Content uploaded by Geoffrey W Sutton
Author content
All content in this area was uploaded by Geoffrey W Sutton on Nov 10, 2014
Content may be subject to copyright.
Pastoral Psychology, Vol. 53, No. 6, July 2005 (C2005)
DOI: 10.1007/s11089-005-4822-7
Can Derailed Pastors be Restored? Effects
of Offense and Age on Restoration
Geoffrey W. Sutton1,3and Eloise K. Thomas2
We studied the restoration beliefs of conservative protestant pastors using an
experimental design. We varied pastor offense (romance, affair) and offender age
(young, middle age) in narratives presented in an online study and at a ministerial
retreat. Both groups rated restoration potential highest for the younger pastor
in the affair condition but there was no difference in the romance condition.
In addition, the participants believed a younger pastor would fare better in his
marriage. We discussed our findings in terms of research on forgiveness and
reconciliation.
KEY WORDS: clergy; restoration; reconciliation; forgiveness; sexual offense.
The sexual abuse by pastors of congregants is a serious problem. Although
the sexual abuse of children by Catholic priests recently gained national attention,
sexual activity between pastors and congregants has harmed the members of many
faith communities. According to Hadman-Cromwell (1991) between 6% and 10%
of clergy sexually violated congregants. Thoburn and Balswick (1998) analyzed
data from 186 male seminarians representing 23 denominations. They found that
15.6% engaged in extramarital sexual activity. Most sexual activity was between a
pastor and a church member, but sexual intercourse was primarily between a pastor
and a counselee. More recently, researchers from the Hartford Institute for Religion
Research (2000) found that 23% of 532 Christian congregations experienced a
pastor-congregation disruption due to a pastor’s sexual activity. Although we
could have studied the relationship between restoration and several variables (e.g.,
deceit, embezzlement, substance abuse), we elected to study restoration responses
1Geoffrey W. Sutton is Associate Professor of Psychology at Evangel University in Springfield, MO.
2Eloise K. Thomas is a graduate student in the Clinical Psychology Program of Evangel University in
Springfield, MO.
3Address correspondence to Geoffrey W. Sutton, Department of Behavioral Science, Evangel
University, 111, N. Glenstone Ave., Springfield, Missouri 65807; e-mail: suttong@evangel.edu.
583
0031-2789/05/0700-0583/0 C
2005 Springer Science+Business Media, Inc.
584 Sutton and Thomas
to sexual offenses because of the harmful effects on victims, spouses, community
members, and the offending pastors.
SEXUAL OFFENSES AND CARING PROFESSIONALS
The Thoburn and Balswick study (1998) provides empirical support for the
problem of pastoral sexual activity beyond the newspaper headlines. The im-
portance of the problem among Christian leaders may be reflected in Conlin’s
(2001) finding that 85% of accredited Christian seminaries and schools of theol-
ogy include sexuality education in their coursework, although only 47% offer an
independent course on sexuality. According to Hadman-Cromwell (1991), pastors
who violated sexual boundaries with congregants were more likely to have been
victims of childhood sexual abuse than were other pastors. Although there are only
a few empirical studies of sexual offenses by pastors, studies of psychotherapists’
sexual behavior provide a basis for understanding the effects of sexual activity
between a person who has been entrusted to function as a guide or counselor and
a person who expects to obtain guidance and counsel.
Beliefs of mental health professionals regarding therapist-client sex may be
reflected in their codes of conduct. The ethics codes of the American Psychological
Association (APA) (2002), the American Counseling Association (ACA) (n.d.),
and the National Association of Social Workers (NASW) (1999), as well as the
American Association of Pastoral Counselors (AAPC) (1994), clearly prohibit all
sexual contact or harassment, both overt and covert, both physical and verbal,
between therapist/counselor and client. Furthermore, they all deal with the issue
of sexual intimacy after the termination of the professional relationship. The APA
and ACA state that such a relationship is unethical until at least two years after
termination, and even then the therapist is responsible to make sure the relationship
is not harmful to the former client. The NASW code is basically the same except
that the two-year recommendation is stated more strongly. However, the AAPC
code makes no reference to the two-year time period. It states that any sexual
relationship with a client, current or former, is always unethical. Additionally, in
all four codes, the responsibility lies solely with the therapist never to engage
in any type of relationship with a client or former client that is harmful to that
person.
In addition to ethics codes, criminal and civil statutes that prohibit sexual re-
lationships between counselors and counselees reflect societal attitudes about such
behavior; in fact, in some states, these statutes include clergy (Haspel, Jorgenson,
Wincze, & Parsons, 1997). Moreover, in some states the language of the statutes
indicates that former counselees can sue their counselors for violating sexual
boundaries even if the violation occurred after a counseling relationship ended.
However, plaintiffs must prove that they were still dependent on their counselors
at the time of the sexual relationship or that deception was involved (i.e., that the
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 585
therapist indicated that the sexual relationship was therapeutic and then ended the
professional relationship in order to engage in the sexual relationship). In Texas,
this provision applies even after the two-year period indicated in most professional
codes of ethics if the counselee can prove emotional dependency (Haspel et al.,
1997). Furthermore, these laws apply even if the counselee is the initiator of the
relationship (Hadman-Cromwell, 1991).
Studies have shown varying percentages of therapists who have had or are
having sexual relationships with clients. In a survey of a stratified random sample
of psychologists, Lamb, Catanzaro, and Moorman (2003) found that 3.5% of the
respondents reported having had one or more sexual boundary violations in a
professional setting and 2% reported at least one violation with a client (the others
were with supervisees or students.) Furthermore, 59% of these relationships with
clients began after the professional relationship ended. However, 50% of the
therapists ended the professional relationship for the purpose of beginning the
sexual relationship (Lamb et al., 2003). In a similar study, Lamb and Catanzaro
(1998) found that 8% of psychologists reported sexual violations in professional
relationships. Of these, 6% were with clients. In the latter study, Lamb et al.
(2003) suggested that perhaps the difference in results was due to the requested
information. The 2003 study requested more details about the offenses, which may
have caused more negative feelings and, consequently, less willingness to report
the incidents.
Some suggested that the numbers proposed in studies are lower than the
actual numbers because most studies used self-report instruments (e.g., Haspel
et al., 1997). In an attempt to address self-report bias, Wincze, Richards, Parsons,
and Bailey (1996) used an indirect method of asking therapists if they had treated
clients who had had sexual relationships with former counselors. They surveyed
all Rhode Island licensed psychologists, psychiatrists, social workers, marriage
and family therapists, and mental health counselors who practiced from 1989 to
1991. They sent a similar survey to all registered psychologists who practiced
from 1989 to 1992 in the state of Western Australia, Australia. The return rate
of surveys was 49% for Rhode Island and 48% for Western Australia. In Rhode
Island 26% and in Western Australia 22% of those who returned surveys indicated
that in the previous three or four years they had treated clients who had been
sexually involved with a therapist. In Western Australia 5% of the cases involved
clergy compared with 17% in Rhode Island.
Some authors have explored factors associated with psychotherapist-client
sex. In a study on therapists’ sexual attraction to clients, Pope, Keith-Spiegel, and
Tabachnick (1986) mailed surveys to 1,000 psychologists in private practice ran-
domly selected from the APA Membership Register. The questionnaire focused
on sexual attraction to clients and the therapists’ reactions to these feelings. Of
the 1,000 psychologists surveyed, 585 (58.5%) responded. Of the respondents,
508 (86.8%) reported having been sexually attracted to at least one client. How-
ever, only 104 (18%) reported that they had ever seriously considered acting out
586 Sutton and Thomas
on their sexual attractions. Furthermore, only 13 (2%) considered it more than
once or twice. In fact, only 6.5% reported having had a sexual encounter with
a client. Following are the top seven reasons cited by participants for not en-
gaging in sexual relationships with clients, in order from most frequently stated
to least frequently stated: “unethical, countertherapeutic/exploitative, unprofes-
sional practice, against therapists’ personal values, therapist already in a commit-
ted relationship, fear of censure/loss of reputation, [and] damaging to therapist”
(p. 153). Jackson and Nuttall (2001) found that therapists with self-reported his-
tories of childhood sexual abuse might be more apt to violate sexual boundaries
with clients. Lamb et al. (2003) reported that of the 13 therapists who had re-
ported sexual relationships with one or more clients, 3 were still involved in these
relationships whereas 10 said the relationship had ended. All of the ongoing rela-
tionships were rated as “mostly to extremely positive,” and they all three had lasted
anywhere from 2 months to 30 years. However, those whose relationships were
over viewed them much less consistently. In addition, looking back on the rela-
tionships, 20% of the respondents indicated that at the time the relationship ended,
they thought that it was not worth having. However, 50% said that, in retrospect,
the relationship was not worth having. In addition, only 20% indicated that the
relationship had little effect on them professionally. The other 80% indicated nu-
merous negative effects, such as lack of trust in self, distance in relationships, and
guilt.
Other studies have examined the impact on the client. Bouhoutsos, Holroyd,
Lerman, Forer, and Greenberg (1983) surveyed licensed psychologists in California
about clients who reported past sexual encounters with therapists. The authors
viewed the numbers concerning negative effects for the clients to be conservative
because the percentages were based on the total number of cases, and all respon-
dents did not answer all questions. The total number of cases reported was 559.
In 34% of the cases, the respondents indicated that the “patient’s personality was
adversely affected” (p. 190), as evidenced in the narratives by symptoms such as
“increased depression, loss of motivation, impaired social adjustment, significant
emotional disturbance, suicidal feelings or behavior, and increased drug or alco-
hol use” (p. 190). Additionally, 11% of the patients required hospitalization, and
1% committed suicide. Furthermore, 37% of the cases resulted in termination of
therapy, and 40% interfered with therapy. When asked how the incidents affected
the therapeutic process between the respondents and clients, 25% said the patient
was negatively affected, 14% said the patient’s view of therapists and therapy was
negatively affected, and 29% said the process of therapy was negatively affected.
However, 2% said that it sped up the process of therapy. Only 5% of the patients
“quickly sought help which was needed to resolve conflict engendered” while 48%
found it more difficult to begin therapy with a new therapist (p.190). According to
the narratives, the evidences of this difficulty were suspicion and mistrust of ther-
apists, trouble establishing the new therapeutic relationship, extreme wariness in
choosing a new counselor, and long delays in returning for therapy. Furthermore,
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 587
25% of the patients chose the new therapist based on gender whereas 1% chose the
new therapist “for continued sex” (p. 190). Pope (1988) suggested that the effects
of patient-therapist sex may constitute a syndrome similar to “Rape Response
Syndrome, Battered Spouse Syndrome, . . . [or] Post-Traumatic Stress Disorder”
(p. 224). The symptoms are somewhat like reactions to incest and child abuse.
He stated that one similarity is that the evidence of damage may be delayed and
present later in response to a triggering event, much like a flashback of Post-
Traumatic Stress Disorder. In Lamb et al.’s (2003) study, 40% of those who had
engaged in sexual boundary violations stated that they viewed the relationship as
unethical or harmful, and 28% saw the relationship as harmful or unethical but
continued anyway. According to the APA’s Committee on Women in Psychology
(1989), sexual contact between therapist and client results in loss of objectivity
and damage to the client’s trust in general and, specifically, in the therapist.
Interpersonal Offenses and Restoration
Forgiveness researchers have explored the constructs of forgiveness and
restoration in an interpersonal context. They have been careful to draw a clear dis-
tinction between the constructs of forgiveness and reconciliation (Enright & Cole,
1998; Worthington, 1998; Worthington, 2000; Worthington & Drinkard, 2000;
Worthington & Wade, 1999). A common starting point is the offense. Following
the offense, a person experiences unforgiveness. One can then view forgiveness as
an experience in which a set of forgiving emotions replaces unforgiving emotions
(Worthington, Berry, & Parrott, 2001). Both cognitive and physiological factors
may be associated with the forgiveness experience (Berry & Worthington, 2001;
Seybold, Hill, Neumann, & Chi, 2001).
McCullough and Worthington (1994a) referred to reconciliation as the restora-
tion of a relationship following a significant interpersonal offense or series of
offenses. They wrote that it involved a reestablishment of norms for acceptable
conduct in the relationship. According to Worthington (2000), restored trust is a
part of reconciliation: “Reconciliation restores trust in a relationship where such
trust has been violated by a transgression” (p. 1725). Enright and Coyle (1998)
separated the concepts of forgiveness and reconciliation when they noted that a
person could forgive an offender and elect not to be in a relationship with the
offender. They suggested that forgiveness might lead to reconciliation between
an injured party and an offender. For Worthington and Drinkard (2000), recon-
ciliation involves trustworthy behaviors that would restore trust in a relationship.
They also suggested that reconciliation might be less than a full restoration.
Smedes (2001) expressed the belief that people can forgive without reconciling,
but reconciliation cannot happen without forgiveness. In contrast to the prior re-
searchers, Aquino, Tripp, and Bies (2001) linked reconciliation to forgiveness
when they defined reconciliation as a behavioral manifestation of forgiveness.
588 Sutton and Thomas
In this study, we accepted the distinction between forgiveness and reconcilia-
tion when we focused on beliefs of community leaders about the likelihood of
a successful restoration of an offending pastor and spouse. In the interpersonal
context of a marriage, we consider the terms restoration and reconciliation to
have the same meaning. In addition, we conceptualized restoration to spouse as
having a range of possibilities from no restoration in the relationship to a full
restoration, which may include an enhanced relationship as a result of therapeutic
processes.
Offenses and Community Restoration
Restoration within a community is a matter of degrees and is not absolute.
Pastors who lost leadership positions following an offense may experience partial
restoration when they are restored to a ministry position that has less responsibility
than one previously held. In a sense, any restoration that does not include a return
to the same position may be considered a partial restoration.
Forgiveness researchers have discussed restoration in association with rec-
onciling and repairing trust damaged as a result of an interpersonal offense (e.g.,
McCullough & Worthington, 1994b; Worthington, 2000; Worthington & Drinkard,
2000; Worthington & Wade, 1999). McCullough and Worthington (1994a) also
referred to restoration of relationship with God as a spiritual effect following an act
of forgiveness. In the context of the justice system, Worthington (2000) addressed
forgiveness related to three parties and their supporters: a victim, an offender, and
society. The justice system involves society in the process of forgiveness when
acts are taken to restore offenders and victims to their respective communities.
The concept of restorative justice was exemplified through programs in which
offenders confessed and apologized to their victims or their representatives. An-
other concept related to restoration is seeking forgiveness. Sandage, Worthington,
Hight, and Berry (2002) viewed seeking forgiveness as a motivational construct
that applies to a person who has offended someone in a relationship. They de-
fined the motivation in terms of acceptance of moral culpability and reparation
effort. They also elaborated on cognitive, affective, and behavioral aspects of the
construct.
Hughes and Armstrong (1995), differentiated between restoration within a
religious community and restoration to pastoral leadership. In particular, they
proposed that clergy who have committed adultery should not be restored to
pastoral leadership. They also differentiated between restoration to the church
and to the ministry. According to Wells (2003), 74% of respondents believed that
sexually offending clergy should be allowed to return to ministry situations in
which they have no access to victims. However, 44% said they should not be able
to return to “active pastoral ministry” (p. 213). We defined restoration to ministry
as reinstatement to public ministry, either in the position previously held or in
another position.
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 589
RESEARCH OVERVIEW AND HYPOTHESES
Our purpose was to explore factors related to restoration for leaders who
have committed an offense within a community that resulted in a loss of position
and relationships. We focused particularly on pastors because of their prominent
caring role within society and the lack of empirical studies regarding factors
affecting restoration decisions concerning pastors. We elected to study sexual
offenses because those were the offenses that seem to cause widespread harm to
the victims, the congregations, the pastors, and their families. We added age as a
factor because Mullet, Houdbine, Laumonier, & Giard (1998) found that victim
age was related to willingness to forgive. Thus, we wondered if offender age might
also affect the related concept of restoration.
Our research plan began with a pilot study (Study 1) designed to establish
viable dimensions of one independent variable (type of offense) presented in a
narrative format and to explore the reliability of the instruments we constructed
to evaluate restoration beliefs. We hypothesized that undergraduate student par-
ticipants would be more likely to favor restoring a pastor who had a romantic in-
volvement than an affair. We used three global ratings and three sets of individual
Likert-type scale items to assess restoration to God, spouse, and public ministry.
In Study 2 we manipulated type of offense (romance, affair) and age of
offender (young, 28; middle age, 42) and measured restoration to public ministry
and spouse using our revised measures. Our participants were experienced pastors
who responded to an email request to read an online narrative and respond to the
measures. We hypothesized that the participants would be more likely to favor
restoration following a romantic offense than an affair. We explored offender age
as a factor affecting restoration but without a predicted direction.
In Study 3 we examined the same variables as in Study 2 with experienced
volunteer pastors who attended a Midwest retreat. These participants worked with
paper forms of the materials, thus serving as a validity check against findings from
the online sample. Our hypotheses for Study 3 remained the same as for Study 2.
STUDY 1
The major purposes of Study 1 were to establish type of offense as a potential
factor affecting beliefs about restoration, assess the validity of the case narrative
procedure, and determine what measures might be sensitive to participants’ beliefs
about restoration.
Method
Participants and Design
Ten students (1 man, 9 women; mean age =20.8 years) participated in the ex-
periment. They were mostly of European descent except for one Native American,
590 Sutton and Thomas
one Hispanic American, and one other. We used a within-subjects design and
presented the two narratives (romance, affair) in a counterbalanced order.
Materials and Procedure
In a Midwestern Christian university classroom, participants completed con-
sent forms, read the one-paragraph, hypothetical narratives, and completed the
dependent measures. We manipulated pastor offense by describing either a non-
sexual romantic involvement between a 28-year-old married male minister and a
married female congregant or a brief adulterous affair between the two persons.
We identified variables that could affect beliefs about restoration and held them
constant in the narratives (e.g., length of relationship (three weeks), apologetic
response of the pastor, and willingness of the pastor’s wife to participate in a
two-year restoration program). The narrative for the romantic encounter follows:
A minister, age 28, had worked closely with a woman of the church on social activities
for the youth. On one occasion he left the residence of the woman as her husband came
home unexpectedly during a workday. After speaking to his wife about the unusual visit,
she confessed to feeling terribly upset about how things had changed between her and
the minister in the last three weeks. They had often worked together on special music
and discussed how much they had in common. Though admitting to what seemed to a
few hugs and kisses, she denied any sexual activity. The husband was both angry and
devastated when he learned of the relationship. Fearing what he might do if he confronted
the minister, he called a board member who subsequently met with the full board and
the minister. The minister appeared overwhelmed by guilt and confessed to having an
inappropriate relationship with the woman. He affirmed the woman’s story in essential
details and repeatedly apologized for his behavior. Initially his wife of four years was
distraught and was advised to seek individual counseling. She spent a month away at her
sister’s home before returning to her husband. After a few months she agreed to participate
with her husband in a rehabilitation program.
Measures
Each participant read one narrative, completed three global ratings, read a
second narrative, and completed three global ratings. The global ratings assessed
beliefs about restoration to public ministry (GRM), to spouse (GRS), and to God
(GRG). In addition, following the affair narrative, participants rated several Likert-
type items dealing with restoration to public ministry (RPM) or to spouse (RS).
The Likert-type ratings assessed what items may be useful in a future Likert-type
rating scale. We constructed the three global measures as three separate one-item
measures of a global belief about restoration. Ratings on the GRM ranged from 1
to8(1=A full return to a similar public ministry position with an enhanced ability
to minister to others as a result of the experience, 8 =In general, no return to a
ministry position is likely). Ratings on the GRS ranged from 1 to 9 (1 =A divorce.
Most relationships would not survive the experience, 9 =A relationship that is
fully restored and enhanced as a result of the experience. There is a high likelihood
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 591
of a long-term successful marriage). Finally, the GRG scale had seven values (1 =
A relationship that is improved and enhanced as a result of the experience, 7 =A
worsened relationship that will take more than two years to experience a modest
degree of restoration). An example from the RPM is:
3. This minister could likely be restored to public ministry within two years
provided he fully cooperated with a fellow minister to complete a rehabilitation
plan.
Strongly disagree Disagree Agree Strongly agree
An example from the RS is:
9. This minister could likely be successfully restored in his marriage with his
spouse within two years provided he fully cooperated with a fellow minister to
complete a rehabilitation plan.
Strongly disagree Disagree Agree Strongly agree
Results and Discussion
To be consistent with the other measures, we reversed the GRM scores so
that the high score reflected a favorable restoration rating. We analyzed the nine
completed GRM ratings using a paired samples t-test which revealed a significant
rating favoring the romance (M=4.78, SD =1.56) over the affair (M=3.44,
SD =1.88) conditions, t(8) =2.67, p=.029. The results of the GRS (romance
M=5.40, SD =1.35; affair M=5.60, SD =1.65) and the GRG (romance
M=5.80, SD =1.40; affair M=5.00, SD =1.70) analyses were not significant
(p>.05). Reliability analysis of the Likert-type items revealed α=.77 for scores
produced by a group of six RPM items and α=.81 for a group of three RS items.
This small sample suggested that the two categories of pastor sexual offense
resulted in reliably different beliefs about restoration potential. The lack of a
difference on the GRG rating was not surprising given general expectations that
God would be forgiving in the situation which described an apologetic offender
desirous of restoration. Thus, we elected to drop the GRG from further study.
The willingness of the spouse to participate in treatment may have influenced the
generally favorable ratings for restoration. However, in the case of the spouse, we
obtained the ratings from college students whose young age may have affected
the favorable restoration outcome. Thus, we elected to include restoration to spouse
in follow-up studies. Finally, we found several Likert-type items that resulted in
adequate internal consistency values for this limited sample.
STUDY 2
In Study 2 we used a between-subjects factorial design to examine the effects
of pastor offense (romance, affair) and offender’s age (young, 28; middle age,
592 Sutton and Thomas
42). We obtained e-mail addresses of pastors in supervisory positions and other
experienced pastors affiliated with the Assemblies of God. We defined experience
as a minimum of five years of pastoral work. We considered their beliefs about
restoration to be most valuable in these exploratory studies.
Method
Participants and Design
Of 58 conservative evangelical male pastors, 53 with a minimum of five
years of ministerial experience produced valid web records. They were primarily of
European descent (88.7%) with some from other ethnic backgrounds (Asian 3.8%,
Hispanic 1.9%, Native American 5.7%). Their ages were below 30 (1.9%), 30–39
(9.4%), 40–49 (30.2%), 50–59 (41.5%), and 60+(17%). All but one had some
college education: two years of college (7.5%), undergraduate degrees (66%),
master’s degrees (18.9%), and doctoral degrees (5.7%). Years of experience were
6–9 (9.4%), 10–19 (17%), 20–29 (43.4%), 30–39 (17%), and 40+(13.2%). We
randomly assigned participants to one of four case groups formed from the two
levels of the two independent variables.
Procedure
We sent a total of 507 emails to pastors requesting their participation in a
web-based study. The emails included a link to the web page that explained the
study and provided a continuation link if they agreed to participate. The web page
was programmed to rotate among the four narrative conditions each time the site
was hit. Following each narrative, the participant viewed a series of 17 items to
be rated. The narratives were the same as those in Study 1 except that we varied
pastor age in addition to type of pastor offense.
Measures
We measured restoration beliefs using the global ratings (GRM, GRS) and
the Likert-type rating scales (6-item RPM and 3-item RS) from Study 1. In Study
2α=.68 for the RPM 6 item scale and α=.60 for the RS 3 item scale.
Results and Discussion
Two records were removed because they were below our cutoff of five years
experience. Three cases were removed due to participant entry errors. Mean
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 593
Tab l e 1 . Means and Standard Deviations of Dependent Variables (Study 2)
Romance Affair
Young age Middle age Young age Middle age
n=14 n=13 n=11 n=15
Mean SD Mean SD Mean SD Mean SD
Variable
GRM 5.14 1.51 5.38 1.56 6.18 .98 4.47 2.17
RPM 17.21 1.58 16.69 2.29 16.42 1.90 16.66 3.39
GRS 5.57 1.99 5.77 2.17 6.27 1.62 5.47 1.68
RS 9.00 .68 8.29 1.17 9.27 .65 8.13 1.64
Note. GRM is Global Restoration to Ministry. Its range is from 1 to 8. RPM (Restore to Public
Ministry) is a 6-item Likert-type scale range 6 to 24. GRS is Global Restoration to Spouse with range
1 to 9. RS (Restore to Spouse) is a 3-item Likert-type scale range 3 to 12.
item values were used for two cases with missing data for only two items. See
Table 1 for means and standard deviations.
We conducted the main analysis using two 2 (offense type) ×2 (offender age)
multifactorial, multivariate procedures. In the first analysis we examined the two
measures of restoration to public ministry, and in the second analysis we examined
the two measures of restoration to spouse. The MANOVA results for restoration
to public ministry were significant for the interaction of offense and pastor’s age,
Wilks’s =.849, F(2, 48) =4.26, p=.02, η2=.15. The main effects were not
significant (p>.05). Next, we computed follow-up ANOVAs on the individual
measures. ANOVA results indicated a significant effect for the offense and pastor’s
age interaction for GRM, F(1, 49) =4.57, p=.038, η2=.085, but not for RPM.
No other effects were significant. The correlation between age of participant and
RS was significant (r=−.24; p=.04); therefore, we included participant age
as a covariate in the analysis. The MANCOVA results for restoration to spouse
were significant for the pastor’s age, Wilks’s =.833, F(2, 47) =4.70, p=
.014, η2=.17. The main effect for offense type and the interaction effect were not
significant (p>.05). Then, we computed follow-up ANCOVAs on the individual
measures. ANCOVA results indicated a significant effect for the pastor’s age for
RS, F(1, 48) =8.66, p=.005, η2=.15, but not for GRS. No other effects were
significant.
The significant interaction between type of offense and pastor’s age revealed
a difference in mean restoration to public ministry ratings between the two age
groups for the affair condition. Experienced pastors believed a successful restora-
tion more likely for the younger pastor than the middle-aged pastor. This could
be due to several expectations. A greater range of impact is possible with a more
senior pastor who is likely more widely known. In addition, greater expectations
may well be associated with maturity. Moreover, the younger pastor was consid-
ered likely to have a more successfully restored marriage, independent of offense
594 Sutton and Thomas
type, compared to that of the middle-aged pastor. This finding may also suggest a
greater impact on the couple as a result of the middle-aged pastor’s offense than
for the couple with fewer years of relationship history. The lack of an interaction
effect suggests that pastor’s age may not moderate beliefs about the effects of an
extramarital relationship on a marriage.
STUDY 3
In Study 3 we examined the effects of pastor offense (romance, affair) and
offender’s age (young, 28; middle age, 42). We used a between-subjects factorial
design as in Study 2, but participants completed paper versions of the measures
while attending a clergy retreat.
Method
Participants and design
Eighteen pastors (15 men, 3 women) produced valid responses. They were
all members of the Assemblies of God and were all of European descent. Ages
varied: <30 (5.6%), 30–39 (11.1%), 40–49 (33.3%), 50–59 (27.8%), and 60+
(22.2%). Educational levels varied: high school (5.6%), two years of college
(22.2%), undergraduate degrees (33.3%), master’s degrees (27.8%), and doctoral
degrees (11.1%). Their years of experience were: 6–9 (22.2%), 10–19 (27.8%),
20–29 (27.8), 30–39 (16.7%), 40+(5.6%). We randomly assigned participants to
one of four narrative groups formed from the two levels of the two independent
variables.
Procedure
Packets of materials contained one of four narratives followed by 17 items that
comprised the four dependent measures and demographic questions. A separate
envelope contained the consent forms. The research assistant who distributed the
materials to volunteer participants was unaware of the order of the narratives. The
assistant was a pastor with research experience who participated in the retreat.
Measures
We measured restoration beliefs using the global ratings (GRM, GRS) and
the Likert-type ratings (6-item RPM and 3-item RS) from Study 2. In Study 3 α=
.60 for the RPM 6 item scale and α=.76 for the RS 3 item scale.
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 595
Tab l e 2 . Means and Standard Deviations of Dependent Variables (Study 3)
Romance Affair
Young age Middle age Young age Middle age
n=3n=4n=5n=6
Mean SD Mean SD Mean SD Mean SD
Variable
GRM 6.00 1.00 4.50 3.51 5.60 .55 4.67 2.80
RPM 15.67 1.53 16.75 1.26 16.80 1.79 14.50 2.81
GRS 4.00 3.00 5.25 2.22 5.22 2.49 6.50 1.87
RS 8.67 .58 7.50 1.73 8.20 .84 7.00 2.19
Note. GRM is Global Restoration to Ministry. Its range is from 1 to 8. RPM (Restore to Public
Ministry) is a 6-item Likert-type scale range 6 to 24. GRS is Global Restoration to Spouse with range
1 to 9. RS (Restore to Spouse) is a 3-item Likert-type scale range 3 to 12.
Results and Discussion
We removed one case due to an incomplete record, leaving 18 complete
records for analysis. See Table 2 for means and standard deviations. We conducted
the main analysis using two, 2 (offense type) ×2 (offender age) multifactorial,
multivariate procedures. We did not limit our sample based on participant age or
gender; thus, we used participant age and gender as covariates in the analyses. In
the first analysis, we examined the two measures of restoration to public ministry,
and in the second analysis, we examined the two measures of restoration to
spouse. The MANCOVA results for restoration to public ministry were significant
for the interaction of offense and pastor’s age with participant age and gender as
covariates, Wilks’s =.451, F(2, 11) =6.71, p=.012, η2=.55. The main
effects were not significant (p>.05). Next, we computed follow-up ANCOVAs
on the individual measures. ANCOVA results indicated a significant effect for
the offense and pastor’s age interaction for RPM with participant age and gender
as covariates, F(1, 12) =5.56, p=.036, η2=.32, but not for GRM. The
two restoration to spouse measures appeared to be independent of each other
(r=.10, p=.35), and participant age was a significant correlate with RS (r=
.76; p=.000); therefore, we analyzed each restoration to spouse measure using
separate ANCOVA’s with participant age as the covariate. The ANCOVA results
for restoration to spouse using the RS items from Study 2 were only significant
for the pastor’s age, F(1,13) =5.04, p=.04, η2=.28. The results for offense,
interaction, and the ANCOVA for GRS were not significant (p>.05).
The significant interaction between type of offense and pastor’s age revealed
a difference in mean restoration to public ministry ratings between the two age
groups for the affair condition. Participants believed a successful restoration to
public ministry more likely for the younger pastor than the middle-aged pastor.
These results were similar to those obtained from the online study but with the
596 Sutton and Thomas
Likert-type restoration scale and not the one-item global restoration scale. Similar
to the online participants, retreat participants also believed that the younger pastor
would fare better in his marriage following either type of offense.
GENERAL DISCUSSION
In the three studies, we found beliefs about restoration were responsive
to narrative material whether rated by students or experienced church leaders.
Our primary goal was to examine beliefs among experienced pastors who would
presumably be in a position to influence the outcomes for peers who become
derailed. For experienced pastors, the age of the offending pastor was a factor
moderating their beliefs about restoration to public ministry. In studies two and
three, experienced pastors were reliably more lenient in their beliefs about restoring
a younger pastor who committed adultery than a middle-aged pastor with the
same offense. Age did not affect their beliefs when the offense was a romantic
relationship. We also found a reliable effect for beliefs about restoration to spouse
indicating an expectation that a younger pastor would fare better than a middle-
aged pastor.
The finding of a significant pastor age and offense interaction for restoration
to ministry, but only a main effect for restoration to spouse, suggests that pastors
view these restoration outcomes differently. In the same manner that forgiveness
researchers have separated the concepts of forgiveness and reconciliation (Enright
& Cole, 1998; Worthington & Drinkard, 2000; Worthington & Wade, 1999), we
also may be able to draw a distinction between restoration within closely held
relationships, such as a marriage, and restoration within a community context.
Why did experienced pastors favor the younger pastor? One explanation might
be that experienced pastors are used to functioning as mentors toward younger
pastors and thus perceive the younger pastor as more likely to respond to guidance
than an older pastor who is more likely to be a mentor rather than a mentee. The
value of participant age as a covariate also appears to support the findings of Mullet
et al. (2000) that a forgiving disposition increases with age. Although differences
exist between forgiving and restoring responses, forgiveness and restoration may
share variance with a more general source trait as suggested by Ashton and Lee
(2001).
We also noted a measurement effect. In the online study, the one-item global
measure of restoration to ministry was sensitive to the differences, but in the retreat
packet research, only the 6-item scale reflected group differences. The order of
the items was the same in both studies. Although we cannot be sure of the reasons
for this difference, we hypothesize that online respondents, likely working from
an office, may be less tolerant of more indepth measures than those who attend
retreats. The slight difference of two items for the spouse measures may have been
too few to produce a differential effect.
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 597
Our studies had several limitations. We explored restoration beliefs within
the context of conservative evangelical pastors who were exclusively (Study 2) or
predominantly (Study 3) men. We limited our assessment of beliefs to self-report
measures. The reliability values for the rating scales were moderate and may have
attenuated the effects obtained.
Based on our limited exploration of restoration beliefs, we tentatively offer
several suggestions for further research. First, there is a need to develop a range of
reliable measures of beliefs about the restoration process. Although our measures
were sensitive to the variables in these studies, the reliability values varied below
desirable levels. Second, there is a need to explore a wide range of variables
that may moderate beliefs about restoration decisions. We held several variables
constant in our narratives, yet there remain factors such as rehabilitation time, other
types of offense, type of apology, prior offense history, participation of the spouse,
self-interest, and personality variables that may affect restoration beliefs. Third,
there is a need to examine the beliefs of congregants who may have experienced
problems within their community of faith and whose beliefs may well affect the
success of any restoration decisions. Fourth, there is a need to address matters of
restoration beliefs in other professions such as health care, business, education,
and so forth. Fifth, there is a need to examine the relationship among the constructs
of forgiveness, reconciliation, and restoration especially in the context of offenses
within a community.
ACKNOWLEDGMENTS
The authors wish to thank Gary Allen and Jeffrey Fulks for their assistance
with data collection, Nathan Sutton for the creation and maintenance of the web
site for the online study, and Kelley White for her extensive email correspondence.
We are also grateful for the funding and support provided by the Faculty Summer
Research Program of Evangel University.
REFERENCES
American Association of Pastoral Counselors. (1994). Codes of ethics online project center for the
study of ethics in the professions. Retrieved July 10, 2003, from Illinois Institute of Technology
Web site: http://www.iit.edu/departments/csep/PublicWWW/codes/coe/ Ethics 2.html
ACA Code of Ethics. (n.d.). Retrieved July 10, 2003, from http://www.fairaccess.org/acaethic.
htm#SECTION%20A
American Psychological Association. (2002). Ethical Principles of psychologists and code of conduct
2002. Retrieved May 14, 2004, from http://www.apa.org/ethics/code2002.pdf
Aquino, K., Tripp, T. M., & Bies, R. J. (2001). How employees respond to personal offense: The
effects of blame attribution, victim status, and offender status on revenge and reconciliation in
the workplace. Journal of Applied Psychology,86, 52–59.
Ashton, M. C., & Lee, K. (2001). A theoretical basis for the major dimensions of personality [Electronic
version]. European Journal of Personality,15, 327–353.
598 Sutton and Thomas
Berry,J. W., & Worthington,E. L., Jr. (2001). Forgivingness, relationship quality, stress while imagining
relationship events, and physical mental health. Journal of Counseling Psychology,48, 447–455.
Bouhoutsos, J., Holroyd, J., Lerman, H., Forer, B. R., & Greenberg, M. (1983). Sexual intimacy
between psychotherapists and patients. Professional Psychology: Research and Practice,14,
185–196.
Committee on Women in Psychology, American Psychological Association. (1989). If sex enters into
the psychotherapy relationship. Professional Psychology: Research and Practice,20, 112–115.
Conlin, S. C. (2001). Seminary sexuality education survey: Current efforts, perceived need and
readiness in accredited christian institutions [Electronic version]. Journal of Sex Education and
Therapy,26, 301–309.
Enright, R. D., & Coyle, C. T. (1998). Researching the process model of forgiveness within psycho-
logical interventions. In E. L. Worthington, Jr. (Ed.), Dimensions of forgiveness: Psychological
research and theological perspectives (pp. 139–161). Philadelphia: Templeton Foundation Press.
Hadman-Cromwell, Y. C. (1991). Sexual misconduct by clergy. Journal of Religious Thought,48,
64–72.
Hartford Institute for Religious Research. (2002). A quick question: How common is clergy sexual
misconduct? Retrieved May 28, 2004, from Hartford Seminary, Hartford Institute for Religious
Research Web site: http://hirr.hartsem.edu/research/quick question18.html
Haspel, K. C., Jorgenson, L. M., Wincze, J. P., and Parsons, J. P. (1997). Legislative intervention
regarding therapist sexual misconduct: An overview. Professional Psychology: Research and
Practice,28, 63–72.
Hughes, R. K., & Armstrong, J. H. (1995, April 3). Why adulterous pastors should not be restored.
Christianity Today,39, 33–36.
Jackson, H., & Nuttall, R. L. (2001). A relationship between childhood sexual abuse and professional
misconduct. Professional Psychology: Research and Practice,32, 200–204.
Lamb, D. H., & Catanzaro, S. J. (1998). Sexual and nonsexual boundary violations involving psy-
chologists, clients, supervisees, and students: Implications for professional practice. Professional
Psychology: Research and Practice,29, 498–503.
Lamb, D. H., Catanzaro, S. J., & Moorman, A. S. (2003). Psychologists reflect on their sexual
relationships with clients, supervisees, and students: Occurrence, impact rationales, and collegial
intervention. Professional Psychology: Research and Practice,34, 102–107.
McCullough, M. E., & Worthington, E. L., Jr. (1994a). Encouraging clients to forgive people who have
hurt them: Review, critique, and research prospectus. Journal of Psychology and Theology,22,
3–20.
McCullough, M. E., & Worthington, E. L., Jr. (1994b). Models of interpersonal forgiveness and their
applications to counseling: Review and critique. Counseling and Values,39, 2–14.
Mullet, E., Houdbine, A., Laumonier, S., & Giard, M. (1998). “Forgivingness:” Factor structure in a
sample of young, middle-aged, and elderly adults. European Psychologist,3, 289–297. Retrieved
May 14, 2004 from PsycARTICLES database.
National Association of Social Workers. (1999). Code of Ethics. Retrieved July 10, 2003, from
http://www.naswdc.org/pubs/code/default.asp
Pope, K. S. (1988). How clients are harmed by sexual contact with mental health professionals: The
syndrome and its prevalence. Journal of Counseling and Development,67, 222–226.
Pope, K. S., Keith-Spiegel, P., & Tabachnick, B. G. (1986). Sexual attraction to clients: The human
therapist and the (sometimes) inhuman training system. American Psychologist,41, 147–158.
Sandage, S. J., Worthington, E. L., Jr., Hight, T. L., & Berry, J. W. (2000). Seeking forgiveness:
Theoretical context and an initial empirical study. Journal of Psychology and Theology,28,
21–35.
Seybold, K. S., Hill, P. C., Neumann, J. K., & Chi, D. S. (2001). Physiological and psychological
correlates of forgiveness. Journal of Psychology and Christianity,20, 250–259.
Smedes, L. B. (2001, December 1). Keys to forgiving: How do you know that you have truly forgiven
someone? Retrieved October, 11, 2003 from EBSCOhost database. Christianity Today,45, 73.
Thoburn, J. W., & Balswick, J. O. (1998). Demographic data on extra-marital sexual behavior in the
ministry. Pastoral Psychology,46, 447–457.
Wells, K. (2003). A needs assessment regardingthe nature and impact of clergy sexual abuse conducted
by the Interfaith Sexual Trauma Institute. Sexual Addiction and Compulsivity,10, 201–217.
Can Derailed Pastors be Restored? Effects of Offense and Age on Restoration 599
Wincze, J. P., Richards, J., Parsons, J., & Bailey, S. (1996). A comparative survey of therapist sex-
ual misconduct between an American state and an Australian state. Professional Psychology:
Research and Practice,27, 289–294.
Worthington, E. L., Jr. (1998). An empathy-humility-commitment model of forgiveness applied within
family dyads. Journal of Family Therapy,20, 59–76.
Worthington, E. L., Jr. (2000). Is there a place for forgiveness in the justice system? Fordham Urban
Law Journal,27, 1721–1734.
Worthington, E. L., Jr., Berry, J. W., & Parrott, L. III. (2001). Unforgiveness, forgiveness, religion,
and health. In T. G. Plante & A. C. Sherman (Eds.), Faith and health: Psychological perspectives
(pp. 107–138). New York: Guilford.
Worthington, E. L., Jr., & Drinkard, D.T. (2000). Promoting reconciliation through psychoeducational
and therapeutic interventions. Journal of Marital and Family Therapy,26, 93–101.
Worthington, E. L., Jr., & Wade, N. G. (1999). The psychology of unforgiveness and forgiveness and
implications for clinical practice. Journal of Social and Clinical Psychology,18, 385–418.
