Social networks among ministry relationships: relational capacity, burnout, & ministry effectiveness.
In recent years, the stress of ministry leaders has become of increasing interest within the academic community and the religious public. Researchers have found that clergy members suffer from hypertension, depression, and obesity at higher rates than most people in the United States (Vitello, 2010). Clergy work is oriented toward serving others and characterized by a devotion to Christian faith and a feeling of divine calling (Proeschold-Bell et al., 2012). Clergypersons occupy many roles including teacher, counselor, preacher, administrator, and fundraiser (Proeschold-Bell et al., 2012; Weaver, Flannelly, Larson, Stapleton, & Koeing, 2002). Though the term "clergyperson" is generally reserved for someone ordained inside of a particular denomination, many other religious professionals and volunteers have occupations similar to that of clergy, such as relational ministers. Relational ministers, who are clergy or laity, are those working for faith-based organizations with the strategy of building personal relationships to carry out the mission statement of their respective organizations. As a group, relational ministers work in ways similar to clergypersons: teaching, mentoring, fund-raising, administrating, organizing events, etc. Due to similarities and overlap with clergypersons, it can be speculated that relational ministers (clergy or laity) may suffer similar physical and emotional health problems as a result of their occupation.
No one cause has been linked to the negative effects associated with clergy or relational ministry jobs, but interpersonal relationships have been implicated as a major stressor for those doing cross-cultural relational ministry (Carter, 1999). What is it about personal relationships that may lead to these negative outcomes? Theoretical developments in the field of evolutionary psychology provide insight into this question.
Social Network Theory
Evolutionary psychologist and anthropologist, Robin Dunbar (1992) found a predictable relationship between relative neocortex size and social group size in a regression analysis of 38 non-human primates. He observed that animals with a larger neocortex size relative to body mass tended to have larger social group sizes. Considering the substantial size of the neocortex in relation to a human's body mass, Dunbar (1993) extrapolated from the regression fit line and estimated the upper limit of a typical human's personal social network to be 148. This figure, known as Dunbar's number, is commonly rounded to 150 and consists of personal, loving social relationships. Other researchers have corroborated this estimation in studies attempting to measure social network sizes, finding that the average active social network size centered around 150 (Hill & Dunbar, 2003; Killworth, Bernard, & McCarty, 1984).
Dunbar's number encompasses the personal relationships in which an individual is in regular contact and feels a level of emotional connectedness (Dunbar, 2008). People in this category may include family members and friends. It should be noted that work and activity related relationships are not included in Dunbar's number unless they are continued outside of work or the particular activity (Dunbar, 2008). Work relationships are temporary in nature; that is, they are formed to serve a mutual interest or address an issue and generally end after the task is completed. However, it is often true that work and activity members move into the social sphere described by Dunbar's number as people spend more time with one another outside of their respective activities. When applying this theory to relational ministers, relational categorization becomes complicated, as the relationships with others are the focus of work. This further blurs the line between a work/personal relationship dichotomy. Overall, however, those included in Dunbar's number are the individuals that one knows and cares about on a personal basis.
Sub-networks characterized by decreasing degrees of intimacy constitute the core of this active social network: an inner circle of trusted individuals consisting of approximately 5 people, another ring of close family and friends of about 15 people, a larger circle of around 50 people, and then a personal active social group of 150. An outer ring of acquaintances that numbers about 500 people (Dunbar, 2008; Roberts & Dunbar, 2011) may surround the active network. These relationships can be thought of as nested within concentric circles, such that the five trusted people in the inner layer are also part of the personal social network of 150 people (Dunbar, 2008; see Figure 1). It is important to note that Dunbar also expresses these social network ring sizes in terms of ratios, such that each social network ring is approximately three times the size of the preceding ring (Dunbar, 2008).
Dunbar (2008) theorized that people are limited to a personal social network of approximately 150 due to constraints of cognition, time, and emotional investment. Cognitive capacities such as memory, empathy, and the ability to predict another's behavior require a larger neocortex and appear related to social network size (Frith & Frith, 1999). Time is a limited resource and does not allow for people to invest in an unlimited number of relationships at an intense level (Dunbar, 2008). Dunbar's number has been shown to be the limit to which people can invest in relationships on an emotional level (Sutcliffe, Dunbar, Binder, & Arrow, 2012). Emotional intensity, like time, is not an unlimited resource and restricts human social abilities (Roberts, Dunbar, Pollet, & Kuppens, 2009). An individual must make decisions regarding those in whom she will emotionally invest, and some relationships may require more emotional attention than others.
Personal Differences in Dunbar's Number
Past research shows fluctuations in social network capacity based on age, gender, and personality factors such as extroversion; however, this research is limited (Lu et al., 2009; Pollet, Roberts, and Dunbar, 2011). Additionally, relational ministry workers may have self-selected their occupations in part because of their unusually large social network capacity. Unlike much of the general population, relational ministers choose a career in which they actively and continually make relationships for the purposes of carrying out a specific goal.
No published research exists that examines the consequences of exceeding one's natural active social network size. Humans are naturally limited by social intelligence and time available to 'groom' relationships; thus, most people do not aim to exceed Dunbar's number (Dunbar, 1998). People working in the context of relational ministry may be an exception to this general rule. Relational ministry workers generally strive to increase their social network in order to further their ministry goals. Those operating beyond their natural capacity may stress cognitive-emotive resources and/or require more time to invest in relationship maintenance than is available without making sacrifices in other areas of life (Dunbar, 2008). Hence, attempting to exceed one's natural relational capacity may contribute to burnout.
Burnout, as defined by Maslach (1982), is a job-related syndrome characterized by emotional exhaustion, lack of accomplishment, and feeling disconnected from those one is working to support. It generally arises from chronic emotional overload that results from extensive interpersonal work with others in a care-giving role, especially with those who have problems or are troubled (Maslach, 1982). Relational ministry workers may be at particular risk for developing burnout as a result of spending the majority of their time ministering to, or caring for, others.
People experiencing burnout are likely to report lower levels of job satisfaction and have higher attrition rates than those not experiencing burnout. In a longitudinal study of Anglican clergy, Randall (2004) found that emotional exhaustion, perceived feelings of low personal accomplishment, and depersonalization were significantly related to participants' desire to leave their job, suggesting that burnout contributes to higher attrition rates. Similarly, Kinman, McFall, and Rodriguez (2011) found that emotional demands are inherent in clergy work and that the increased intensity and frequency of these demands lead to lower levels of job satisfaction and higher levels of psychological distress. These findings corroborate Maslach's (1982) assertion that burnout leads to a decrease in work performance, ultimately leading to a physical withdrawal from the source of burnout. When these findings are applied to relational ministry workers, it appears that that they are at risk for job dissatisfaction, attrition, and, ultimately, decreased ministry effectiveness.
Relational Capacity, Burnout, &: Ministry Effectiveness
Relational ministers as a group seek out and develop relationships in order to achieve their ministry goals and, as such, may average active social networks of more than 150 relationships. Engaging in this many invested personal relationships on a regular basis is likely to lead to emotional exhaustion and burnout. Since burnout leads to job dissatisfaction and decreases job effectiveness, it is likely that those relational ministers with more relationships than predicted by social network theory are less effective in their ministry goals, rendering them less effective in their respective ministries. It is the purpose of this study to investigate the following hypotheses:
1. Full-time relational ministers are predicted to average over 150 people in their active social network.
2. Ministries with leaders exceeding 150 people in their active, personal social networks are predicted to have lower perceived ministry effectiveness scores than those who do not exceed this alleged limit.
3. Ministers who exceed 150 in their active social network are predicted to experience greater levels of burnout than those with fewer than 150.
Participants were from two samples: a missionary group and a youth pastor group. Participants in the missionary group were 84 national and international Christian ministry field workers recruited from mission agencies that achieve ministry goals through a relational model. Participants in the youth pastor group were 153 youth pastors serving in Protestant Christian churches throughout the United States. The combined group totaled 237 participants, of which 69.2% were male and 30.8% were female. The sample was predominantly Caucasian (78.5% Caucasian, 9.7% Asian/ Pacific Islander, 5.5% Black/African American, 4.2% Hispanic/Latino, and 2.1% Bi-racial/Other), and participants had a mean age of 31.9 years with a range of 17 to 69 years. Of the 237 participants, 63.3% reported having a "full-time" ministry position, whereas 36.7% reported being "part-time." See Table 1 for further demographic information. (1)
Participants were recruited through large organizations. (2) For the missionary group, administrative staff for the respective organizations provided access to international and national staff email lists. Emails containing a brief description of the study were sent, and those who indicated interest in the study were sent a follow-up email containing a link to an online survey. Participants were compensated for their participation in the study, and compensation varied from $25-540 depending upon level of participation in the study. Participants were given the option to complete a follow-up survey approximately one year after their initial participation at which time they were compensated up to an additional $60 as part of a larger longitudinal study. Youth pastors were recruited in the same way as ministry workers, but through churches. Additional measures were added at the time of data collection for the missionary sample. As such, youth pastors were not asked to complete the Big Five Inventory's extroversion subscale (John, Donahue, & Kentle, 1991). All other procedures were the same.
Social Network Structure. Participants were asked to generate a list of people who they considered to be part of their relational network. Participants were asked to categorize each relationship (spouse/ partner, family, friend, ministry-derived relationship, organizationally-demanded relationship, etc.), classify the level of relationship demand, identify the level of emotional closeness (EC), and estimate the frequency of contact (FC) for each person listed. Participants rated EC on a seven-point scale where the low end of the scale (1) meant "Not At All Close" and the high end (7) meant "Very Close." Options for frequency of contact included the following: daily, weekly, monthly, and yearly. A social network structure was assembled for each participant based on the ratings of EC and FC, similar to previous research (Roberts et al., 2009; Pollett et al., 2011). Each social network structure consisted of a support group (EC = 7 & FC = daily-monthly), sympathy group (EC = 6 & FC = daily-monthly), close group (EC = 7 or 6 & FC = yearly, EC = 5 & FC = daily-yearly, EC = 4 & FC = daily-monthly), and active group (EC = 4 & FC = yearly, EC = 3 & FC = daily-yearly). Individuals contacted less frequently and/or who were emotionally more distant, were considered "acquaintances" and outside of the active network.
Extroversion. Extroversion was measured by the eight extroversion-introversion items taken from the Big Five Inventory (John et al., 1991), a 44-item self-report measure of personality including extroversion, agreeableness, conscientiousness, neuroticism, and openness scales. These items were rated on a five-point Likert scale, anchored 1 = disagree strongly and 5 = agree strongly. Reliability of this scale was acceptable ([alpha] = .85).
Subjective Ministry Effectiveness Measure. This scale, created for the current study, consisted of six statements regarding subjective ministry effectiveness and ministry outcomes. Participants were asked to rate the level at which they agreed or disagreed with these statements on a seven-point scale (with one meaning strongly disagree and seven meaning strongly agree), including items like the following: "this ministry has been effective at fostering relationships with the community we serve," "this ministry has been effective at fulfilling the organization's mission statement," and "this ministry has been effective at stimulating growth in the faith of [people ministered to]." Item scores were summed, and internal consistency of this scale was acceptable ([alpha] = .76).
Professional Quality of Life Scale. The Professional Quality of Life Scale (ProQOL; Stamm, 2010) is a 30-item measure that is intended to assess how an individual feels regarding their work and role as a helper. The ProQOL contains questions within the categories of compassion satisfaction, burnout, and secondary trauma. The burnout subscale, used for the current study, measures difficulties and hopelessness as they relate to work effectiveness, such as feeling disconnected from the relationships they are working to foster. Responders rate statements on a five-point scale, one meaning "never" and five meaning "very often." Reliability of the scale was acceptable ([alpha] = .77).
Relational Network Ring Sizes
Based upon descriptive statistics, for both the missionary and youth pastor group, the support (innermost, most intimate) and sympathy (next most intimate) network rings appeared enlarged, whereas the active network ring appeared atrophied when compared with Dunbar's and others' previous observations (Table 2 and Figure 1). As a whole, relational ministry workers (both part-time and full-time) averaged significantly fewer than 150 relationships in their active social network, M = 119.29, SD = 65.32, t(236) = -7.24, p < .001. Similar one-sample t-tests comparing this sample to the interior network ring sizes with greatest empirical and theoretical support, support (M = 5) and sympathy (M = 15), indicated these network sizes significantly deviated from norms in the missionary group, support ring: t(83) = 4.76,p < .001; sympathy ring: t(83) = 5.23, p < .001, and youth pastor group, support ring: t(152) = 5.46, p < .001; sympathy ring: t(152) = 6.37, p < .001.
Due to both theoretical and demographic similarities, after failing to find significant differences in social network sizes, participants from the missionary group and the youth pastor group were merged to form a combined group for analysis of the three hypotheses.
Additional Research Questions. Due to the considerable differences between the current study's support and sympathy ring sizes with those of other studies (Hill & Dunbar, 2003; Sutcliffe et al., 2012) the authors proposed an additional research question: do relational ministry workers with support and sympathy network sizes different from those established in previous research perceive their ministry to be less effective and report more burnout than those with support and sympathy network sizes similar to previous research?
Outcome Variable Scores
Perceived Ministry Effectiveness. Responses (M = 5.54, SD = 1.36) suggest that participants regard their ministries as generally effective, given that responses averaged above the mid-point (4).
Burnout. Using the ProQOL scale metric (1-5 scale), the average for burnout was 2.09 (SD = 0.84), which is below the midpoint of the original scale (3). When adding all burnout scores together and dividing by the number of items (different from using the scale metric as a base), the average total burnout score for participants in the combined group of the current study was 50.00 (SD= 10.00) with 23.2% of the sample scoring below 43.00 and 22.4% of the sample with scores above 57.00. As a whole, the current study's sample reflected levels of burnout similar to that of the general population as measured by Stamm (2010), t(236) = 0.04,p = .971. (3)
Bivariate correlations for individual non-concentric network ring sizes (i.e., each ring minus the rings inside it) and outcome variables were conducted (see Table 3). The ring sizes used in the correlations were expressed as ratios of the specific network ring size over total number of relationships (e.g., non-concentric close ring/total network). Ring sizes were expressed in this manner to control for participants' individual total reported network sizes, as more enthusiastic participants may have simply listed more people generally, and less diligent participants may have under-counted all of their relationships. Using ratios also more closely respects Social Network Theory in which the ratios of inner rings of intimacy (each ring .3 - .33 of the next larger ring) is thought to remain constant (Dunbar, 2008).
For both the missionary and youth pastor groups, support network size was significantly related to sympathy network size (missionary group: r = .42,p < .001; youth pastor group: r = .18,p = .023). These relationships were positive, demonstrating that participants with larger support networks had larger sympathy networks. Additionally, for the youth pastor group, sympathy network size was significantly correlated to close network size (r = .32, p < .001), such that participants who reported larger sympathy network sizes had larger close networks. Interestingly for both groups, support (missionary group: r = -.41; youth pastor group: r = -.34), sympathy (missionary group: r = -.51; youth pastor group: r = -.45), and close (missionary group: r = -.42; youth pastor group: r = -.34) network size were each negatively related to active network size at the p < .001 significance level, demonstrating that participants with larger support, sympathy, and/or close networks had smaller active networks.
Among participants in both groups, support network size (innermost ring of intimacy) was significantly related to burnout, such that participants with larger support networks reported less burnout than those with smaller support networks. Additionally, burnout was significantly related to ministry effectiveness among participants in both groups such that those who reported higher levels of burnout perceived their ministry to be less effective than those who reported lower levels of burnout.
Data regarding levels of extroversion was only collected for the missionary group (Table 3). This variable was included in the bivariate correlational analysis in order to determine whether extroversion was related to predictor and outcome variables. Correlations were run, and extroversion was significantly and positively related to support, sympathy, and close network sizes, indicating that participants with larger support, sympathy, and/or close network sizes reported higher levels of extroversion. Extroversion was also significantly and negatively related to active network size, such that participants with larger non-concentric active networks (i.e. a larger number of less intimate but still personal relationships) reported lower levels of extroversion. Correlational analyses did not reveal a significant relationship between the outcome variables and extroversion.
Supplemental Preliminary Analyses
Larger reported support and sympathy rings than in previous research, along with the results of the correlational analyses, motivated a supplemental analysis to determine whether larger support and sympathy rings lead to smaller close and active rings. That is, does having more relationships that are characterized by high levels of emotional closeness and frequency of contact lead to having fewer relationships characterized by lower levels of emotional closeness and less frequent contact? In order to answer this question a multiple regression was run in which non-concentric support and sympathy network sizes were used to predict non-concentric close and active network size. Nonconcentric support and sympathy network sizes were added together and expressed in a ratio of support and sympathy size sum over total network size. Non-concentric close and active network sizes were also added together and expressed in a ratio of close and active size sum over total network size. These ratios were created to help control for the varying amount of reported relationships by different participants, such that the non-concentric support-sympathy ratio served as the independent variable, and the non-concentric close-active ratio served as the dependent. The regression showed that support and sympathy network size ([beta] = -.25, p < .001) was a significant predictor of close and active network size, [R.sup.2] = .06, F(1, 235) = 15.53, p < .001, such that a larger support and sympathy network size predicted a smaller close and active network size.
Hypothesis 1. As reported above, the first hypothesis, that relational ministers would exceed 150 relationships in their active network, was not supported. In the combined group the total number of active relationships (i.e., excluding those that were less intimate and/or frequent in contact) averaged 119.29 (SD = 65.32).
Hypothesis 2. Initial bivariate correlations did not show a significant relationship between active relational network size and perceived ministry effectiveness, r(235) = .02, p = .806. However, most of the sample (76.8%) identified as having 150 or fewer relationships in their active network. In an attempt to better detect the consequences of exceeding Dunbar's number, participants were grouped into those exceeding 150 in their active network versus those having 150 or fewer reported active personal relationships. This variable was used to predict perceived ministry effectiveness, controlling for covariates. Results of a hierarchical regression did not find that dichotomized active network size was significantly related to perceived ministry effectiveness, and only years with current organization was found to significantly predict perceived ministry effectiveness ([beta] = .13, p = .042). (4)
Hypothesis 2: Additional Research Questions. Due to the unusually large support and sympathy rings among the current study's participants when compared with previous samples (Dunbar, 2008; Hill & Dunbar, 2003; Sutcliffe et al., 2012), and their large variability (sec Table 2), the question of whether typical support and/or sympathy network sizes predicted perceived ministry effectiveness and burnout was explored through hierarchical multiple regression analyses. Previous research demonstrates that an individual's support network generally consists of about five relationships (Hill & Dunbar, 2003; Sutcliffe et al., 2012), however, no literature exists regarding a predicted or observed normal range of support network size. Hence, one whole value above and below the mean (5) was used to create a hypothesized optimal range of four to six individuals. The predictor variable for support network was dummy coded into two different variables, such that Support Network Below 4 indicated participants with fewer than four relationships in their support network (n < 4 = 63; 26.6%), and the variable Support Above 6 indicated participants with more than six relationships in their support network (n > 6 = 124; 52.3%).
A similar procedure was conducted with the sympathy network predictor. Previous research suggested a typical sympathy network ranging from 12 to 15 relationships (Sutcliffe et al, 2012), thus, the predictor sympathy network was dummy coded into two variables: the variable Sympathy Below 12 indicating a sympathy network with fewer than 12 relationships (n < 12 = 53; 22.4%), and the variable Sympathy Above 15 indicating a sympathy network with more than 15 relationships (n > 15 = 156; 65.8%).
Support on Perceived Ministry Effectiveness. A hierarchical regression showed that the support network size was significantly related to perceived ministry effectiveness. In the first step, years with current organization was identified as a significant predictor of perceived ministry effectiveness, [R.sup.2] = .02, ,F(1, 235) = 3.97, p = .047 (see Table 4). After adding Support Below 4 and Support Above 6 to the model while controlling for years with current organization, step two of the model indicated a significant relationship between support network size and perceived ministry effectiveness, [R.sup.2] = .08, [DELTA][R.sup.2] = .07, F(3, 233) = 6.93, p < .001 (see Table 4). The significant effect in step two of the model appeared to be mostly driven by Support Below 4, indicating that participants with fewer than four relationships in their support network perceived their ministry to be less effective than those with four to six relationships in this network.
Sympathy on Perceived Ministry Effectiveness. Results of the hierarchical multiple regression showed that the sympathy network size was significantly related to perceived ministry effectiveness. In the first step of the regression, years with current organization was identified as a significant predictor of perceived ministry effectiveness, [R.sup.2] = .02, F(l, 235) = 3.97, p = .047 (see Table 4). After adding Sympathy Below 12 and Sympathy Above 15 to the model while controlling for years with current organization, step two of the model indicated a significant relationship between sympathy network size and perceived ministry effectiveness, [R.sup.2] = .09, [DELTA][R.sup.2] = .07, F(3,233) = 7.51,p < .001 (see Table 4). Similar to the relationship between support network size and perceived ministry effectiveness, sympathy network size for those with fewer than 12 relationships in their sympathy network appeared to account for most of the variance accounted for in the model and was negatively related to perceived ministry effectiveness. That is, it appears that participants with fewer than 12 relationships in their sympathy network reported lower levels of ministry effectiveness than participants with 12 to 15 relationships in their sympathy networks.
Hypothesis 3. The hypothesis that participants exceeding 150 in their active network would report higher levels of burnout than participants who did not exceed this number was not supported. An initial bivariate correlation did not show a significant relationship between active relational network size and burnout; r(235) = 0.01, p = .441. As in the previous analyses conducted for Hypothesis 2, a cutoff of 150 relationships was established, since the majority of the sample (76.8%) identified having 150 or fewer relationships in their active network. Nevertheless, the resulting analysis detected no significant relationship between exceeding 150 and reported burnout.
Hypothesis 3: Additional Research Questions. Linear regressions were conducted with both support and sympathy network sizes as predictors of burnout.
Support on Burnout. A multiple linear regression was conducted with the predictor variables of Support Below 4 and Support Above 6. The model did not show that support network size was significantly related to burnout, [R.sup.2] = .02, F(2,234) = 2.47, p = .087.
Sympathy on Burnout. An additional multiple linear regression was run with Sympathy Below ([beta] = .34, p < .001) and Sympathy Above 15 ([beta] = .22, p = .03) as predictor variables. The model showed that sympathy network size was significantly related to burnout, [R.sup.2] = .05, F(3, 233) = 6.55, p = .002. Effect sizes of Sympathy Below 12 and Sympathy Above 15 were moderate, and these results demonstrated that participants with fewer than 12 and more than 15 relationships in their sympathy network reported higher levels of burnout.
This project was motivated by three primary hypotheses derived from Social Network Theory and by the general finding that burnout and negative health outcomes are high among clergy. Perhaps those ministers who feel considerable pressure to do the "work of God" via forming personal relationships find the pressure to do so overwhelming because of natural limitations on the number of such relationships anyone can maintain. What researchers did not anticipate was that a trade-off between very intimate relationships (those with whom one interacts daily, weekly, or monthly and rated as a 6 or 7 on a 7-point emotional closeness scale), would drive down the number of total relationships in these ministers' active networks.
Rather than unwieldy and diffuse relational networks leading to burnout and ministry dissatisfaction, these ministers had swollen numbers of very close relationships and relatively atrophied total active social network sizes. Results support the idea that humans have limited capacity for amount and intensity of relationships. When comparing the current sample to Dunbar's proposed network structure, it appears that there may be trade-offs that occur between network rings, which effect overall relational capacity. That is, it could be that a great deal of time, emotional energy, and neurocognitive power are put into developing intimate relationships, which leave relational ministers with minimal reserves from which they can increase the size of their active network. If this is indeed true, it suggests that capacity has been reached. Proportionally large inner rings of relationships--those high in frequency and emotional closeness--were strongly associated with relatively small outer rings. In a typical sample, the outer ring of a social network would number 100 individuals and comprise 2/3 of the total active network. In this sample characterized by large inner social network rings, the outer ring (active ring, not including the inner rings) only averaged 32 individuals and 27% of the total active network.
Total active network sizes in these samples of ministers were smaller than expected, approximately 20.7% smaller on average than Dunbar's number, but within the range found by some other studies. The greatest deviance from previous samples was the size of the support and sympathy rings, which ranged from 66.7%-80% larger than normal, and with enormous variation. Hence, we focused our analytical attention on these two rings. Does the size of the support and/or sympathy ring bear upon perceived ministry effectiveness or burnout?
Results from the current study suggest that very intimate "best friend" relationships of the support ring may be important to one's perception of ministry effectiveness. Relational ministers with fewer than four relationships in their support ring appeared to perceive their ministries to be less effective than those who had relationships within the normal range (4 to 6). Perhaps this finding may reflect that a relational minister may sense that she is succeeding with her ministry if she is forming and maintaining a high level of intimate relationships. These findings could also indicate increased feelings of social support that in turn help the relational minister seem to feel as if his ministry is effective. What is unknown is the makeup of relational ministry workers' support and sympathy rings. Are they made up of mainly supportive relationships, or are they made up of relationships that are characterized by a high level of need from the other? Further research is needed to better understand this dynamic.
Considering the next ring of intimacy--the support ring of (typically) 15 good friends and emotionally close family members--results suggest the importance of an optimum size. Though this sample of relational ministers reported levels of burnout similar to those in the general population, having more or fewer than 12 to 15 relationships in one's sympathy network predicted significantly higher burnout scores. Furthermore, having too few (fewer than 12) people in this ring of intimacy predicted lower perceived ministry effectiveness.
An ideal range of relationships in the sympathy network is explicable if the role of social grooming is considered. Social grooming among humans occurs in the physical presence of another (Dunbar, 2010; Dunbar & Schultz, 2010). Laughter, touch, and synchronous activities are carried out in face-to-face reciprocal interactions and provide feelings of warmth and closeness to all individuals involved, perhaps facilitated by endorphin release. At levels of intimacy characterizing the sympathy network (typically 12-15 individuals), considerable in-person investment temporally and emotionally is required. It may be that a relational minister does not receive enough benefit from social grooming unless he or she has at least 12 people involved in "grooming" him or her. It may also be that a relational minister cannot invest in more than 15 relationships requiring this level of social grooming without making sacrifices to other, less intimate relationships in the active network or neglecting other areas of life. Stated simply, relational ministers may feel under supported with fewer than 12 people in their sympathy ring, but they may also feel overwhelmed by the burden to care if more than 15 people are in their sympathy ring. Either of these can lead to burnout.
An additional note should be made regarding the relationship between extroversion and social network size. In previous research, higher levels of extroversion have been linked to larger active network sizes (Lu et al., 2009; Pollet et al., 2011). The opposite was found amongst participants in the current study, as those with lower levels of extroversion reported more non-concentric active network relationships. This is counterintuitive and appears to contradict previous research. However, it may be that the population of the current study defines emotional closeness in a different manner than the world at large, rating their relationships at much higher levels of emotional closeness than actuality. If this is the case then this finding may not contradict previous research, but reflect that as extroversion increases the sheer number of relationships increase.
Enthusiasm for any purported practical implications for ministry should be muted by a number of considerations. First, this study has inherited the vagueness of previous research in determining the number of people making up one's active social network (in exclusion of mere acquaintances, for instance). Measurement error is possible, and unintended (or intended, see below) deviations from previous practice may have impacted the size and make-up of participants' social networks reported here. However, conservatism was the guiding principle for inclusion in the inner rings. A relationship was only considered to be in the support or sympathy network ring if ratings of frequency of contact and emotional closeness were high. Therefore, it may be the case that the sizes of the inner-most social network rings were underestimated, suggesting that the support and sympathy network rings may be even larger than calculated.
The data collection process was time intensive and entirely electronic, which contributed to smaller sample sizes than originally desired and possible participant fatigue. Participants spent approximately one to two hours typing and categorizing a list of names that coincided with all of their current relationships. This was a tedious task, after which participants completed outcome measure surveys for about another half hour. The time involved in answering data could have led to test fatigue and ultimately omission of data. Participants were able to save their results and finish them at another time, however this posed its own problem, as it was possible to lose track of where one left the survey. This likely also led to the omission of data. Nevertheless, an omission of data would have led to smaller network sizes, thus it may actually be that participants' social networks are larger than what was reported in this study--perhaps closer to previously observed totals for active networks but even larger support and sympathy networks.
Similarly, diverging from previous studies concerning social network theory, work relationships were included in this study because developing enduring personal relationships is the work of relational ministers. Boundaries between work and personal life are fluid for this population. Given that previous protocols excluded work-based relationships, the result of including work relationships should have been greater numbers of relationships being reported in the total active network than in previous studies, not fewer. The fact that the opposite pattern was found demonstrates that the reduced network sizes are not merely the result of participant fatigue (a problem in previous protocols, too), or other protocol differences.
Additionally, caution should be taken when interpreting results in regards to the Subjective Ministry Effectiveness Measure, as this was a measure created for the study and does not have any previous research establishing its validity in capturing perceived ministry effectiveness. An assumption was made that an above average score on the 7-point scale meant that participants felt their ministries were more effective than not, however this may not be the case. The scale was made as an attempt to capture ministry effectiveness, a notion that has been elusive to capture in previous research, since baselines such as ministry growth and budget do not adequately capture the personal development of those in ministry (Ferguson, Qiabing, Spruijt-Metz, & Dyrness, 2006; McKenna & Eckard, 2009; Nauss, 1994). Though the measure of perceived ministry effectiveness in the current study may not be as robust as desired, findings related to burnout could be interpreted in regards to ministry effectiveness. Work effectiveness and burnout have a well-documented history, with higher levels of burnout related to lower levels of work effectiveness (Maslach, 1982; Kinman et al., 2011; Randall, 2004).
Implications for Future Study
This study is one of the first to explore the consequences of deviating from normal social network structure. It is also one of the first to examine whether Dunbar's number and the other rings' intimacy in social networks theory are more than statistical regularities and represent actual capacity limitations. Thus, future research in testing the limits of relational capacity as it relates to social network structure and sizes is needed.
It could also be valuable to identify the makeup of relational network rings for participants who exceed capacity. Who is identified in each ring and is he/she a source of support or one of strain? More valuable information may include the influencing factors behind an individual's relational capacity. Does trauma history or attachment style limit the number of relationships one can maintain? One such study has been conducted, implicating trauma history in larger support and sympathy networks, but further studies should be conducted to parse out the nuances of psychosocial history as it relates to relational capacity (Wilkins, Eriksson, Coppinger Pickett, & Barrett, 2017). If these questions can be answered, relational ministry and other relationship-dependent fields would benefit, as there would be practical implications for forming and tending to relationships in ways that optimally benefit the worker.
Tentative Conclusions, Questions, & Implications
With these caveats, study investigators offer a few tentative conclusions and implications for relational ministry that should be regarded as working hypotheses more than policy recommendations. More research is needed, but it is not premature to begin asking these questions.
First, if indeed organizational and ministry pressures lead relational style ministers (including church-based youth ministers) to have large numbers of intimate relationships at the expense of a full quiver of personal relationships, that fact should give pause. What are the consequences for that minister's personal wellbeing? Will such ministers have enough individuals to support them as they navigate life's challenges, particularly if a large proportion of their relatively small active networks are tied to their current employment? If these ministers are reassigned or otherwise lose their jobs, their social networks could be gutted, too.
Second, assuming that ministers generally have a sense for when things are not going well, it may be that successful effective ministry is bolstered by enough (12 - 15) emotionally close and available friends and family members, and enough best friends (at least 4). Churches, ministry organizations, and others who care for professional relational ministers may consider how they should support these ministers in developing and maintaining their circles of most intimate support and trust.
The third suggestion taps the brakes on the simple idea that more emotionally close relationships are always better. Church and other ministry communities may create an abundance of relationship opportunities, but the finding that too few or too many individuals in the sympathy group (more than 15 or fewer than 12 good, close friends and family), predicted higher burnout scores suggests a Goldilocks solution: maybe ministers need the "just right" number of these trusted supporters and encouragers. Because the design was correlational, we do not know whether increased signs of burnout drive ministers to more very close relationships, or if too many very close relationships lead to burnout, or if both are caused by a third factor. Nevertheless, given the amount of time and emotional energy required to develop and maintain such intimate relationships, it would be no surprise if ministers with too many such relationships feel like they have nothing to give to the ministry-demanded relationships that they are called to develop and, therefore, begin moving toward burnout. Alternatively, if they have too few trusted, emotionally close individuals supporting and encouraging them, other vulnerabilities may arise, also leading to burnout.
Using Dunbar's social network theory as a guiding principle, researchers hypothesized that relational ministers would exceed Dunbar's number (Dunbar, 2008) and in so doing would report higher levels of burnout and lower levels of perceived ministry effectiveness. Contrary to initial hypotheses, as a whole, relational ministers reported significantly smaller active relational networks than those of previous studies (Hill & Dunbar, 2003; Killworth, Bernard, & McCarty, 1984). Relational ministers reported having more relationships at a more intimate level, and fewer relationships marked by looser connections. These atypical social networks were correlated with higher levels of burnout when outside of the ideal range, likely suggesting either the need for support (if intimate relationships were too few) or the overtaxing of investing in intimate relationships (if intimate relationships were too many). Having too few best-friend type relationships was also correlated with lower levels of perceived ministry effectiveness, again suggesting that relational ministers need a certain level of support and/or intimate relationships to feel as if they are doing well in ministry. It should be noted that this study is the first of its kind and as such should be a starting point for further research. Results from this study beg the questions: what is it about relational ministers that leads to atypical social network structures, and how can this knowledge be used to implement change in overall relational minister health?
Candace Coppinger Pickett, Justin L. Barrett, Cynthia B. Eriksson, and Christina Kabiri Fuller Theological Seminary
Author Note; Candace Coppinger Pickett, Graduate School of Psychology, Fuller Theological Seminary; Justin L. Barrett, Graduate School of Psychology, Fuller Theological Seminary; Cynthia Eriksson, Graduate School of Psychology, Fuller Theological Seminary; and Christina Kabiri, Graduate School of Psychology, Fuller Theological Seminary.
Candace Coppinger Pickett is now at Avant Ministries. We thank all of the missionaries, youth pastors, and ministry organizations that participated in this study. We also thank the John Templeton Foundation for funding this study. The views of the authors do not necessarily reflect those of the John Templeton Foundation.
Correspondence concerning this article should be addressed to Candace Coppinger Pickett, Avant Ministries, 10000 N Oak Trafficway, Kansas City, MO 64155. Email: email@example.com
Carter, J. (1999). Missionary stressors and implications for care. Journal of Psychology and Theology, 27, 171-180.
Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 22, 469-493.
Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language in humans. Behavioral and Brain Sciences, 16, 681-735. doi:10/1037/h0088251
Dunbar, R. I. M. (1998). Grooming, gossip, and the evolution of language. Cambridge, MA: Harvard University Press.
Dunbar, R. 1. M. (2008). Cognitive constraints on the structure and dynamics of social network s. Journal of Group Dynamics: Theory, Research, and Practice, 12, 7-16.
Dunbar, R. I. M. (2010). The social role of touch in humans and primates: Behavioural function and neurobiological mechanisms. Neuroscience and Biohehavioral Reviews, 34, 260-268. doi:10.1016/j. neubiorev.2008.07.001
Dunbar, R. I. M., & Schulcz, S. (2010). Bondedness and sociality. Behaviour, 147, 775-803. doi:10.1163/000579510X501151
Ferguson, K. M., Qiabing, W., Spruijt-Metz, D., & Dyrness, G. (2006). Outcomes in faith-based social services: Are we evaluating faith accurately? Research on Social Work Practice, 16, 1-13. doi: 10.1177/1049731505283698
Frith, C. D., & Frith, U. (1999). Interacting minds: A biological basis. Science, 286, 1692-1695.
Hill, R. A., & Dunbar, R. I. M. (2003). Social network size in humans. Human Nature, 14, 53-72.
John, O. P., Donahue, E. M., & Kentle, R. I. (1991). The Big Five Inventory --versions 4a and 54. [Self-report inventory]. Berkeley, CA: UC Berkeley, Institute of Personality and Social Research.
Killworth, P. D., Bernard, H. R., & McCarty, C. (1984). Measuring patterns of acquaintanceship. Current Anthropology, 25, 381-397.
Retrieved from http://nersp.nerdc.ufl.edu/~ufruss/doeuments/patterns%20of%20acquaintanceship.Current%20Anthrop%201984.pdf
Kinman, G., McFall, O., & Rodriguez, J. (2011). The cost of caring? Emotional labour, wellbeing and the clergy. Pastoral Psychology, 60, 671-680. doi:10.1007/s11089-011-0340-0
Lu, Y., Roberts, S., Cheng. T., Dunbar, R. I. M., Lio, P., & Crowcroft, J. (2009). Variation in personal network size, personality and effect on information transmission. Computational Science and Engineering, 4, 188-193. doi: 10.1109/CSE.2009.179
Maslach, C. (1982). Burnout, the cost of caring. Englewood Cliffs, NJ: Prentice-Hall, Inc.
McKenna, R. B., & Eckard, K. (2009). Evaluating pastoral effectiveness: To measure or not to measure. Pastoral Psychology, 58, 303-313. doi:10.1007/s11089-008-0191-5
Nauss, A. (1994). Ministerial effectiveness in ten functions. Review of Religious Research, 36, 58-69.
Poller, T., Roberts, S., & Dunbar, R. I. M. (2011). Extraverts have larger social networks: But do not feel emotionally closer to individuals at any layer. Journal of Individual Differences, 32, 161-139. doi: 10.1027/1614-0001/a000048
Proeschold-Bell, R., LeGrand, S., Wallace, A., James, J., Moore, H., Swift, R., & Toole, D. (2012). Tailoring health programming to clergy: Findings from a study of United Methodist clergy in North Carolina. Journal of Prevention & Intervention in the Community, 40, 246-261. doi: 10.1080/10852352.2012.680423
Randall, K.J. (2004). Burnout as a predictor of leaving Anglican parish ministry. Review of Religious Research, 46, 20-26.
Roberts, S. G. B., & Dunbar, R. I. M. (2011). Communication in social networks: Effect of kinship, network size and emotional closeness. Personal Relationships, 18, 439-452. doi: 10.1111/j.14756811.2010.01310.x
Roberts, S. G. B., Dunbar, R. I. M., Pollet, T. V., & Kupens, T. (2009). Exploring variation in active network size: Constraints and ego characteristics. Social Networks, 21, 138-146. doi: 10.1016/ j.socnet.2008.12.002
Stamm, B. H. (2010). The Concise ProQOL Manual, 2nd Ed. Pocatello, ID: ProQOL.org.
Sutcliffe, A., Dunbar, R. I. M., Binder, J., & Arrow, H. (2012). Relationships and the social brain: Integrating psychological and evolutionary perspectives. British Journal of Psychology, 103, 149-168. doi:10.111 l/j.2044-8295.2011.02061.x
Vitello, P. (2010, Aug. 1). Taking a break from the Lord's work. The New York Times. Retrieved from http://www.nytimes.com/2010/08/02/nyregion/02burnout.html?pagewanted=all&_r=0
Weaver, A. J., Flannelly, K. J., Larson, D. B., Stapleton, C. L., & Koeing, H. G. (2002). Mental health issues among clergy and other religious professionals: A review of research. The Journal of Pastoral Care & Counseling, 56, 393-403.
Wilkins, A., Eriksson, C., Coppinger Pickett, C., & Barrett, J. (2017). Early trauma as a predictor of burnout and social network structure in mission workers. Journal of Psychology & Theology, 45(2).
PICKETT, CANDACE COPPINGER. PhD. Address: Avant Ministries, 10000 North Oak Trafficway, Kansas City, MO 64155. Title: Director of Member Care at Avant Ministries. Degrees: BA (Psychology and History), Baylor University; MA (Clinical Psychology), Fuller Theological Seminary; MA (Theology and Ministry), Fuller Theological Seminary; PhD (Clinical Psychology), Fuller Theological Seminary. Specializations: Member care (Missionary psychological health).
BARRETT, JUSTIN L. PhD. Address: Fuller Theological Seminary, Graduate School of Psychology, 180 North Oakland Ave., Pasadena, CA 91101. Title: Professor of Psychology; Program Director for PhD in Psychological Science. Degrees: BA (Psychology), Calvin College; PhD (Experimental Psychology), Cornell University. Specializations: Cognitive science of religion, Psychology of religion, Cognitive study of culture, Positive psychology, Cognitive development.
ERIKSSON, CYNTHIA B. PhD. Address: Fuller Theological Seminary, Graduate School of Psychology, 180 North Oakland Ave., Pasadena, CA 91101. Title: Associate Professor of Psychology. Degrees: BA (Psychology), Wheaton College; MA (Theology), Fuller Theological Seminary; PhD (Clinical Psychology), Fuller Theological Seminary. Specializations: Trauma, Spirituality, Ministry stress.
KABIRI, CHRISTINA. PhD. Address: 23200 S. Western Ave., Apt. 251, Harbor City, CA 90710. Title: Ordained Minister, International Church of the Foursquare Gospel. Degrees: BS (Psychology), California Lutheran University; MA (Clinical Psychology), Fuller Theological Seminary; MA (Theology), Fuller Theological Seminary; PhD (Clinical Psychology). Fuller Theological Seminary.
(1) Of those recruited in the missionary group, 84 completed surveys. Though 163 youth ministers completed surveys, due to missing data regarding social network size, only 154 could be used. The missing data were not characterized by a discernable pattern of any sort. Out of the 154 remaining participants, one additional participant was eliminated due to irregular reporting, leaving the youth pastor sample at 153 participants.
(2) For confidentiality reasons, organizations in this study are unnamed. However, participants in the missionary group worked for one of two large national/international Protestant Christian mission agencies, with the majority of participants working directly in youth services.
(3) In the ProQOL manual it is indicated that the average score on burnout items is 50.00 (SD: 10.00), with 25% of people scoring below 43.00 and 25% of people scoring above 57.00. A score above 57.00 is indicative of burnout, however researchers warn against over interpretation: "Your score may reflect your mood; perhaps you were having a 'bad day' or are in need of some time off. If the high score persists or if it is reflective of other worries, it may be a cause for concern" (Stamm 2010, p. 28).
(4) Other variables were initially covariates including group difference, gender, age, education level, etc., but only years with current organization was found to have a significant relationship with perceived ministry effectiveness.
Caption: FIGURE 1 Relational structure means. Means are rounded to the nearest whole number to more accurately represent people. Circles are representative of the Support, Sympathy, Close, and Active Network sizes.
TABLE 1 Participant Demographics Missionary Youth Pastor N=84 N=153 Variables n % n % Marital Status Single 25 29.8 60 39.2 Married 58 69.0 86 56.2 Divorced/Separated 1 1.2 7 4.6 Children Yes 49 58.3 54 35.3 No 35 41.7 99 64.7 Education High School 6 7.1 4 2.6 Some College 24 28.5 35 22.9 Bachelor's Degree 45 53.6 60 39.2 Graduate Degree 9 10.7 54 35.3 Hours per Week 1-5 8 9.5 2 1.3 6-10 9 10.7 7 4.6 11-20 10 11.9 22 14.4 21-30 6 7.1 23 15.0 31-40 9 10.7 49 32.0 40+ 42 50.0 50 32.7 Compensation Paid by salary 14 16.7 123 80.4 Paid by raising support 54 64.3 10 6.5 Volunteer 16 19.0 20 13.1 Years with Current Org. 0-4 42 50.0 105 68.6 5-9 23 27.4 40 26.8 9+ 19 22.6 8 4.6 Combined N=237 Variables n % Marital Status Single 85 35.9 Married 144 60.8 Divorced/Separated 8 3.4 Children Yes 103 43.5 No 134 56.5 Education High School 10 4.2 Some College 59 24.9 Bachelor's Degree 105 44.3 Graduate Degree 63 26.6 Hours per Week 1-5 10 4.2 6-10 16 6.8 11-20 32 13.5 21-30 29 12.2 31-40 58 24.5 40+ 92 38.8 Compensation Paid by salary 137 57.8 Paid by raising support 64 27.0 Volunteer 36 15.2 Years with Current Org. 0-4 147 62.0 5-9 63 26.6 9+ 27 11.4 Note. Years with Current Org. = Years with Current Organization TABLE 2 Concentric Social Network Structure Descriptives Missionary Group Networks Support Symp. Close Active M 9.52 25.79 85.61 119.12 SD 8.70 18.98 44.89 59.89 Mdn 7.50 21.00 82.50 103.50 Mode 3.00 8.00 59.00 90.00 Range 50.00 93.00 193.00 279.00 N 84 84 84 84 Youth Pastor Group Networks Support Symp. Close Active M 8.66 24.50 86.85 119.39 SD 8.29 18.46 57.82 68.31 Mdn 7.00 21.00 80.00 107.00 Mode 1.00 21.00 30.00 94.00 Range 47.00 122.00 357.00 373.00 N 153 153 153 153 Note. Symp. = Sympathy TABLE 3 Correlations, Means, and Standard Deviations for Network Circles & Outcome Variables among Participants Missionary Group Variables Min. Eff. Burnout Extrov. Networks Support .29 ** -.23 * .38 *** Sympathy .11 -.02 .29 ** Close .10 -.02 .11 Active -.07 -.00 -.34 ** Outcomes Min. Eff. -- -.39 *** Burnout -- -- .12 M 34.10 50.00 28.18 SD 4.85 10.00 6.49 N 82 84 84 Youth Pastor Group Variables Min. Eff. Burnout Networks Support .11 -.26 ** Sympathy .27 ** -.04 Close .21 ** -.04 Active -.10 .04 Outcomes Min. Eff. -.40 *** Burnout M 32.69 50.04 SD 5.78 9.96 N 153 153 Note. Figures for the different sizes of network circles are not concentric. The network circles are depicted as ratios of network circle size over total network size. Extrov. = Extroversion, Min. Elf. = Perceived Ministry Effectiveness. * p < .05, ** p < .01, *** p < .001 TABLE 4 Hierarchical Multiple Regression Analyses Predicting Perceived Ministry Effectiveness from Network Sizes Predictors B SE P Step 1 Years with Current Org. 0.13 0.06 .13 * Step 2 Years with Current Org. 0.13 0.06 .13 * Active Above 150 -0.76 0.84 -.06 Step 1 Years with Current Org. 0.13 0.06 .13 * Step 2 Years with Current Org. 0.12 0.06 .11 Support Below 4 -2.64 1.01 -.21 ** Support Above 6 0.68 0.89 .06 Step 1 Years with Current Org. 0.13 0.06 .13 * Step 2 Years with Current Org. 0.11 0.06 .11 Sympathy Below 12 -3.03 1.23 -.23 * Sympathy Above 15 0.57 1.09 .05 Note. N=237, * p < .05, ** p < .01, ** p < .001
|Printer friendly Cite/link Email Feedback|
|Author:||Pickett, Candace Coppinger; Barrett, Justin L.; Eriksson, Cynthia B.; Kabiri, Christina|
|Publication:||Journal of Psychology and Theology|
|Date:||Jun 22, 2017|
|Previous Article:||Cultural differences in family affection and coping abilities for missionary kids.|
|Next Article:||Early trauma as a predictor of burnout and social network structure in mission workers.|