Content uploaded by Daniel Brossart
Author content
All content in this area was uploaded by Daniel Brossart
Content may be subject to copyright.
Effectiveness of Telephone Counseling: A Field-Based Investigation
Robert J. Reese
Abilene Christian University Collie W. Conoley and Daniel F. Brossart
Texas A&M University
The use of the telephone has become an increasingly popular mode for providing counseling. However,
little is known about its effectiveness. This study is an initial effectiveness study on telephone counseling.
Participants (N⫽186) who had received counseling from a telephone counseling service rated the
effectiveness of telephone counseling and the quality of their counseling relationship. Generally,
respondents indicated that telephone counseling was helpful for both global and specific improvement
and that they were satisfied with the counseling they received. Respondents also rated the counseling
relationship and level of interpersonal influence similarly to face-to-face counseling studies measuring
the same attributes. The telephone counseling results are given in the context of face-to-face counseling
data from other studies.
Technological advances and emphasis on time-limited, inexpen-
sive treatment has ushered in psychotherapy through use of tele-
communications and the Internet (Rosenfield, 1997). The utiliza-
tion of technology and telecommunications for health care
purposes is commonly referred to as “telehealth,” which has in-
creased in popularity and usage in clinical practice (Jerome et al.,
2000). The Internet and the use of video transmission have re-
ceived more attention for the provision of telehealth services, but
the telephone is overwhelmingly the most popular telehealth me-
dium for delivering clinical services among licensed psychology
practitioners (Williams, 2000). Little research exists, however, on
whether the effectiveness of psychotherapy extends to telephone
counseling. Additionally, the ethics of providing telephone coun-
seling services is questioned in the literature (Haas, Benedict, &
Kobos, 1996).
Telephone counseling has received attention in the literature as
an attractive alternative to traditional face-to-face counseling. The
four strengths that have been attributed to telephone counseling are
as follows: It is less expensive compared to face-to-face counsel-
ing, it allows anonymity, it gives a sense of control, and it is
convenient (Mermelstein & Holland, 1991; Ranan & Blodgett,
1983; Shepard, 1987; Zhu et al., 1996). These strengths are be-
lieved to make counseling possible for some who would not
otherwise seek or receive counseling services.
For example, telephone counseling may be an attractive option
for individuals who cannot afford traditional psychotherapy. Peo-
ple of low socioeconomic status and rural residents have been
underserved in the mental health system (Mays & Albee, 1992).
Although low socioeconomic status individuals are underserved,
evidence indicates that increased psychological distress is related
to low income (Ulbrich, Warheit, & Zimmerman, 1989). Clearly,
if telephone counseling could fulfill some of its promise in pro-
viding effective mental health counseling for underserved people,
it is worth investigation.
Telephone counseling has been found to be effective for certain
circumscribed problems, such as smoking cessation (Orleans et al.,
1991; Zhu et al., 1996), but the research has not been extended to
address adequately the effectiveness of telephone counseling for
general mental health concerns. The research supporting the ef-
fectiveness of telephone counseling for general mental health
concerns consists of case examples that reflect authors’ favorable
experiences (e.g., Fish, 1990; Hines, 1994). Likewise, negative
reactions to telephone counseling have been based on suppositions
and intuitive reasoning (e.g., Buie, 1989; Haas et al., 1996).
The most accepted method to establish that a model of psycho-
therapy works is by clinical trials that use monitored, manualized
treatments, random assignment of participants, and a control/
comparison group—that is, an efficacy study (Howard, Moras,
Brill, Martinovich, & Lutz, 1996). Alternately, a study that focuses
more on whether psychotherapy works as it is practiced in the field
in its unaltered state is referred to as an effectiveness study
(Howard et al., 1996). In an effectiveness study, individuals
choose the type of treatment they prefer and the nature of treatment
is not controlled. Consumer Reports (CR; 1995) conducted an
effectiveness study that received a great deal of attention, both
positive and negative (e.g., Brock, Green, Reich, & Evans, 1996).
Criticism of the CR study has primarily centered on the lack of
controlled clinical trials. However, Seligman (1995), a consultant
to the CR study, heralded the study as the most extensive psycho-
therapy effectiveness study ever conducted and felt that the study
provided solid evidence for psychotherapy’s effectiveness.
Just as there was a need to know whether face-to-face psycho-
therapy as it is done in the field is effective, telephone counseling
as currently provided needs a similar effectiveness study, a goal we
pursued in this study.
Another issue in psychotherapy research is the dose–effect
response curve. There appear to be different results depending on
Robert J. Reese, Department of Psychology, Abilene Christian Univer-
sity; Collie W. Conoley and Daniel F. Brossart, Department of Educational
Psychology, Texas A&M University.
Correspondence concerning this article should be addressed to Robert J.
Reese, Box 28180, Abilene Christian University, Abilene, Texas 79699, or
to Collie W. Conoley, Department of Educational Psychology, 704 Har-
rington Education Center, Texas A&M University, College Station, Texas
77843-4225. E-mail: jeff.reese@acu.edu or collieconoley@tamu.edu
Journal of Counseling Psychology Copyright 2002 by the American Psychological Association, Inc.
2002, Vol. 49, No. 2, 233–242 0022-0167/02/$5.00 DOI: 10.1037//0022-0167.49.2.233
233
the methodology and the sample (e.g., CR, 1995; Howard, Kopta,
Krause, & Orlinsky, 1986; Salzer, Bickman, & Lambert, 1999;
Steenbarger, 1994). A dose–effect response curve has appeared
when studying an adult sample with general mental health prob-
lems using a general outcome measure at the end of therapy. For
example, the CR study and the Howard et al. (1986) meta-analysis
found that greater number of sessions received was associated with
better client outcome. Both of those studies had a mixture of
psychotherapy models included in their sample. However, the
dose–effect curve may be different depending upon the therapy
model (Steenbarger, 1994). Therefore, examining the dose–effect
response for different models of psychotherapy may yield impor-
tant information for evaluating the model efficiency.
In addition to investigating effectiveness, discovering if the
same psychotherapeutic processes exist in telephone counseling
that have been associated with the successful outcome in face-to-
face counseling is crucial. Understanding the processes allows for
the development of efficient, effective psychotherapy and training.
Two important process issues in the psychotherapy literature are
the therapeutic alliance and the counselor’s social influence.
From the early psychotherapy theories of Freud and Rogers,
therapeutic alliance has been central. Even though disagreement
about the function and definition of the therapeutic alliance has
flourished, researchers have consistently found it to be an impor-
tant component within the therapeutic process (Martin, Garske, &
Davis, 2000). Martin et al. found a consistent, strong relationship
between therapeutic alliance and outcome. The collaborative and
affective bond between therapist and client is considered pantheo-
retical (Horvath & Luborsky, 1993), or a common core aspect of
psychotherapy (Wampold, 2001). The research to date suggests
that therapeutic alliance should be considered mandatory for all
therapeutic approaches, including telephone counseling.
Social influence has been an important counseling process issue
since translated from social psychology to become the social
influence theory of psychotherapy (Strong, 1968). In this concep-
tual framework the psychotherapist must gain influence with the
client to be therapeutic. Like the therapeutic alliance, social influ-
ence appears to become pantheoretical because psychotherapy
requires the therapist to have some influence, notwithstanding
Strong’s two-stage model of change. The bulk of the social influ-
ence research has been analogue, which may only be relevant to
the very early psychotherapy sessions (Heppner & Claiborn,
1989). The social influence of the telephone counselor seems
important because nothing is known about the client’s perceptions
of the counselor’s influence. Additionally, the use of brief therapy
over the telephone might not give the telephone counselor much
time to create a power base for influence.
Several authors have asserted that the telephone modality pro-
vides no roadblock to interpersonal communication. For example,
Grumet (1979) postulated that telephone clients could project what
they needed or wanted onto the therapist given the absence of
visual cues. Miller (1973) asserted that telephone communication
creates the feeling of intimacy because the speaker is seemingly
only inches away from the listener’s ear. Spiro and Devenis (1991)
reported mixed experiences from the reports of psychodynamic
psychotherapists regarding the amount of transference generated
on the telephone. They reported that some clinicians found less
transference over the telephone, whereas others experienced in-
creased transference given the client’s need to fantasize to fill in
the visual gaps. With the variety of opinions, a researched inquiry
into the influence the telephone has on therapeutic processes seems
important.
The purpose of this study was to investigate whether telephone
counseling as used in the field is effective for providing general
mental health counseling and to investigate two important process
variables. For the purposes of this study, effectiveness was defined
as (a) the client’s self-reported satisfaction with telephone coun-
seling’s treatment of problems; (b) the client’s self-reported im-
provement in specific work, social, and emotional issues; and (c)
the client’s self-reported global improvement in emotional state.
Second, counseling process was investigated by examining the
counseling relationship and counselor influence in telephone coun-
seling. For the purposes of this study, the counseling relationship
was measured by the client’s perceptions of the therapeutic bond
between the client and telephone counselor, and the counselor’s
influence was measured by the client’s self-reported perceptions of
the counselor’s social influence in the telephone counseling.
Specifically, this study asks five questions: (a) Do telephone
counseling clients perceive counseling to be effective? Effective-
ness is defined as the client’s self-reported satisfaction with tele-
phone counseling’s treatment of problems, the client’s self-
reported improvement in specific work, social, or emotional issues,
and the client’s self-reported global improvement in emotional
state. (b) Do telephone counseling clients perceive a therapeutic
relationship bond with the counselor? (c) Do telephone counseling
clients perceive their counselors to have social influence (i.e., to be
expert, attractive, and trustworthy)? (d) Do the process variables of
social influence and the therapeutic relationship bond predict the
telephone counseling client’s perceived outcome? (e) Is there a
dose–response relationship in terms of the relationship between the
amount of time in telephone counseling and the general outcome
measure? The telephone counseling results in this study are com-
pared to face-to-face counseling results from other studies in the
literature to provide a context for the telephone counseling clients’
responses. Comparative face-to-face studies were selected based
upon their use of the same measures and use of actual clients.
When several studies were available in the literature, the studies
with the more positive results were used for comparison to this
study.
Method
Participants
The nonrandom sample was composed of 186 adult clients; 500 ques-
tionnaires were sent out, resulting in a response rate of 37.2%. This return
rate was considered acceptable because it was similar to the CR (Seligman,
1995) study. When CR sent out their initial questionnaire they received a
return rate of 13%. When compensating the respondents in their validation
study, they received a return rate of 38%. The CR validation study con-
sisted of comparing the answers from the two samples. They found no
differences between the data gathered from the two sets. Because our study
achieved a similar return rate as the CR study and used a similar procedure,
we argue that our responses are likely to be nonbiased as well.
The sample in this study was drawn from clients who received services
from a private telephone counseling agency that serves as an employee
assistance program for several large corporations. There were 136 female
participants (73.1%), 43 male participants (23.1%), with 7 (3.8%) not
indicating gender. The ethnic composition of the sample was as follows:
African American (n⫽14, 7.5%), Hispanic (n⫽10, 5.4%), Native
234 REESE, CONOLEY, AND BROSSART
American (n⫽6, 3.2%), White (n⫽142, 76.3%), other (n⫽7, 3.8%),
and 7 (3.8%) who did not indicate ethnicity (Mdn age ⫽36 years, SD
⫽10.18, range ⫽18–60).
The majority of the respondents (n⫽117, 62.9%) reported a family
income of less than $30,000 a year, with 55 (29.6%) making less than
$15,000 annually. Seventy-four (39.8%) of the respondents had a high
school education or less, 70 (37.6%) had attended some college, 21
(11.3%) had completed college, 10 (5.4%) had attended graduate school, 4
(2.2%) had completed graduate school, and 7 (3.8%) did not provide
information.
The respondents reported a variety of presenting problems, including
depression, anxiety, panic attacks, marital/relationship problems, family
problems, drug/alcohol, job-related/career problems, grief, eating disor-
ders, anger, and stress. The majority of the sample (n⫽131, 70.4%)
reported having more than one presenting problem, and the median number
of presenting problems for the entire sample was four (M⫽3.77,
SD ⫽2.34). Respondents also reported their level of functioning prior to
counseling: 29% said their emotional state was very poor (“I barely
managed to deal with things”); 30.6% reported a fairly poor emotional state
(“Life was usually pretty tough”); 34.9% reported a so-so emotional state;
4.3% reported that life was quite good; and 1.1% reported that life was very
good.
The sample’s median number of sessions was 3, ranging from 1 to 99
sessions (M⫽11.82, SD ⫽15.66); 46 (24.7%) respondents had only
received 1 session; 63 (33.9%) had received 2 to 4 sessions; 29 (15.6%)
had received 5 to 10 sessions; 39 (21%) had received 11 or more sessions,
and 9 (4.8%) did not specify. Most of the respondents were still receiving
telephone counseling at the time the questionnaire was completed (n⫽
138, 74.2%).
In comparing the sample to the population using the service, the agency
had a total of 15,525 people who used the service in 1998, with 67.1% of
clients being female and 32.9% being male. Approximately 79% of the
clients were first-time callers. The average number of sessions per client
was 1.5 sessions.
Telephone Counseling Agency
The telephone counseling agency provides free counseling for the em-
ployees of three large Fortune 500 companies as well as other smaller,
regional companies. The agency provides counseling for individuals across
the United States, Canada, and parts of Mexico, with over 2 million people
having access to these services. Any employee or employee’s immediate
family member has access to the agency’s toll-free number. A caller
initiates counseling by talking with an intake worker to provide initial
information and to set an appointment time with a counselor. The coun-
selor, who is not the intake worker, then calls the person back at a
scheduled time for the session. At the end of the session the client is
transferred back to the intake worker to schedule the next appointment with
the same counselor. There is no limit to the number of sessions a client may
have; the average session lasts 30 min.
The agency provides counseling for a variety of issues including, but not
limited to, depression, anxiety, marital and relationship problems, grief,
and work-related difficulties. Callers with problems that are more severe or
chronic in nature are encouraged to seek local mental health services and
are advised on how to obtain appropriate help. Although the service
provides crisis intervention for callers, it is primarily considered a coun-
seling service and not a “crisis line.”
The agency provides their telephone counselors with training and su-
pervision in the use of solution-focused therapy (SFT; De Jong & Berg,
1998). SFT, based on the work of de Shazer (1985) and Berg (1995), is a
brief, strength-based model. This approach focuses on the individual’s
strengths rather than on the problematic behaviors or their causes. SFT
focuses on times when the problematic behaviors are not occurring to
discover a person’s strengths in promoting change (de Shazer, 1985). SFT
claims to respond to individual differences (Robbins, Alexander, Newell,
& Turner, 1996) while taking into consideration the systemic whole (Beye-
bach & Carranza, 1997). The length of SFT has been typically around six
sessions (Lee, 1997; Zimmerman, Jacobsen, MacIntyre, & Watson, 1996;
Zimmerman, Prest, & Wetzel, 1997). Counselor adherence to SFT was not
assessed.
Counselors
All counselors working for the telephone counseling agency have a
master’s degree and are licensed in the field of psychology, counseling,
marriage and family therapy, or social work. Each counselor receives
training in SFT and live supervision to ensure that he or she is capable of
implementing the solution-focused approach with clients. All counselors
receive weekly supervision.
Instrumentation
Consumer Reports Annual Questionnaire (CRAQ; CR, 1994). The
CRAQ was used in the CR (1995) effectiveness study and was chosen for
this study so that comparisons could be made for therapeutic effectiveness.
This study used the 10 mental health questions of the CRAQ, which were
totaled for an overall effectiveness score and can be used as three sub-
scales: Specific Improvement (“How much did treatment help with the
specific problem that led you to counseling?”), Satisfaction (“Overall how
satisfied were you with the counselor’s treatment of the problems?”), and
Global Improvement, or how respondents described their “overall emo-
tional state”at the time the questionnaire was completed compared with at
the start of treatment (Seligman, 1995). The Specific Improvement sub-
scale uses a 5-point Likert-type response, the Satisfaction subscale uses a
6-point Likert-type response, and the Global Improvement subscale uses a
5-point Likert-type response. The three subscales are weighted on a 0–100
scale and combined for a composite scale of 0–300 for total effectiveness.
The methodology for score transformation is identical to that of the CR
study (M. Kotkin, personal communication, September 1998).
No information on reliability or validity is available on the CRAQ from
the CR (1995) study. However, reliability using the total effectiveness data
from our sample achieved a coefficient alpha (internal consistency coeffi-
cient) of .72. Although there is no validity study as such, construct validity
of the CRAQ can be inferred because the measure responds as expected in
the CR study. Perhaps the most persuasive finding for validity purposes
was the dose–response curve that was found in the CR (1995) study using
the CRAQ, where the rate of improvement was found to be a function of
length of treatment. The CR finding paralleled the dose–response curve of
general mental health services that used other measures of outcome
(Howard et al., 1986).
A confirmatory factor analysis using these data supports the validity of
the CRAQ Specific Improvement subscale. (The Global Improvement and
Satisfaction subscales each had only one item so they were not included in
the factor analysis.) The eight items that make up the Specific Improve-
ment subscale of the CRAQ had a comparative fit index (Bentler, 1990)
coefficient of .993 (the closer to 1, the better the fit). The correlation
coefficients between the Specific Improvement subscale score and the
scale items ranged from .67 to .83 (intercorrelations between the items of
the Specific Improvement subscale are .33–.69). The intercorrelations
between the CRAQ subscales are: Global Improvement and Satisfaction,
.28; Global Improvement and Specific Improvement, .42; and Specific
Improvement and Satisfaction, .43.
In summary, the CRAQ Specific Improvement subscale appears to be
made up of items that measure a similar construct. All of the CRAQ
subscales have a strong correlation with each other, but they also appear to
include some unique parts of the construct. Therefore, the psychometric
properties of the data are consistent with a valid measure.
Working Alliance Inventory—Bond Scale (WAI–B). The client version
of the WAI–B (Horvath & Greenberg, 1986) consists of 12 questions
235
TELEPHONE COUNSELING
referring to the strength of the interpersonal relationship. It is one of three
subscales from the WAI, which has 36 questions, 12 each for the Tasks,
Goals, and Bond subscales, respectively. The questions are rated on a
7-point Likert-type response scale ranging from never (1) to always (7).
The WAI–B was used to assess the interpersonal aspect of the therapeutic
alliance.
Reliability estimates for the WAI–B have been generally adequate.
Internal consistency coefficients have ranged from .68 to .92 in previous
studies, as measured by Cronbach’s alpha (e.g., Horvath & Greenberg,
1989; Plotnicov, 1990). The Task, Goal, and Bond subscales have had
strong intercorrelations in previous studies, ranging between the low .60s
to the high .80s (e.g., Horvath & Greenberg, 1989). Coefficients for
test–retest reliability have ranged between .66 and .74 for each of the
subscales (Plotnicov, 1990). Using coefficient alpha, the internal consis-
tency for our sample was .69.
Previous studies have provided evidence of convergent validity for the
WAI–B (e.g., Horvath & Greenberg, 1989; Safran & Wallner, 1991). For
example, the Bond subscale possessed a correlation of .72 with the Cali-
fornia Psychotherapy Alliance Scales, demonstrating reasonable conver-
gent validity (Safran & Wallner, 1991). Results for discriminant validity,
however, are mixed. Discriminant validity has been investigated by con-
trasting the WAI to other inventories that measure the therapeutic alliance
differently, such as the Counselor Rating Form (CRF; LaCrosse, 1980) and
the Empathy subscale of the Relationship Inventory (Barrett-Lennard,
1962). Horvath and Greenberg (1989) constructed a multitrait–
multimethod matrix with the WAI subscales and the Empathy subscale.
The WAI–B and the Empathy subscale had a correlation of .83, suggesting
little discriminant validity. In a second study, however, evidence for
discriminant validity was found with correlations of .38, .28, and .05
between the WAI–B and the CRF subscales of Attractiveness, Expertness,
and Trustworthiness, respectively (Horvath, 1981).
Counselor Rating Form—Short Version (CRF–S). The CRF–S (Cor-
rigan & Schmidt, 1983) measures the client’s view of the counselor’s
influence. The measure consists of 12 adjectives that are subsumed under
the factors of Expertness, Attractiveness, and Trustworthiness. The adjec-
tives are rated on a 7-point Likert-type scale ranging from not very (1) to
very (7). Each factor makes up a four-item subscale on the CRF–S.
Reliability for the CRF–S has generally been shown to be acceptable,
with the three subscale coefficients ranging from .82 to .94 (Mdn ⫽.91;
Corrigan & Schmidt, 1983). Epperson and Pecnik (1985) reported more
conservative estimates of internal consistency for the subscales with coef-
ficients ranging from .63 to .89 (Mdn ⫽.82). Using coefficient alpha, the
internal consistency for the Expertness, Attractiveness, and Trustworthi-
ness subscales for the current sample was .95, .96, and .96, respectively.
Construct validity findings for the CRF–S are mixed. Initially, Corrigan
and Schmidt (1983) found three factors utilizing factor analysis that re-
flected the hypothesized constructs of expertness, attractiveness, and trust-
worthiness. Subsequent validation studies have only found two factors
(Attractiveness–Trustworthiness and Expertness–Trustworthiness) indi-
cated by factor-analytic methods (Tryon, 1987; Wilson & Yager, 1990).
Demographic questionnaire. Participants were asked questions regard-
ing their gender, ethnicity, age, length of time in counseling, and whether
counseling was completed. An additional CR (1995) question was asked
regarding emotional state prior to counseling.
Procedure
The telephone intake worker asked each agency client to participate in
the study for approximately 3 weeks. Some of the clients received the
questionnaires prior to completion of telephone counseling. Both first-time
callers and established clients were included. The telephone counseling
clients received the questionnaires through the mail after receiving at least
one telephone counseling session. A first-time caller speaks with the intake
worker just prior to telephone counseling with the ongoing counselor. An
established client speaks with the intake worker after the telephone coun-
seling session to schedule the next appointment. The only stipulation for
client inclusion was that he or she must have been calling about a problem
related to mental health, relationships, or job/career and been at least 18
years old.
A packet containing the questionnaires was mailed by the telephone
counseling service to each participant. The packet was received about a
week after the recruitment session. The participants filled out the packet,
indicated a return address, and mailed the completed packet to Robert J.
Reese, who is not associated with the telephone counseling agency. The
packet took clients approximately 20–25 min to complete. Participants
who returned the completed questionnaires were eligible for a prize draw-
ing (one $100 prize and two $50 prizes).
Results
Several analyses were performed to explore the structure of the
data. At the time of responding to the questionnaires some clients
(26%) had completed telephone counseling, but the rest had not.
Analyses were performed to determine if completion of telephone
counseling made a difference in the perception of effectiveness,
social influence, or therapeutic bond. A multivariate analysis of
variance (MANOVA) was computed comparing individuals who
had completed counseling to those who had not on the three
effectiveness subscales and the total effectiveness score (Wilks’s
⫽.99), F(3, 178) ⫽.263, p⫽.85,
2
⫽.004, and the WAI–B
and the CRF (Wilks’s
⫽.98), F(2, 164) ⫽1.405, p⫽.25,
2
⫽
.017. No statistically significant differences were found. The re-
sults indicate that clients who had finished telephone counseling
did not answer differently than those who were continuing coun-
seling. Combining the sample was justified, and comparing this
data to data from studies with clients who had completed coun-
seling also was justified based upon this analysis.
Question 1: Client Perceptions of Telephone Counseling’s
Effectiveness
To aid in understanding the effectiveness ratings of the tele-
phone counseling sample of this study, we used the CRAQ data
from the CR (1995) face-to-face counseling study for comparison
to the current telephone counseling data. The CR study was se-
lected as a comparison study because of the similarities between
the CR study and our telephone counseling study. Both studies
used actual clients as the sample, the CRAQ for measuring effec-
tiveness and data gathered through the mail. The CR study was
different because it had a higher socioeconomic sample (middle vs.
lower), a sample that probably included only clients who finished
contact with their counselor and clients who reported a higher level
of functioning prior to counseling (see Figure 1). The level of
functioning was significantly worse in the telephone counseling
sample (Somers’sdfor ordinal data ⫽.113, p⫽.000, Spearman
r⫽.154). Statistical options are limited because Consumers
Union, the publisher of CR, was unwilling to provide standard
deviations. The telephone sample means and standard deviations
as well as the CR (1995) means for the three effectiveness sub-
scales and the total effectiveness scores are reported in Table 1.
Specific improvement. When asked about the specific problem
that led to telephone counseling, 30% endorsed feeling “a lot
better,”53% reported doing “somewhat better,”and 15% reported
“no difference.”Over 80% of the participants felt the specific
236 REESE, CONOLEY, AND BROSSART
problem that led them to counseling had improved. The Specific
Improvement mean scores reported by the telephone counseling
group were slightly lower than the face-to-face counseling scores
(77.4 to 82.0). Figure 2 displays the frequency of response com-
parisons between the telephone counseling data and the CR (1995)
percentages for responses in the categories that CR supplied. Using
the frequency data the CR face-to-face counseling results were
statistically significantly higher than the telephone counseling
data, but the effect size was small (Somers’sdfor ordinal data ⫽
.068, p⫽.003, Spearman r⫽.096).
Global improvement. When the telephone counseling clients
were asked how they felt at the time they completed the question-
naire compared with how they felt before therapy, 25.1% of the
participants indicated that they were doing “quite good,”
with 4.4% doing “very good.”The majority of clients reported
doing “so-so”(52.5%), and 18.1% of the sample reported feeling
“poor.”The Global Improvement mean scores reported by the
telephone counseling group were slightly lower than the face-to-
face counseling scores (61.5 to 64.0; see Table 1).
Satisfaction. Over two thirds of the telephone counseling sam-
ple (68%) reported being either “very satisfied”or “completely
satisfied”with telephone counseling. Another 27% were “fairly
well satisfied,”and 5% indicated dissatisfaction. The Satisfaction
mean scores reported by the telephone counseling group were
slightly higher than the face-to-face counseling scores (79.4
to 75.0; see Table 1). Figure 3 contains the comparative percent-
ages between the telephone counseling and the CR (1995) re-
sponses. Using the frequency data, the telephone counseling re-
sults were statistically significantly higher than the CR face-to-face
counseling data but the effect size was small (Somers’sdfor
ordinal data ⫽.051, p⫽.042, Spearman r⫽.064).
Effectiveness ⫻Initial Functioning. The telephone counseling
clients were asked to rate their level of functioning before they
began counseling. An analysis of variance (ANOVA) was used on
the CRAQ total effectiveness scores to compare the three groups
of initial functioning (very poor, M⫽231, SD ⫽37.5; fairly poor,
M⫽214, SD ⫽38.7; so-so or better, M⫽205, SD ⫽30.7). The
differences were statistically significant, F(2, 182) ⫽8.99, p⫽
Figure 1. Comparison of the telephone counseling and the Consumer
Reports (1995) percentile responses to level of functioning prior to
counseling.
Table 1
Descriptive Statistics of the Telephone Counseling and
Consumer Reports Samples Using the Consumer Reports
Annual Questionnaire (CRAQ) Measures of
Psychotherapeutic Effectiveness
Measure
Telephone counseling Consumer
Reports
nMSDnM
Specific Improvement 174 77.40 19.13 2,606 82
Global Improvement 183 61.50 13.93 2,606 64
Satisfaction 185 79.35 18.99 2,606 75
Total Effectiveness 171 218.32 39.00 2,606 221
Figure 2. Comparison of the telephone counseling and the Consumer
Reports (1995) percentile responses to how much counseling helped the
problem that led to counseling.
Figure 3. Comparison of the telephone counseling and the Consumer
Reports (1995) percentile responses to overall satisfaction with counseling.
237
TELEPHONE COUNSELING
.000,
2
⫽.09. Telephone counseling clients who judged them-
selves to be doing the poorest prior to counseling rated total
effectiveness higher than the fairly poor or the so-so or better
groups, using the Tukey post hoc statistic.
Comparing the telephone counseling to the CR (1995) respon-
dents’specific problem improvement that led to counseling by
their level of functioning data reveals a shortcoming in the tele-
phone counseling (see Figure 4). For respondents who rated their
functioning at the very poor level, the CR data had more responses
at the “counseling helped a lot”and “helped somewhat”levels than
did the telephone counseling sample (Somers’sd⫽.052, p⫽
.000, Spearman r⫽.12). Similarly, for respondents who rated
their functioning at the fairly poor level, the CR data had more
responses at the “counseling helped a lot”and “helped somewhat”
levels than did the telephone counseling sample (Somers’sd⫽
.020, p⫽.009, Spearman r⫽.047).
Question 2: The Counselor’s Social Influence
The second question addressed how telephone counseling cli-
ents perceived their counselor’s social influence characteristics of
expertness, attractiveness, and trustworthiness as measured by the
CRF–S (Corrigan & Schmidt, 1983; Table 2 presents the descrip-
tive statistics).
To aid in the interpretation of the telephone counseling data, we
compared the social influence scores of this study to those of
another study with individuals who received traditional, face-to-
face therapy. The face-to-face counseling study utilized for com-
parison was the original Corrigan and Schmidt (1983) study con-
taining 155 clients who saw experienced community counselors.
Their study was chosen because it had the highest mean scores
compared with the other studies reviewed, which sets the highest
standards for comparison purposes. Four independent ttests and
effect sizes were computed to compare telephone counseling and
face-to-face counseling for differences in mean scores on the
Expertness, Attractiveness, and Trustworthiness subscales and the
total score on the CRF–S. Telephone counseling respondents rated
their counselors no differently than did individuals who saw ex-
perienced therapists face-to-face in community health centers.
None of the ttests were statistically significant (p⬎.05) and
standardized effect sizes were very small (Cohen, 1988), as seen in
Table 2. All of the CRF–S scores in Table 2 are near the maximum
positive scores for Expertness, Attractiveness, and Trustworthi-
ness; the possible range is 4 to 28 for these subscales.
Lawson, Gaushell, McCune, and McCune (1995) provided an-
other manner of understanding the level of social influence that the
telephone counseling CRF–S scores represent. In their study, the
counselors in the highest credibility group were those who re-
ceived CRF–S total scores between 72 and 82. The telephone
counselors would be members of the highest credibility group,
with 76.66 as the mean CRF–S score.
Question 3: Counseling Relationship
The purpose of the third question was to assess whether a
therapeutic bond was perceived by the telephone counseling client.
To aid in the interpretation of the telephone counseling data, we
compared the WAI–B scores of our study with those from another
study with individuals who received traditional face-to-face ther-
apy. The face-to-face counseling study used for comparison was
the Mallinckrodt, Coble, and Gantt (1995) study because it pro-
vided a mean and standard deviation for the Bond subscale, pro-
vided data from real sessions, and had the highest mean of the
other studies reviewed, which provided the highest standard for
comparison. The sample was composed of 76 women who re-
ceived, on average, eight sessions either at a university counseling
center, a community college counseling center, an outpatient hos-
pital clinic, or a training clinic for a graduate program.
The mean Bond score for the telephone counseling sample
(M⫽72.14, SD ⫽9.79) was a little higher than the normative
sample mean (M⫽70.34, SD ⫽9.58), but the difference was not
statistically significant, t(245) ⫽1.33; the standardized effect size
was small (d⫽.14).
Question 4: Therapeutic Alliance, Social Influence, and
Outcome
The fourth question addressed the relationship between the
therapeutic alliance, social influence, and outcome. Multiple re-
gression was used to examine the ability of the CRF–S total score
and the WAI–B (Bond) score to predict the criterion variable, the
Total Effectiveness score from the CRAQ.
Both the WAI–B and the CRF–S were moderately correlated
with the Total Effectiveness score, r⫽.49, (pⱕ.001) and r⫽
.34, (pⱕ.001), respectively. The two predictors also were highly
correlated with one another, r⫽.56 (pⱕ.001). Both variables in
the regression equation were statistically significant in predicting
total effectiveness, F(2, 152) ⫽22.69, pⱕ.001. The R
2
was .23
(adjusted R
2
⫽.22), which is considered a medium-to-large effect
size. Both variables accounted for nearly a quarter of the total
variance. The standardized beta weights and structure coefficients
indicate that both predictors do well, especially as indicated by the
structure coefficients, .79 for the CRF–S and .94 for the WAI–B.
When the structure coefficients are squared, the amount of vari-
ance that each predictor accounts for in the predicted criterion can
be seen (Thompson, 1992). The WAI–B explains .88 of the total
criterion variance and the CRF–S accounts for .62 of the variance.
Figure 4. The telephone counseling and the Consumer Reports (1995)
percentile responses to how much counseling helped the primary present-
ing problem for two initial levels of functioning prior to counseling.
238 REESE, CONOLEY, AND BROSSART
Each of the predictors is correlated with each other, which explains
why the sum of the predictors’squared structure coefficients
exceeds 100%.
Question 5: Effectiveness ⫻Treatment Length
Clients who received telephone counseling for a year or longer
reported more improvement than clients receiving less time in
treatment. Clients’mean total effectiveness score was 213.97
(SD ⫽37.02) for those in counseling 2 months or less, 215.21
(SD ⫽33.61) for those in counseling 3 to 6 months, and 220.38
(SD ⫽36.74) for those in counseling 7 to 11 months. The effec-
tiveness score increased to 243.53 (SD ⫽33.22) for those who had
been in counseling 12 months or longer. An ANOVA indicated a
statistically significant difference, F(3, 178) ⫽4.48, p⬍.01. A
Scheffe´post hoc analysis found a statistically significant differ-
ence for respondents who had been in counseling 12 months or
longer compared with respondents who had received counseling
for shorter periods of time.
Discussion
The results of the study support the utility of telephone coun-
seling for general mental health concerns using an SFT model.
Telephone counseling tended to be viewed by clients as effective
and satisfactory. The telephone counseling clients tended to per-
ceive the counselor’s influence and relationship bond as strong.
The more successful telephone counseling outcomes were associ-
ated with higher estimations of counselor’s social influence and
higher estimations of the counseling relationship bond. Finally,
there is evidence that SFT telephone counseling that lasts longer
than 12 months is more beneficial than shorter term counseling.
Treatment Effectiveness
The most important finding was that respondents reported sat-
isfaction with their telephone counseling experience and believed
that telephone counseling helped them improve their lives. The
positive findings are consistent with case studies of telephone
counseling (e.g., Grumet, 1979) and the few empirical studies
available (e.g., Lynch, Tamburrino, & Nagel, 1997; Zhu et al.,
1996).
This study begins to answer some of the concerns about tele-
phone counseling. The ethical concerns (e.g., Haas et al., 1996)
about outcome that have haunted telephone counseling perhaps
now will evolve into investigations addressing more specific out-
come questions and process questions. Although telephone coun-
seling has been used for limited applications (e.g., crisis counsel-
ing), telephone counseling for general mental health concerns may
open up new vistas for counseling services. Telephone counsel-
ing’s natural niche may be to help individuals who may not
otherwise receive counseling. In this study approximately two
thirds of the respondents indicated that their total family income
was under $30,000, with only 35% of the respondents in the study
having insurance for mental health services. Probably most of the
respondents would not have been able to receive face-to-face
counseling because of the cost.
The comparisons of telephone counseling with face-to-face
counseling revealed some strengths and concerns. In compari-
son to the CR (1995) data, the telephone counseling received
statistically significant higher satisfaction scores but the effect
sizes were small. The face-to-face counseling received statisti-
cally significant higher scores on improvement for the specific
problem that led to counseling but the effect size was small.
Because the effect sizes were small these results are seen as
tenuous and needing further research support, and they are not
discussed here.
A concern surfaced about the effectiveness for clients who
began counseling feeling very poorly. When compared to the
face-to-face counseling (CR, 1995) results, the telephone counsel-
ing respondents who reported functioning very poorly at the be-
ginning of counseling reported significantly less improvement
(based on the Specific Improvement subscale). This finding was
accompanied by a small-to-moderate effect size.
The poorer results of telephone counseling suggest the need for
assessing a client’s level of functioning to determine the suitability
for telephone treatment. The finding also indicates the importance
of a referral network of face-to-face counseling in conjunction
with, or instead of, telephone counseling. Additionally, this finding
highlights an area in which telephone counseling may need to
improve.
Effectiveness ⫻Treatment Length
The telephone counseling clients who used the service for over
a year reported higher functioning than those who used it for less
time. Our results can be considered as similar to the dose–response
curves found by Howard et al. (1986) and in the CR (1995) study,
where the rate of improvement was found to be a function of
treatment length.
Table 2
Descriptive t-test Statistics Comparing the Telephone Counseling and the Corrigan
and Schmidt (1983) Samples Using the Counselor Rating Form—Short Version (CRF–S)
Measure of Counseling Social Influence
CRF–S measure
Telephone sample Face-to-face sample
tdnMSDnMSD
Expertness 183 24.87 4.63 155 24.82 3.46 0.11 .02
Attractiveness 185 25.92 3.64 155 25.42 3.85 1.25 .13
Trustworthiness 181 25.76 4.25 155 25.96 3.56 0.48 .05
Total 181 76.66 11.97 155 76.19 9.76 0.40 .03
239
TELEPHONE COUNSELING
Counseling Processes
The telephone counseling processes examined in this study
appeared to function no differently than traditional psychotherapy
processes. The telephone counselor was viewed as emotionally
invested and influential in the counseling relationship. When we
compared the telephone counseling data gathered in this study to
face-to-face studies, there was no difference in amount of thera-
peutic bonding or social influence. These findings are important
because the most consistent predictor of outcome has been the
therapeutic relationship between the client and counselor. Addi-
tionally, these findings suggest that the process measures of ther-
apeutic bond and social influence can be meaningfully used to
investigate telephone counseling further.
The lack of visual cues could change counselor training tech-
niques for telephone counseling as compared with face-to-face
counseling. For example, Heppner and Claiborn (1989) surmised
that the variables influencing clients’early ratings of social influ-
ence were nonverbal behaviors, most of which were visual. The
telephone removes many of the counselor’s opportunities for
amassing social influence through visual means. The voice of the
telephone counselor communicates what an office, clothes, and
physical appearance typically offer a counselor. Training the tech-
niques of telephone counseling may provide extra focus on use of
the voice. Future research may find that focus on the use of voice
may have a salutary effect on face-to-face counseling as well. Or,
future research may find that with the visual stimuli removed, the
telephone client becomes more focused on what the therapist says.
Limitations of Study
An effectiveness study examines psychotherapy in the field,
which contributes to external validity. An effectiveness study
provides valuable information about how counseling is actually
performed and provided to clients who have a wide variety of
problems. However, effectiveness studies are known for weak-
nesses in internal validity. This study attempted to remedy some of
those inherent limitations.
The central limitation of this study is the lack of a treatment
integrity measure. We have no evidence that the telephone coun-
selors provided SFT adequately beyond the assurances of the
supervisors. However, this study did investigate telephone coun-
seling as actually performed when not controlled by researchers.
The counselors were supervised once a week.
Other limitations inherent to an effectiveness study were ad-
dressed to ameliorate their negative influences: potential sampling
bias, the lack of a control group, and the number of people sampled
who had not completed therapy. Survey studies suffer from the
threat of sampling bias because random sampling does not occur.
In this study every individual who called to use telephone coun-
seling, except crisis callers and persons under age 18, was given
the opportunity to participate in the study; only 8 persons initially
declined to participate. Seligman (1995) described the CR proce-
dure to test the validity of their sample. In their initial uncompen-
sated response questionnaire, they received a return rate of 13%.
When compensating the respondents in their validation study they
received a return rate of 38% and there were no differences
between the data gathered from the two sets. Our return rate was
a similar 37% using a similar approach to that of CR.
Other indications that the responders were similar to the typical
users of the telephone counseling service were the demographic
comparisons. The similarities are in the female to male respon-
dents, the breadth of the presenting problems, and the high fre-
quency of clients using telephone counseling two or fewer times.
A second concern is the lack of control groups. This is an
inherent flaw for an effectiveness study in comparison to an
efficacy study. However, we dealt with that limitation by compar-
ing our results to those of similar studies. In the case of the process
variables, the highest scoring comparison groups from the litera-
ture substituted for control groups.
Our sample may systematically differ from the comparison
groups in some way other than face-to-face and telephone coun-
seling that influenced the assessment scores. An obvious differ-
ence was the income level of the clients. Our sample was com-
posed primarily of people earning less than $30,000 a year and
with less than a college education. The CR study sampled persons
who were likely to be college graduates and have higher incomes.
These issues should have made for conservative claims for tele-
phone counseling. If socioeconomic status did effect treatment or
measures, the literature suggests that lower socioeconomic status
places more hardships on persons, which may suppress their ability
to use therapy effectively.
Another characteristic of this sample was the mix of individuals
who had completed (26%) and not yet completed (74%) telephone
counseling. However, as noted previously, there was no statistical
difference and only a small effect size when comparing the scores
of respondents who had completed counseling with those who had
not. Therefore, the mix of individuals who had not completed
counseling does not appear to be a limitation.
Further research is necessary to substantiate the findings pre-
sented here. Future outcome studies need to focus on the level of
initial functioning when comparing telephone counseling to face-
to-face counseling. The telephone counseling data presented in this
study reveal that clients starting with a poorer level of functioning
gained more from counseling. However, the data show that face-
to-face counseling (CR, 1995) is more helpful for poor functioning
clients than telephone counseling.
Process studies examining the techniques used in telephone
counseling seem especially important because of ethical issues.
Our ethical standards require us to provide service only in areas in
which we are competent; process studies could identify skills that
are necessary and possibly unique to telephone psychotherapy. It
may be that certain techniques should be avoided in telephone
counseling.
Examining the theoretical orientation used in telephone coun-
seling could be important. Perhaps some counseling approaches
are not suitable to telephone counseling if they contain techniques
that could endanger the therapeutic alliance. For example, it may
be that a psychodynamic therapist may have difficulty with a client
when interpretations are received negatively by the client.
Another area of future research is the possible interaction of
client socioeconemic status and telephone counseling. Wark
(1982) noted that a potential advantage of telephone counseling
was the absence of visual biases for both the client and the
counselor. For example, neither the physical characteristics of the
therapist or the client could negatively influence the counseling
relationship over the telephone. Also, the office surroundings (e.g.,
diplomas on the wall, a nice office) or clothing could negatively
240 REESE, CONOLEY, AND BROSSART
influence the relationship, based on a perceived inequality in
power, values, or socioeconomic status. The removal of these
potential biases may have been important in this study, given the
low socioeconomic status of the sample.
In conclusion, the good news about telephone counseling is that
over 82% of the respondents reported specific improvement on
their presenting problem. Clients reported strong levels of satis-
faction with telephone counseling. Therapeutic relationships and
influence processes appear to occur at efficacious levels over the
telephone. The therapeutic outcome of counseling was predicted
by the process variables that are predictive for face-to-face coun-
seling. The final good news is that persons of a socioeconomic
status that does not typically have money or insurance for face-
to-face counseling used telephone counseling. We are uneasy
about the fate of persons who view themselves as doing very
poorly prior to treatment. Telephone counseling clients who rated
themselves as doing very poorly prior to treatment did not report
that counseling was as helpful as did clients in face-to-face coun-
seling. Clearly, the processes involved in telephone counseling
using SFT merit and require continued study.
References
Barrett-Lennard, G. T. (1962). Dimensions of therapist response as causal
factors in therapeutic change. Psychological Monographs, 76(43, Whole
No. 562).
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psy-
chological Bulletin, 107, 238–246.
Berg, I. K. (1995). Solution-focused brief therapy with substance abusers.
In A. M. Washton (Ed.), Psychotherapy and substance abuse (pp.
223–242). New York: Guilford Press.
Beyebach, M., & Carranza, V. E. (1997). Therapeutic interaction and
dropout: Measuring relational communication in solution-focused ther-
apy. Journal of Family Therapy, 19, 173–212.
Brock, T. C., Green, M. C., Reich, D. A., & Evans, L. M. (1996). The
Consumer Reports study of psychotherapy: Invalid is invalid. American
Psychologist, 51, 1083.
Buie, J. (1989). Therapy by telephone: Does it help or hurt? APA Monitor,
20(3), 14–15.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences
(2nd ed.). Hillsdale, NJ: Erlbaum.
Consumer Reports. (1994). Annual questionnaire.
Consumer Reports. (1995, November). Mental health: Does therapy help?
734–739.
Corrigan, J. D., & Schmidt, L. D. (1983). Development and validation of
revisions in the Counselor Rating Form. Journal of Counseling Psychol-
ogy, 30, 64–75.
De Jong, P., & Berg, I. K. (1998). Interviewing for solutions. Pacific
Grove, CA: Brooks/Cole.
de Shazer, S. (1985). Keys to solution in brief therapy. New York: Norton.
Epperson, D. L., & Pecnik, J. A. (1985). Counselor Rating Form—Short
version: Further validation and comparison to the long form. Journal of
Counseling Psychology, 32, 143–146.
Fish, S. L. (1990). Therapeutic uses of the telephone: Crisis intervention vs.
traditional therapy. In G. Gumpert & S. L. Fish (Eds.), Talking to
strangers: Mediated therapeutic communication (pp. 154–169). Nor-
wood, NJ: Ablex.
Grumet, G. W. (1979). Telephone therapy: A review and case report.
American Journal of Orthopsychiatry, 49, 574–584.
Haas, L. J., Benedict, J. G., & Kobos, J. C. (1996). Psychotherapy by
telephone: Risks and benefits for psychologists and consumers. Profes-
sional Psychology: Research and Practice, 27, 154–160.
Heppner, P. P., & Claiborn, C. D. (1989). Social influence research in
counseling: A review and critique. Journal of Counseling Psychol-
ogy, 36, 365–387.
Hines, M. H. (1994). Using the telephone in family therapy. Journal of
Marital and Family Therapy, 20, 175–184.
Horvath, A. O. (1981). An exploratory study of the working alliance: Its
measurement and relationship to outcome. Unpublished doctoral disser-
tation, University of British Columbia, Vancouver, British Columbia,
Canada.
Horvath, A. O., & Greenberg, L. S. (1986). The development of the
Working Alliance Inventory. In L. S. Greenberg & W. M. Pinsoff (Eds.),
The psychotherapeutic process: A research handbook (pp. 529–556).
New York: Guilford Press.
Horvath, A. O., & Greenberg, L. S. (1989). Development and validation of
the Working Alliance Inventory. Journal of Counseling Psychology, 36,
223–233.
Horvath, A. O., & Luborsky, L. (1993). The role of the therapeutic alliance
in psychotherapy. Journal of Consulting and Clinical Psychology, 61,
561–573.
Howard, K. I., Kopta, S. M., Krause, M. S., & Orlinsky, D. E. (1986). The
dose–effect relationship in psychotherapy. American Psychologist, 41,
159–164.
Howard, K. I., Moras, K., Brill, P. L., Martinovich, Z., & Lutz, W. (1996).
Evaluation of psychotherapy: Efficacy, effectiveness, and patient
progress. American Psychologist, 51, 1059–1064.
Jerome, L. W., DeLeon, P. H., James, L. C., Folen, R., Earles, J., &
Gedney, J. J. (2000). The coming of age of telecommunications in
psychological research and practice. American Psychologist, 55, 407–
421.
LaCrosse, M. B. (1980). Perceived counselor social influence and coun-
seling outcomes: Validity of the Counselor Rating Form. Journal of
Counseling Psychology, 27, 320–327.
Lawson, D. M., Gaushell, W. H., McCune, S. L., & McCune, E. D. (1995).
Perception of counselor behavior and current intergenerational family
relationships. Counselor Education and Supervision, 34, 356–368.
Lee, M. (1997). A study of solution-focused brief family therapy: Out-
comes and issues. The American Journal of Family Therapy, 25, 3–17.
Lynch, D. J., Tamburrino, M. B., & Nagel, R. (1997). Telephone counsel-
ing for patients with minor depression: Preliminary findings in a family
practice setting. The Journal of Family Practice, 44, 293–298.
Mallinckrodt, B., Coble, H. M., & Gantt, D. L. (1995). Working alliance,
attachment memories, and social competencies of women in brief ther-
apy. Journal of Counseling Psychology, 42, 79–84.
Martin, D. J., Garske, J. P., & Davis, M. K. (2000). Relation of the
therapeutic alliance with outcome and other variables: A meta-analytic
review. Journal of Consulting and Clinical Psychology, 68, 438–450.
Mays, V. M., & Albee, G. W. (1992). Psychotherapy and ethnic minorities.
In D. K. Freedheim (Ed.), History of psychotherapy: A century of change
(pp. 552–570). Washington, DC: American, Psychological Association.
Mermelstein, H. T., & Holland, J. C. (1991). Psychotherapy by telephone:
A therapeutic tool for cancer patients. Psychosomatics, 32, 407–412.
Miller, W. B. (1973). The telephone in outpatient psychotherapy. American
Journal of Psychotherapy, 27, 15–26.
Orleans, C. T., Schoenbach, V. J., Wagner, E. H., Quade, D., Salmon,
M. A., Pearson, D. C., et al. (1991). Self-help quit smoking interven-
tions: Effects of self-help materials, social support instructions, and
telephone counseling. Journal of Consulting and Clinical Psychol-
ogy, 59, 439–448.
Plotnicov, K. H. (1990). Early termination from counseling: The client’s
perspective. Unpublished doctoral dissertation, University of Pittsburgh.
Ranan, W., & Blodgett, A. (1983). Using telephone therapy for “unreach-
able”clients. Social Casework: The Journal of Contemporary Social
Work, 64, 39–44.
Robbins, M. S., Alexander, J, F., Newell, R. M., & Turner, C. W. (1996).
241
TELEPHONE COUNSELING
The immediate effect of reframing on client attitude in family therapy.
Journal of Family Psychology, 10, 28–34.
Rosenfield, M. (1997). Counselling by telephone. London: Sage.
Safran, J. D., & Wallner, L. K. (1991). The relative predictive validity of
two therapeutic alliance measures in cognitive therapy. Psychological
Assessment, 3, 188–195.
Salzer, M. S., Bickman, L., & Lambert, E. W. (1999). Dose–effect rela-
tionship in children’s psychotherapy services. Journal of Consulting and
Clinical Psychology, 67, 228–238.
Seligman, M. E. P. (1995). The effectiveness of psychotherapy: The
Consumer Reports study. American Psychologist, 50, 965–974.
Shepard, P. (1987). Telephone therapy: An alternative to isolation. Clinical
Social Work Journal, 15, 56–65.
Spiro, R. H., & Devenis, L. (1991). Telephone therapy: Enhancement of
the psychotherapeutic process. Psychotherapy in Private Practice, 9,
31–55.
Steenbarger, B. N. (1994). Duration and outcome in psychotherapy: An
integrative review. Professional Psychology: Research and Practice, 25,
111–119
Strong, S. R. (1968). Counseling: An interpersonal influence process.
Journal of Counseling Psychology, 15, 215–224.
Thompson, B. (1992, April). Interpreting regression results: Beta weights
and structure coefficients are both important. Paper presented at the
meeting of the American Educational Research Association, San
Francisco.
Tryon, G. S. (1987). The Counselor Rating Form—Short version: A factor
analysis. Measurement & Evaluation in Counseling Development, 20,
122–126.
Ulbrich, P. M., Warheit, G. J., & Zimmerman, R. S. (1989). Race, socio-
economic status, and psychological distress: An examination of differ-
ential vulnerability. Journal of Health and Social Behavior, 30, 131–
146.
Wampold, B. E. (2001). The great psychotherapy debate: Models, meth-
ods, and findings. Mahwah, NJ: Erlbaum.
Wark, V. (1982). A look at the work of the telephone counseling center.
The Personnel and Guidance Journal, 61, 110–112.
Williams, S. (2000). How is telehealth incorporated into psychology prac-
tice? APA Monitor, 31(4), 15.
Wilson, F. R., & Yager, G. G. (1990). Concurrent and construct validity of
three counselor social influence instruments. Measurement and Evalu-
ation in Counseling and Development, 23, 52–66.
Zhu, S.-H., Stretch, V., Balabanis, M., Rosbrook, B., Sadler, G., & Pierce,
J. P. (1996). Telephone counseling for smoking cessation: Effects of
single-session and multiple-session interventions. Journal of Consulting
and Clinical Psychology, 64, 202–211.
Zimmerman, T. S., Jacobsen, R. B., MacIntyre, M., & Watson, C. (1996).
Solution-focused parenting groups: An empirical study. Journal of Sys-
temic Therapies, 15, 12–25.
Zimmerman, T. S., Prest, L. A., & Wetzel, B. E. (1997). Solution-focused
couples therapy groups: An empirical study. Journal of Family Ther-
apy, 19, 125–144.
Received November 14, 2000
Revision received August 20, 2001
Accepted August 21, 2001 䡲
242 REESE, CONOLEY, AND BROSSART
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Counseling Psychology
This content is subject to copyright. Terms and conditions apply.
