything I want to achieve because I’m still just a guy” (Participant Q). Two participants who realized that they might not behave as fair toward others as they intended to expressed a motivation to become fairer, as in “I think if I don’t like people, for example, and they would ask me for help, I would help them but if someone asks me for help that I like, I will help them more, for example. I think that’s something I could work on to improve in the future to be fairer to everyone” (Participant E). In turn, if the informants’ score was lower than expected, participants were often unsure what to do about it. An exception was apparent for gratitude, where some participants mentioned they could try to just say “thank you” more often.

Discussion

The present study combined quantitative and qualitative analysis to explore in-depth responses to personality feedback about moral characteristics and to gain insight into the ability of feedback to instill goals to change in socially desirable ways. First and foremost, participants generally considered the feedback helpful and reacted quite positively to it. This was especially true when the feedback came from well-acquainted informants who tended to judge targets more positively than they themselves did – a common finding known as the “letter of recommendation effect” that is often apparent when informants are nominated by targets themselves (Leising et al., 2010). Correspondingly, whenever participants voiced disagreement with the feedback, it most often referred to their self-reported rather than informant reported trait levels. This is in line with self-maintenance theory (Mazar et al., 2008) suggesting that individuals should strive to maintain a positive moral self-image – which, in the context of obtaining personality feedback, can be most easily realized by disagreeing with undesirable information.

In line with the idea that participants indeed learned something new through the feedback, they tended to be somewhat surprised by the information they received. This was particularly evident in the post-feedback interviews where surprise was one of the two major recurring themes. The qualitative analysis revealed that surprise was pronounced for honesty and humility. These observations align with the Self-Other Knowledge Asymmetry (SOKA) model which suggests that individuals have blind spots in self-perception (i.e., aspects that are known to others but not to the self) especially for traits that are highly evaluative while being difficult to see from the outside (Vazire, 2010) – which applies to honesty in particular (Thielmann et al., 2017). Many participants intensively reflected on the reasons for their low scores on these two dimensions in particular, and this search for explanations was the other main theme brought up in the interviews. For example, surprisingly low honesty scores were considered to stem from a lack of authenticity in some situations and a tendency to tell white lies to avoid conflicts with others. As such, participants’ lay views reflected well the current scholarly discourse on the complex relationship between honesty, authenticity, and morality (Bailey & Iyengar, 2022): Although white lies are sometimes considered the ethical thing to do (Levine, 2022), this may come at the cost of authenticity (not being “real” with others) and therefore be seen as dishonest and undesirable.

Irrespective of the level of surprise, participants tried to explain their low scores with a variety of justifications, which again complies with the notion that morality is a central feature of identity that people are motivated to protect from threats (Mazar et al., 2008; Strohminger & Nichols, 2014). Justifications included minimizing their actions (“(…) you would be extra nice to someone to get profits, for example, at work. I would say I’d do that but not in an extreme way”) and pointing to the unobservability of some behaviors (“They just see the one personality that I show”). In the rare cases where participants were unable to find explanations for negative feedback coming from informants, participants reported feelings of hurt.

Participants rated the feedback as a useful reminder of something they already knew (as in “That’s actually good feedback. I know I have to work on it. I am working on it”) but also as a wake-up call (as in “If that is how I presented myself and how other people perceive me even less than that, I want to do better”), especially resulting in desires to be more “real,” humble, and fair in the future. In this sense, it is conceivable that, by pointing out blind spots in self-perception and/or discrepancies between self- and other-perceptions, the feedback acted as a catalyst for a motivation to change, in line with broader identity change processes proposed by identity control theory (conflict of meaning; Burke, 2006) or the three-factor identity model (in-depth exploration and reconsideration of commitment; Crocetti et al., 2008). That is, it might be that the feedback triggers a conflict between one’s identity of being an honest person and others’ perceptions of that person being less honest than average, which could then lead to desires to change to resolve the discrepancy. Similarly, negatively discrepant feedback on honesty might prompt an in-depth exploration of one’s current commitment to being honest with others and also stimulate that person to reconsider such commitment, for example by developing desires to become more genuine. These considerations are aligned with our findings of small to medium increases in change goals in reaction to the feedback. The increases were descriptively greater for the facet-level traits compared to general morality. Together, these findings further support the idea that personality feedback can be used to instill change goals (Bollich et al., 2011; Thielmann & de Vries, 2021), and suggest that narrow traits may be better candidates for doing so. In this sense, providing people with concrete, specific advice on how to improve on certain (narrow) traits may be particularly helpful, as several participants in our study seemed rather clueless about how to implement change. Indeed, it appears that some people tend to formulate specific goals (e.g., “donate more to charity”) change goals, while others express broad goals (e.g., “be more compassionate,” Sun et al., 2023). When not specifically trained to do so, many people generate vague, non-measurable goals (e.g., “be more sociable,” Hudson & Fraley, 2015).

Although providing new insights into the reactions to personality feedback, our results should be interpreted in light of some limitations. The very small sample size did not allow us to run statistical tests with sufficient power. Our sample was also relatively young, which may have inflated change goals slightly (Hudson & Fraley, 2016). Additionally, we only measured change goals rather than actual behavior. Given the well-known intention-behavior gap (Sheeran & Webb, 2016), a next step would be to study whether these goals can be used to leverage actual personality change, which – according to theories on personality change (see, e.g., the TESSERA framework, Wrzus & Roberts, 2017) – will further require repeated enactment of trait-congruent behaviors to eventually change one’s habits. Lastly, since we had no control group who did not receive personality feedback, we cannot rule out the possibility that (descriptive) increases in change goals after the feedback may, at least in part, be due to socially desirable responding. However, we consider this latter limitation unlikely for two reasons: First, the results align with prior evidence on the positive influence of personality feedback about morality-related characteristics on change goals (Thielmann & de Vries, 2021) where participants received feedback in written form as part of an online study, thus arguably reducing socially desirable responding by design. Second, we only found pre-post differences in change goals for some traits (i.e., honesty, fairness, and to a lesser degree, humility and compassion) but not for others (i.e., morality and gratitude), even though all traits under scrutiny can be considered socially desirable.

Overall, our main findings support the notion that personality feedback on moral traits can be reasonably well received by targets and elicits some introspection, leading to explanations and motivation to change for the better. In this sense, our results provide novel and rich information about how people perceive and receive personality feedback about traits that are key to their identity, and thus offer important insights for future research aimed at using personality feedback as an intervention to sustainably promote positive change goals. Furthermore, our findings can inform the growing research area on identity interventions (Barnett et al., 2021) by showing that personality feedback may be a valid strategy for triggering reflections about one’s overly positive moral identity and, by instilling goals to change identity, eventually changing (moral) behavior. We show that negative feedback can make people aware of discrepancies between their actual self and their ideal self. Even though it may be painful, it can help people to become better persons, which should also have positive consequences for their well-being (Olaru et al., 2023).

Acknowledgments

This research was funded by a grant from the European Union (ERC) awarded to the last author (acronym: KNOW-THYSELF; project no.: 101039433). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

Disclosure statement

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

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15283488.2024.2340488

Additional information

Funding

The work was supported by the European Research Council [101039433].

Notes

1. The comparison sample stem from the Prosocial Personality Project (see https://osf.io/m2abp; N = 1,327 participants, 586 females, aged 18 to 66 years, M_age = 45.9, SD_age = 11.7).

2. Our difficulties were due to the semester break taking place at the time of data collection and availability of the coach who conducted the interviews. Also, we had several inclusion criteria (i.e., a minimum age of 21, fluency in English, and at least three informants) that arguably made it even more difficult to recruit participants.