University of Southern California Dissertations and Theses

Assessing therapeutic change mechanisms in motivational interviewing using the articulated thoughts in simulated situations paradigm

(USC Thesis Other)

Assessing therapeutic change mechanisms in motivational interviewing using the articulated thoughts in simulated situations paradigm

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content ASSESSING THERAPEUTIC CHANGE MECHANISMS IN MOTIVATIONAL
INTERVIEWING USING THE ARTICULATED THOUGHTS IN SIMULATED
SITUATIONS PARADIGM
by
Jed P. Grodin
___________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)
December 2006
Copyright 2006 Jed P. Grodin
ii
Acknowledgements
I would like to sincerely thank my committee chair and advisor, Dr. Gerald
Davison, and the members of my committee: Dr. Stanley Huey, Dr. David Walsh,
and Dr. Rodney Goodyear, for their support, guidance, and encouragement during
this project.
I would also like to thank the many USC undergraduates who served as research
assistants for this study. I am truly indebted to them all and thank them for their
excellent work. I would like to specifically recognize the project coordinators,
Daniel Goldman and Lauren Baron, who worked tirelessly to help ensure the study
was conducted professionally and efficiently.
This project was supported in part by funds from the Kellerman Research Fund and
I would like to thank Dr. Jonathan Kellerman for his support.
iii
Table of Contents
Acknowledgements ii
List of Tables iv
Abstract v
Chapter 1: Introduction 1
Chapter 2: Method 34
Chapter 3: Results 47
Figure 1: Plot of significant quadratic within subjects
contrast for therapist style as measured by the Readiness
to Change Questionnaire. 63
Chapter 4: Discussion 75
References 85
Appendices 95
Appendix A: Frequency/Quantity Screening and 95
Follow-Up Measure
Appendix B: Alcohol Timeline Followback Instructions 97
and Example Calendar
Appendix C: Readiness to Change Questionnaire (RTCQ) 100
Appendix D: Alcohol Stages of Change Ruler (ASCR) 101
Appendix E: Frequency/Quantity (F/Q) Intention 102
Appendix F: Overall Impression of Drinking Alcohol (OIDA) 103
Appendix G: Overall Impression of Quitting Alcohol (OIQA) 104
Appendix H: Experimental Check for Subjects (ECS) 105
Appendix I: Demographics Questionnaire 106
Appendix J: Therapeutic Reactance Scale (TRS) 107
Appendix K: Experimental Check-Experts on the scenario 108
Appendix L: Condition One ATSS Script 109
Appendix M: Condition Two ATSS Script 111
Appendix N: Condition Three ATSS Script 113
Appendix O: Condition Four ATSS Script 115
Appendix P: ATSS Pleasant Scenario Transcript 117
Appendix Q: Coding Manual 118
Appendix R: Non-disclosure Agreement for research project 125
iv
List of Tables
Table Page
1 The four experimental conditions in the 2X2 experimental design 2
2 The four experimental conditions in the 2X2 experimental design 34
3 Means and stand deviations for each dependent
variable grouped by condition 47
4 Demographic differences between semester 1,2, and 3 53
5 Demographic differences among the experimental groups 53
6 Pre-ATSS Paper-and-Pencil Measures Inter-item Correlations 56
7 ATSS Dependent Variable Correlations 57
8 Post-ATSS Paper-and-Pencil Measures Inter-item Correlations 58
9 Follow-Up Paper-and-Pencil Measures Inter-item Correlations 59
10 Demographic data for the sample. 60
11 Paired Samples T-tests associated with significant
within subjects quadratic contrast for the
readiness to change questionnaire 64
12 The four experimental conditions in the 2X2 experimental design 66
13 Interaction effect statistics for therapist style X decisional
balance for change resistance cognitions 68
14 Inter-item correlations for the participant experimental
check scale (ECS) 70
15 Inter-item correlations for expert experimental check scale 72
v
Abstract
Current evidence suggests that Motivational Interviewing (MI) can help to
create behavior change in a variety of contexts, especially for those with alcohol
use problems, but evidence is less clear on how and why MI seems to work. Using
the Articulated Thoughts in Simulated Situations paradigm (ATSS), this study
aimed to determine the mechanisms by which MI may modify the cognitions and
behaviors of college-aged binge drinkers. Two mechanisms typical of an MI
intervention were isolated, simulated, and analyzed: the presence of a
warm/Rogerian therapist communication style and the use of a decisional balance
exercise. By using a 2X2 design to compare these mechanisms to commonly used
alternative therapeutic approaches, we sought to determine which mechanisms have
the greatest impact on cognitions related to binge drinking and on intention to
change, intention to drink, overall impressions of drinking, as well as actual
changes in drinking behavior at a 30 day follow-up. As in a previous study, we
found moderately strong evidence supporting the validity of the ATSS scenarios for
our purposes. No significant effects were found on drinking behavior or any other
variables at 30-day follow-up. However, as in the earlier study, the data show that,
as compared to a more directive and confrontational communication style, the
warm/Rogerian therapist style was associated with significantly fewer participant
cognitions reflecting resistance to change during the ATSS scenario (a simulated
therapy session) and with a temporary increase in readiness to change immediately
after the scenario. We also found that, regardless of condition, higher levels of trait
vi
reactance increase the degree to which participants report that the therapist
influenced them to think about not changing their drinking behavior. This study
provides evidence that the warm/Rogerian therapist style advocated by MI is an
important and active ingredient in psychotherapeutic process and, as such, should
be a focus of clinical training and that attention to trait-like variables like reactance
may help improve our understanding of MI and psychotherapy processes and
outcomes in general.
1
Chapter 1: Introduction
Specific Aims
The objective of this study was to identify therapeutic change mechanisms
in a type of therapy called Motivational Interviewing (Miller & Rollnick (1991,
2002)). To do so, the current study utilized the Articulated Thoughts in Simulated
Situations paradigm (ATSS; Davison, Robins, & Johnson, 1983) and paper-and-
pencil measures to assess the cognitions and behavior of participants exposed to
simulations of various therapy strategy and style conditions, including one designed
to reflect the approach used in Motivational Interviewing (MI). This study
attempted to replicate and expand upon findings of a previous study by Grodin,
Davison, & Earleywine, (2004). Like the previous study, the simulations (or
“ATSS scenarios”) in each experimental condition were constructed to reflect the
components of frequently used, but conceptually opposite approaches to treating
the same problem: problematic alcohol consumption in the college student
population (W.R. Miller, personal communication, November 18, 2002).
The first condition was meant to simulate the approach used in MI and
featured a combination of a warm, Rogerian-type “non-specifics” communication
style (Rogers, 1951, 1961) advocated by the developers of MI and the use of a
decisional balance exercise, a common technique used in MI. The second
condition featured the use of only the warm/Rogerian communication style without
the use of the decisional balance exercise in the context of a general discussion
about alcohol use. The third condition featured the use of a
2
confrontational/educational communication style (often used in non-MI alcohol
counseling) coupled with the decisional balance exercise. The fourth condition
featured the use of a confrontational/educational communication style in the
context of a general discussion of alcohol similar to that in condition two
(see table 1).
Table 1. The four experimental conditions in the 2X2 experimental design
Decisional Balance NO Decisional
Balance
Warm/Rogerian Condition One (MI) Condition Two
Confrontational/Educational Condition Three Condition Four
In the previous study, we assessed a number of variables at pre-ATSS,
during the ATSS procedure, and at post-ATSS to examine within and between
group differences for our dependent variables related to readiness to change and to
participants’ ATSS-measured responses to the conditions. For the current study, in
addition to seeking replication on those same dependent variables, we extended our
analyses with the addition of an assessment session one month (30 days) after the
initial meeting. The purpose of this follow-up was to assess any associations
between the conditions and actual drinking behavior as well as to assess whether
any group differences observed at pre-ATSS and post-ATSS were maintained or
changed in some way over time.
We assessed participant responses on three levels: first, we compared the
ATSS-coded cognitions and responses to pre-ATSS, post-ATSS, and follow-up
3
paper-and-pencil measures on the basis of their response to the therapist
communication style (“Warm/Rogerian” vs. “Confrontation/Educational”).
Second, we examined whether the presence or absence of the decisional balance
exercise was associated with differences in the dependent variables. Third, we
assessed whether there is an interaction effect between therapist communication
style and the presence or absence of the decisional balance exercise on these
variables.
By analyzing both a specific technique used in MI (the decisional balance)
as well as by making a broader comparison of therapist communication styles, we
aimed to illuminate particular cognitive and behavioral changes taking place in
participants. Taking the cognitive and cognitive-behavioral models described by
Beck (Beck, A. 1979; Beck, J. 1995) and Ellis (Ellis & Dryden, 1997) as a
theoretical basis and given the variety of research supporting the validity of the
cognitive-behavioral model (and specifically the support garnered for the
contention that cognitions can influence behavior), we expected that a detailed
examination of the cognitions and behaviors of those engaged in the MI process
would yield insight into the mechanisms that might account for the positive
outcome data for MI in the treatment of a wide variety of behavioral problems (see
Burke, Arkowitz, & Dunn (2002) and Miller (2002) for reviews). While such
research on MI outcomes is plentiful, the research into the mechanisms of MI has
just begun.
4
Hypotheses
The hypotheses of this study were as follows:
Main Effects Hypotheses:
1. There will be a main effect for decisional balance such that, as compared to
those not exposed to the decisional balance, participants exposed to the
decisional balance will:
a. Articulate a greater number of “change cognitions” and more
“change resistance cognitions” as measured by the ATSS coding
format.
b. Indicate a higher stage of change, i.e., readiness or intention to drink
less immediately after exposure to the ATSS stimulus and at follow-
up, as measured by the Readiness to Change Questionnaire (RTCQ;
Rollnick, Heather, Gold, Hall, 1992) and the Alcohol Stages of
Change Ruler (ASCR; CASAA, 1995)
c. Indicate that they intend to consume less alcohol as measured by the
Frequency/Quantity Intention to Drink measure (F/Q Intention) at
post-ATSS and at follow-up.
d. Indicate a less favorable attitude towards drinking as measured by
the Overall Pros and Cons of Drinking measure (OIDA) and indicate
a more favorable attitude towards quitting drinking as measured by
the Overall Pros and Cons of Quitting Alcohol measure (OIQA) at
post-ATSS and at follow-up.
5
e. Indicate a decrease in binge drinking episodes, days of drinking,
total number of drinks, maximum number of drinks on any one
occasion, and average number of drinks per drinking occasion at
follow-up.
2. There will be a main effect for therapist communication style such that, as
compared to those who are exposed to the confrontational/educational
communication style, participants exposed to the “warm-Rogerian”
condition will:
a. Articulate fewer “change resistance cognitions”, fewer “therapy
resistance cognitions”, more “change cognitions”, and more
“therapy cooperation cognitions” as measured by the ATSS coding
format.
b. Indicate a higher stage of change, i.e., readiness or intention to drink
less, as measured by the RTCQ and the ASCR at post-ATSS and at
follow-up.
c. Indicate that they intend to consume less alcohol as measured by the
F/Q Intention at post-ATSS and at follow-up
d. Indicate a less favorable attitude towards drinking as measured by
the OIDA and indicate a more favorable attitude towards quitting
drinking as measured by the OIQA at post-ATSS and at follow-up.
e. Indicate a decrease in binge drinking episodes, days of drinking,
total number of drinks, maximum number of drinks on any one
6
occasion, and average number of drinks per drinking occasion at
follow-up.
Interaction Effects Hypothesis
1. There will be an interaction effect of receiving both the decisional balance and
the “Warm-Rogerian” therapist communication style (illustrated by condition
one, the scenario designed to simulate MI) such that, as compared to those in
the other three conditions, participants in this condition will:
a. Articulate fewer “change resistance cognitions”, fewer “therapy
resistance cognitions”, more “change cognitions”, and more “therapy
cooperation cognitions” as measured by the ATSS coding format.
b. Indicate a higher stage of change, i.e., readiness or intention to drink
less, as measured by the RTCQ and the ASCR at post-ATSS and at
follow-up.
c. Indicate that they intend to consume less alcohol as measured by the
F/Q Intention at post-ATSS and at follow-up.
d. Indicate a less favorable attitude towards drinking as measured by the
OIDA and indicate a more favorable attitude towards quitting drinking
as measured by the OIQA at post-ATSS and at follow-up.
e. Indicate a decrease in binge drinking episodes, days of drinking, total
number of drinks, maximum number of drinks on any one occasion, and
average number of drinks per drinking occasion at follow-up.
7
2. Additionally, participants in the condition receiving the
confrontational/educational therapist communication without the decisional
balance (condition four) will, as compared to the other three conditions:
a. Articulate fewer “change cognitions”, fewer “therapy cooperation
cognitions”, more “change resistance cognitions”, and more “therapy
resistance cognitions” as measured by the ATSS coding format.
b. Indicate a lower stage of change, i.e., readiness or intention to drink
less, as measured by the RTCQ and the ASCR at post-ATSS and at
follow-up.
c. Indicate that they intend to consume more alcohol as measured by the
F/Q Intention at post-ATSS and at follow-up.
d. Indicate a more favorable attitude towards drinking as measured by the
OIDA and indicate a less favorable attitude towards quitting drinking as
measured by the OIQA at post-ATSS and at follow-up.
e. Indicate no decrease in binge drinking episodes, days of drinking, total
number of drinks, maximum number of drinks, and average number of
drinks per drinking occasion at follow-up.
BACKGROUND
Before addressing background information about MI and the related
research, we begin with a brief discussion of the study that we attempted to
replicate and extend. That study (Grodin et al., 2004) also used the ATSS
paradigm and paper-and-pencil measures to assess the cognitions of binge drinking
8
college students exposed to simulated therapy interactions in a 2X2 factorial
design. We found support for several of our hypotheses and, in so doing, an
experimental foundation for the current project.
To summarize our findings, the warm/Rogerian therapist style was
associated with significantly more therapy cooperation cognitions and significantly
fewer therapy resistance cognitions than the confrontational/educational therapist
style. The presence of the decisional balance exercise was associated with
significantly more change cognitions and resistance to change cognitions than the
absence of the decisional balance. Also, therapist style and the decisional balance
interacted such that the confrontational/educational therapist style without the
decisional balance condition (condition four) was associated with a lower intention
to change as measured by the Alcohol Stage of Change ruler (ASCR); and this
interaction was partially mediated by change cognitions.
We drew several conclusions based on these findings. The association
between the warm/Rogerian therapist style and therapy cooperation cognitions may
help to explain why those exposed to MI interventions seem to participate and
engage more in the therapeutic process as compared to a more confrontational
counselor style (see the MI outcome section below for a discussion). Our evidence
for the partial mediation effect of change cognitions in the interaction effect may
help to explain why an approach like MI that makes use of an evocative technique
like the decisional balance can be associated with an increase in intention to
change. To the extent that MI or other interventions can help clients generate their
9
own thoughts about change, these interventions may help to increase client
motivation to change.
We also conducted several checks on our methodology and, on the basis of
that data we had empirical support for the soundness of the experimental design for
the current project. We conducted checks on the coding reliability for our ATSS
data (we found inter-rater reliability, as measured by an intra-class correlation
coefficient, to exceed .9 for all ATSS coding variables); on the extent to which
participants found the scenarios to be realistic and engaging (we found a
moderately strong level of engagement (mean/std dev of 8.4/1.97 out of a possible
12 for all participants); and on whether the participants perceived the two different
therapist style conditions as we intended (the warm/Rogerian therapist style was in
fact perceived by participants as more warm, empathic, and genuine than the
confrontational/educational therapist style with an effect size, d=1.1, sig<.001).
Again, based on these data, we were reasonably confident that the ATSS model
proposed for this study is a relevant, engaging, controllable, and valid method to
assess cognitions of college drinkers engaged in therapeutic interactions.
This previous study was one of a very few in the literature attempting to
identify and isolate the therapeutic mechanisms of action for MI. The current
study, with the addition of a follow-up assessment session, allows us the chance to
not only replicate the findings described above, but also to assess whether
differences in the conditions maintain over time and are related to self-reported
changes in actual drinking behavior.
10
Motivational Interviewing: Background Information
In a general sense, MI is a “directive, client-centered style of counseling
that helps clients to explore and resolve their ambivalence about changing”
(Rollnick & Miller, 1995). MI was initially developed as a way to prepare people
with substance abuse problems (specifically problem drinkers (Miller, 1983)) to
change their behavior and it was intended for use in conjunction with subsequent
substance abuse treatment. MI views motivation as a dynamic, state-like variable
necessary to the process of change and MI was designed to help the client build and
maintain the motivation to change. Because many clients enter treatment feeling
ambivalent about changing their behavior, highly directive advice-giving can often
backfire and result in psychological reactance (Brehm & Brehm, 1981) and a state
of decreased motivation to change (Rollnick, Heather, & Bell, 1992). As Rollnick,
Heather, and Bell (1992) note, in MI it is the therapist’s role to help the client build
motivation by helping to him/her to explore his/her own ambivalence, express
concerns about the behavior, and articulate his/her own reasons for change.
Empirical research, as we will see below, has supported the efficacy of MI
in a wide variety of contexts and for a variety of clients. It has quickly grown in
popularity since it was originally developed and is now used as both a stand-alone
therapy as well as in conjunction with other techniques with numerous populations
for a growing variety of problems (Miller & Rollnick, 2002).
While research has shown that MI can be effective, there has been little
work done to examine how and why it seems to work. Indeed, Miller and Rollnick
11
have stated that, “There is reasonable evidence that MI works in certain
applications, but the data thus far are less clear in documenting how and why it
works” (Miller & Rollnick, 2002, p.26). This project attempted to address this
topic, i.e., it seeks to determine how MI may change a participant’s cognitions
about a target behavior, the cognitions relating to changing that behavior, and the
relationship between these cognitions and the actual target behavior. More
specifically, this project examined MI in a context with which it is commonly
associated: problematic drinking behavior in a college population. The decision to
use this target behavior was bolstered by the prevalence and reported negative
consequences of heavy drinking reported by college students (Wechsler,
Davenport, Dowdall, Moeykens, & Castillo, 1994) and by the findings (described
below) that support the efficacy of MI for problematic alcohol use. In essence, we
know that the problem (binge drinking in the college population) exists and we
know that MI appears to help alleviate the problem, and thus we are provided with
an ideal context in which to analyze how and why it works.
In the following review of relevant literature, we will begin with an
overview of some of the basic concepts of MI before describing the relevant
findings in the MI research. As well, we will address the topics of drinking
behavior in college students, research on the decisional balance, and research on
ATSS.
12
Motivational Interviewing: A Definition
We begin our explication of MI with a description of the most general philosophical
principles of the method. These principles guide what Miller and Rollnick call, the “spirit of
motivational interviewing” (Rollnick & Miller, 1995; Miller & Rollnick, 2002). We will then
move on to describe the four resultant clinical principles before moving on to a brief
discussion of specific therapy techniques.
The Spirit of Motivational Interviewing
MI grows out the interrelated principles of collaboration, evocation, and
autonomy (Miller & Rollnick, 2002). MI is collaborative in that it requires the
therapist to foster a partner-like relationship with the client as opposed to an
authoritarian or expert-novice relationship. The interpersonal atmosphere in MI is
one of support and exploration and not confrontation and argumentation.
Additionally, resistance and client interest in change are seen not as client traits, but
as a product of the interaction between therapist and client. The research on the
impact of therapist communication style (Miller, Benefield, & Tonigan, 1993;
Patterson & Forgatch, 1985) appears to support the claim that a more
confrontational approach results in greater client resistance.
MI also emphasizes the role of evocation. Motivation to change is not
imposed from the outside through persuasion, confrontation, or threat, but rather is
evoked from the client (Rollnick & Miller, 1995). This approach assumes that the
resources for the motivation to change are intrinsic and that it is the therapist’s job
to draw on the client’s thoughts, values, and feelings to help bring out this
13
motivation from within the client. This approach stands in opposition to one in
which a counselor seeks to provide information and insight that the therapist
believes the client does not yet possess.
Finally, the client is understood to be autonomous in that he/she has the
ability and the responsibility to articulate and resolve his/her ambivalence about
change (Rollnick & Miller, 1995). MI respects the client as an autonomous
individual who is free to take action in any direction he/she may choose. Such a
view of the client as autonomous goes hand in hand with MI’s emphasis on
building intrinsic motivation for change.
Clinical Principles of Motivational Interviewing
These philosophical principles underlying the spirit of MI give rise to four
more specific clinical principles of MI. These principles (express empathy,
develop discrepancy, roll with resistance, and support self-efficacy) serve as the
general guiding strategies for MI and give rise to specific techniques used by
therapists.
Expressing Empathy
“Expressing Empathy” is typified by an understanding of the client’s
ambivalence as a natural state of being and by a stance of acceptance of, but not
necessarily agreement with, the client’s perspective. Accurate empathy, as
expressed most often with non-judgmental and skillful reflective listening, is
fundamental to MI’s validation of the client’s autonomy and the collaborative and
evocative spirit of the process. The atmosphere of acceptance cultivated when the
14
therapist expresses empathy facilitates the change process as conceived by Miller
and Rollnick (2002). This client-centered approach, drawn originally from the
work of Carl Rogers (Rogers, 1951, 1961), is the root of MI’s emphasis on non-
directive, empathic interactive skills.
Developing Discrepancy
The emphasis on evocation and client autonomy manifests itself in the
second clinical principle of MI, “Developing Discrepancy”. It is the therapist’s job
to help draw out and illuminate for the client a perceived discrepancy between
present behavior and important and personal goals, values, or attitudes. Such a
discrepancy serves to motivate the client to change. For example, taking a cue
from Rokeach’s work in human values (Rokeach (1973), Miller and Rollnick assert
that when a behavior conflicts with a deeply held value, it is usually the behavior
that changes (Miller & Rollnick, 2002). While a MI therapist maintains a client-
centered approach, she seeks ways to identify and amplify discrepancies in order to
override the inertia of the status quo and therefore provide motivation from the
client’s perspective (Miller & Rollnick, 2002). In this way, MI is more directive
than classic, Rogerian client-centered approaches; there is an agenda to help the
client to change his/her behavior in a particular, healthier direction. While
ultimately it is the client who must present the arguments for change, the therapist
uses techniques like the decisional balance exercise and skillful reflections to
develop the discrepancy that gives rise to these arguments for change (“change
talk”). “The key,” according to Miller, “lies in the process of making the conflict
15
conscious, more salient, bringing the client face-to-face with the discrepancy in an
atmosphere that makes it safe to confront and feel the reality” (Miller, 1994, p.118).
Miller originally conceived of this principle as a tool to create “cognitive
dissonance” as conceived of by Festinger (Miller, 1983), but, partly in order to
avoid invoking an inherent drive towards cognitive consistency, this discrepancy is
now thought of more broadly as discrepancy between how things are and how one
wants them to be (Miller & Rollnick, 2002). It is interesting to note that Draycott
and Dabbs (1998) re-examine Miller’s move away from cognitive dissonance and
seek to demonstrate that cognitive dissonance is clearly helpful in understanding
the mechanism of change of MI. However, the question as to whether MI is
invoking cognitive dissonance and what the implications would be in either case is
at this time unresolved.
Rolling with Resistance
Given the philosophical and functional importance of evocation,
collaboration, and client autonomy as well as the evidence provided by the
previously cited research on therapist-client interactions, it follows that
argumentation with the client or the forceful imposition of an overt change agenda
are to be avoided. “Rolling with Resistance”, the third clinical principle of MI,
reminds the therapist to refrain from opposing client resistance and to see such
resistance as a product of the interaction between therapist and client and not a sign
of some innate stubbornness or inflexibility on the part of the client. Resistance
becomes a cue to the therapist to observe and to change his/her own behavior.
16
Moreover, client resistance can be seen as an opportunity for further skillful
reflection and open-ended questioning to subtly redirect the client towards change.
Supporting self-efficacy
“Supporting self-efficacy” plays an important role in the execution of
behavior change once the decision to do so has been made. Miller and Rollnick
(2002) describe three conditions necessary for change: a person has to be “ready”
(the degree to which a person places a relatively high priority on changing),
“willing” (the degree to which a person desires change or sees it as important), and
“able” (the degree to which a person is confident that she can make change). Self-
efficacy (Bandura, 1977, 1997) plays a role particularly in the degree to which a
client feels “able” to make a change. A client may see change as an important
priority, but without a sense of self-efficacy, i.e., the belief that she can actually
make the change, the client may not make the change. By, for example, asking
evocative questions, empathizing accurately, reframing, and affirming, the therapist
can support the client’s sense of self-efficacy, strengthen the client’s confidence in
his/her ability to change, and in so doing help the client build and maintain the
momentum necessary to make and maintain behavior change.
A Specific Technique: The Decisional Balance
In many senses, MI is as much a “way of being” with the client as much as
a collection of highly specific tools (Miller & Rollnick, 2002). At the same time, in
addition to the specific techniques already mentioned (open-ended questioning,
reflective listening, affirming, and reframing), the decisional balance exercise
17
stands as an example of a concrete and specific technique that exemplifies the MI
process and one that is a focus of the current study.
MI and this study in particular use concepts of decisional balance as
described initially by Janis and Mann (Janis & Mann, 1977). In the Janis and Mann
model, decision-making is a process of weighing the pros and cons of a particular
behavior. Once one “side” outweighs the other, behavior likely changes in that
direction. Janis and Mann’s model meshes particularly well with MI because both
understand ambivalence to be natural. A person engaging in heavy drinking
behavior, for example, might both suffer and benefit from that behavior. Whether
the person continues to drink or not depends in large part upon the relative balance
of the costs and benefits of the drinking behavior.
This decisional balance process can be operationalized in a decisional
balance worksheet: a written list of the pros and cons of a target behavior. The list
can also be arrived at through an open-ended process of conversation and
exploration by therapist and client. In this case, the therapist elicits from the client
the pros and cons of changing or not changing a behavior in a supportive, non-
judgmental, empathic manner. In whichever manner a decisional balance is
constructed, it can serve as an assessment picture of the client’s cognitions relating
to the target behavior as well as an intervention on its own, i.e., it may motivate the
client to make a behavior change (Miller & Rollnick, 1991; LaBrie, Pedersen,
Earleywine, & Olsen, 2006).
18
The validity of the decisional balance construct has been supported in
research on a variety of behaviors. Migneault, Pallonen, Velicer (1997) and
Migneault, Velicer, Prochaska et al. (1999) found that a decisional balance measure
for immoderate drinking in college students had satisfactory external validity as
indicated by correlation with drinking outcome measures and stages of change
measures. Rossi, Greene, Rossi et al. (2001) found support for the validity of a
decisional balance measure in a study on adolescent dietary fat reduction, and
Plummer, Velicer, Redding and colleagues (2001) validated a decisional balance
measure for adolescent smoking.
In a number of studies (Migneault et al., (1997); Migneault et al. (1999);
Prochaska, Velicer, Rossi et al., 1994; Rossi et al. (2001); Share et al. (2004)),
researchers have found that, for a variety of populations and behaviors, a
participants’ decisional balance “profile” (each individual’s balance between pros
and cons for a behavior) correlate with particular “stages of change” (e.g.,
precontemplation, contemplation, preparation, action, maintenance) as described by
the transtheoretical model of change (Prochaska and DiClemente, 1986). For
example, for a person in the precontemplation stage, the cons of changing a
behavior would outweigh the pros whereas for a person in the action stage, the pros
of changing a behavior would outweigh the cons. This indicates that decisional
balance exercises can help a therapist assess a client’s stage of change.
Furthermore, there is empirical support for the contention that the decisional
balance exercise can be seen as a mechanism by which clients can progress from
19
one stage of the model to the next (LaBrie et al., 2006). The Transtheoretical
Model of Change provides a helpful context in which to examine the process of
change and the role played by decisional balance and MI in general.
The Context for Change: The Transtheoretical Model of Change
Prochaska and DiClemente’s Transtheoretical Model of Change (TTM;
Prochaska & DiClemente, 1986) is a theory of the change process that is closely
allied with MI. It can be seen as the change process “context” in which MI is
theorized to operate. The model states that people move through a series of stages
(precontemplation, contemplation, preparation, action, maintenance), each of which
indicates a state of motivational “readiness” or intention to change (Prochaska &
DiClemente, 1986). The TTM has found support in empirical research on a wide
variety of behaviors, such as alcohol, smoking cessation, exercise, dietary change,
and more (DiClemente & Hughes, 1990; Prochaska et al., 1994). In the model,
people move from a stage of precontemplation (not considering change), to
contemplation (considering and possibly intending to change), to preparation
(intending and planning to change), to action (making overt change), and to
maintenance (consolidating and maintaining change). People can move from one
stage to the next as well as relapse and begin again at any one of the stages.
Prochaska et al. (1994) found that the stages of change construct was valid and
consistent for people across twelve different problem behaviors (smoking
cessation, quitting cocaine, weight control, high-fat diets, adolescent delinquent
behaviors, safer sex, condom use, sunscreen use, radon gas exposure, exercise
20
acquisition, mammography screening, and physicians’ preventive practices with
smokers) and, again, that each stage of change was predictably related to the
patterns of pros and cons for the participants.
In addition to underscoring the importance of motivation in the change
process (it is a primary mechanism by which a person advances to the next stage of
change), the TTM also provides a framework for the MI therapist to determine
which styles and techniques are most appropriate for the client at a particular stage.
For example, the decisional balance may be particularly useful in the contemplation
stage whereas affirmation work focusing on self-efficacy might become a priority
after a relapse. Furthermore, given that change can be difficult, that relapse does
happen, and that relapse can have a great impact on a client’s thoughts and feelings
about attempting to change a target behavior again, MI and its use of empathy and
its acceptance of ambivalence, is particularly well suited to helping clients
throughout the TTM’s stages of change.
Motivational Interviewing Outcome research
To date there are well over 50 studies on the outcome of MI interventions
and two meta-analyses on the topic have been published since 2001. Dunn,
DeRoo, and Rivara (2001) reviewed 29 randomized clinical trials of therapies
purported to use a MI approach. They found the greatest effect sizes in the studies
of MI with substance abuse and overall found that 60% of the studies “yielded at
least one significant behavior change effect size.” (Dunn et al., 2001). Although
Rollnick and Miller both raise the criticism that Dunn and colleagues had inclusion
21
criteria that were too broad and thus included a number of studies that used
interventions that should not be considered proper “Motivational Interviewing”, the
meta-analysis provides at least some evidence that MI is efficacious (Rollnick,
2001; Miller, 2001).
Burke, Arkowitz, and Dunn (2002) conducted a meta-analysis of 26 studies
fitting the authors’ definition of a controlled clinical trial. The studies under review
all made use of therapy techniques that relied upon MI principles and some specific
techniques, but the therapy techniques varied in many respects, such as duration of
treatment, whether the MI-related treatment was used in concert with another
treatment, etc. In order to distinguish these therapies from “pure” MI (to the extent
that such a “pure” form was defined by Miller and Rollnick (1991, 2002)), Burke et
al. (2002) referred to them as “Adaptations of Motivational Interviewing” (AMIs).
Burke and colleagues found that the greatest support for AMIs lies in the
domain of alcohol problems. In this domain, AMIs appear to be efficacious across
a number of settings (hospitals, outpatient clinics, general medical practice, etc.)
and even in as few as one or two sessions. While the authors mention that
treatment adherence and fidelity were difficult to ascertain for the studies under
analysis, Burke et al. note that these studies were in general methodologically very
sound.
Significant effects were found also in the domains of drug addiction,
cigarette smoking, psychiatric outpatient treatment attendance, hypertension,
diabetes, and eating disorders. Across all domains except for HIV risk behaviors
22
and for a number of populations including college students, AMIs seem in many
cases to work in a relatively small number of sessions, e.g., six sessions or less. In
many cases, AMIs have resulted in long lasting behavior change (as long as four
years in Marlatt et al. (1998)). Moreover, AMIs were found to be effective as
pretreatment preludes, in comparison with no treatment controls, and were found to
be at least as effective as alternative and more extensive treatments (Project
MATCH, 1997).
Miller has compiled a PowerPoint presentation on controlled outcome
studies of MI for the website, www.motivationalinterview.org (Miller, 2002). The
vast majority of the studies across a range of behaviors such as alcohol
consumption, gambling, HIV risk, etc. show MI to be superior to a wide variety of
comparison conditions.
More recent outcome studies for MI have shown mixed results. In a study
by Miller, Yahne, and Tonigan (2003), MI did not have any effect on drug use
when added to either inpatient or outpatient treatment. Mullins et al. (2004)
showed no difference between a three session MI intervention and an educational
videotape on treatment retention for a population of women involved in the child
welfare system who were being treated for substance abuse. Murphy et al. (2004)
concluded that the addition of MI techniques did not enhance the effect of a one
session personalized drinking feedback intervention. While the first two studies are
not in the area of alcohol treatment (where most of the MI efficacy data have been
established) and while the authors of these studies offer various explanations for
23
possible confounding factors, these findings show that MI is obviously not
universally applicable and effective. However, Tait and Hulse (2003) and
McCambridge and Strang (2004) showed significant effects for MI-type
interventions for the treatment of a variety of substance abuse. Particularly
important for our study, both of these studies show that these effects are occurring
in young adult and adolescent populations in as little as one session. Taken as a
whole, the literature shows that researchers’ and clinicians’ enthusiasm for MI
should be somewhat moderated at this time, but is nonetheless reasonable and that
research into the mechanisms by which MI works is certainly warranted.
Motivational Interviewing Mechanism Research
Empirical investigations of the change mechanisms in MI have only just
begun. One of the first examples was Sanchez’s work on using value-card sorts as
a way of developing and measuring discrepancy (Sanchez, 2000). Sanchez built on
Miller’s hypothesis that a discrepancy between behavior and a deeply held value is
a primary mechanism of MI (Miller, 1994). Sanchez compared two groups; one
that received alcohol counseling and went through the values card sort designed by
Miller (Miller, n.d.) and one that received only the alcohol counseling. Sanchez
found significantly greater improvement in drinking measures for the values card
sort group compared with the control group. These results indicate that an exercise
like the values card sort that highlights discrepancies between important personal
values and behavior may work as a mechanism of behavior change.
24
Two more recent studies have looked at the use of particular language by
the client taking part in a MI intervention as a predictor of treatment outcome.
While these studies are not examples of “mechanism” research, they are related in
the sense that, to the degree that particular aspects of MI are found to be associated
with the client language that predicts positive outcomes, these studies indicate the
aspects of MI that might generate behavior change. Strang and McCambridge
(2004) found that clients’ self-motivational statements, i.e., the amount of “change
talk” articulated by the client in the session, was predictive of outcomes at three
month follow-up. These findings should be interpreted with caution, however,
because the process of data collection was limited (amount of client change talk
was summarized by the therapist after the session and was not coded by an outside
observer).
Amrhein, Miller et al. (2003) conducted a much more in-depth study of the
relationship between client language in therapy sessions and drug use outcome. In
this case, the researchers used a highly structured observer coding system to code
both the frequency of client utterances about changing the target behavior (drug
use) as well as the strength or intensity of each utterance, e.g., a statement like, “I
am definitely going to change” would be coded by category (in this case,
“commitment” talk) as well as with a numerical code for strength (in this case,
perhaps a 5 out of 5, indicating that it is a very strong statement relative to others).
They found that the strength of client statements reflecting a commitment to change
was predictive of behavioral outcomes at follow-up. For our purposes, we can
25
conclude from this that those aspects of MI associated with generating strong client
commitment language could be a mechanism of action for MI.
More recently, Moyers, Miller, and Hendrickson (2005) examined and
coded audiotapes of MI sessions to try to identify therapist behaviors associated
with client engagement with therapy. They found that the degree to which
therapists exhibited interpersonal skills considered to be important to MI (therapist
warmth, therapist empathy, therapist egalitarianism/respect of client autonomy,
therapist acceptance, and therapist display of the “spirit of MI”) was positively
associated with client engagement with therapy, as measured by client cooperation,
client disclosure, and client expression of affect. Interestingly, they also found that
therapist behaviors considered inconsistent with MI (confrontation, warning, and
direction-giving) were also associated with increases in client engagement when
such therapist behaviors were done in the context of an otherwise “skillful” session.
The authors suggest that this unexpected finding might be explained by the notion
that therapist confrontation, warning, and direction are consistent with the quality
of genuineness (also important to MI but not measured by the study) and that
clients value this quality and respond to it well in any form when the therapist is
also warm, accepting, egalitarian, empathic, and acting with the “spirit of MI”. As
regards the mechanisms of change for MI, the authors assert that these therapist
behaviors and the subsequent client engagement will build the working alliance.
Citing the literature on working alliance, the authors argue that improvement on
working alliance will be associated with better client outcomes. Thus, therapist
26
interpersonal skills are at the beginning of a chain leading to client change, though
the degree to which certain or all of these skills are necessary or sufficient has not
been precisely determined.
Very relevant to the current study, LaBrie et al. (2006) found that the
decisional balance exercise on its own was associated with increased motivation to
decrease drinking and decreases in the number of drinks that they intended to drink
as well as in the actual number of days that they drank, the maximum number of
drinks consumed on one occasion, and the average number of drinks they
consumed per occasion at follow-up. Though the authors stop short of theorizing
or researching about how and why the decisional balance has its effects, their
findings suggest that the decisional balance technique used in MI may explain a
good part of the changes in client motivation and behavior.
General Therapist Communication Style and Strategy Research
On a more general level, a number of empirical studies have explored the
impact of different therapist communication styles or strategies on client behavior
and treatment outcome. The early findings described in Truax and Carkhuff (1967)
highlight the importance of therapist empathy as a factor in the success of the
therapeutic process. Luborksy, McLellan, Woody, O’Brien, and Auerbach (1985)
found further empirical support for the assertion that therapist empathy can impact
treatment outcome in their comparison of therapy factors in a population of non-
psychotic veterans. In a study of therapies for problem drinkers, Miller, Taylor,
and West (1980) found that the degree of therapist empathy accounted for two-
27
thirds of the variance in drinking outcome at a 6 month follow-up. More recently,
Burns and Nolen-Hoeksema (1992) found that therapist empathy accounted for a
significant portion of the effect of CBT for depression and Greenberg and
colleagues (2001) conducted a meta-analysis on 47 studies and found a consistent
medium effect size for therapist empathy in a variety of therapies for a variety of
clients.
Building on the work of Chamberlain, Patterson, Reid, Kavanagh, and
Forgatch (1984) on client resistance, Patterson and Forgatch (1985) studied
therapist behavior as a determinant of client noncompliance and found that client
resistance behavior increased more in sessions when a therapist engaged in a
direct/confront style of therapy and decreased when the therapist engaged in a
support/reflect style of interaction. These findings suggest that therapist
communication style can have at least some impact on client behavior in session.
Miller, Benefield, and Tonigan (1993) extended these findings with their study of
therapist style in the context of a MI-type intervention called the Drinker’s Check-
Up (DCU). The comparison groups differed only in the style with which therapists
conducted the second session of the intervention. The authors found that an
empathic, non-confrontational approach yielded less client resistance in session and
showed that such an approach predicted better outcomes at one year than did a
directive-confrontational approach. These findings mesh well with and provide
empirical support for the principles that guide the behavior of MI therapists,
although the work of Moyers et al. (2005) suggests that defining and describing
28
MI-consistent and MI-inconsistent therapist behaviors may be more complicated
than researchers originally thought.
College drinking research
Numerous studies have detailed the prevalence of problematic drinking on
college campuses. Such drinking not only has a direct impact on the drinker’s
physical health, but also increases the likelihood of negative drinking-related
consequences for those involved (Wechsler, Davenport, Dowdall, Moeykens, &
Castillo, 1994; Wechsler, Lee et al., 2000).
In a national survey of American college students, Johnston, O’Malley, &
Bachman (2001) found that 83% of full-time college students reported using
alcohol in the past year, 67% reported alcohol use at some time in the past 30 days,
40.9% reported drinking 5 or more drinks in one session (a binge episode) at least
once in the past 2 weeks, and 4.7% reported drinking on a daily basis in the past 30
days. Johnston et al. (2001) found that college students have a higher rate of heavy
drinking or binge drinking (40.9% vs. 36.9%), but a lower rate of daily drinking
(4.7%) when compared to age-group peers who do not attend college (5.2%).
There is a large body of research elucidating the ties between heavy
drinking and negative physical and social consequences for college students. In a
survey of 17,096 students at 140 American 4-year colleges, Wechsler et al. (1994)
found that frequent binge drinkers were 7 to 20 times more likely to engage in
unplanned or unprotected sex, to have problems with campus police, or to get
physically injured. The CORE Institute surveyed 55,026 American college
29
students on alcohol and other drug use on campuses for the year 2000 (CORE
Institute, 2000) and also found a range of negative consequences for binge
drinking. For example, 62.6% of the students reported having a hangover, 23.5%
reported performing poorly on a test or other project, 13.7% reported getting into
trouble with police or other authorities, 30.8% reported getting into a fight or
argument, 30.3% reported being criticized by someone they knew, 32.5% reported
driving a car while under the influence, 31.7% reported having a memory loss,
38.2% reported doing something they later regretted, 11.7% reported being taken
advantage of sexually, 4.6% reported taking advantage of another sexually. In
addition, the CORE statistics cite secondary effects of drinking, i.e., negative
effects of someone else’s drinking on a student. For example, 28.1% reported
interrupted studying, 16% reported being made to feel unsafe, and 10.9% reported
group activities being adversely affected. Drinking, and occasions of heavy
drinking in particular, can have a negative impact on the drinker’s health, academic
performance, and social environment as well as on those who surround the drinker.
As for assessing the degree to which individuals in this population are
aware of their problems and are ready to change, the picture is complex. While Vik
et al. (2000) found that two-thirds of their heavy-drinking college sample did not
recognize a need to change despite evidence that they were experiencing negative
consequences of drinking and were showing signs of developing alcohol tolerance,
the CORE institute showed that 5.8% of heavy drinkers reported that they had tried
unsuccessfully to stop using alcohol (CORE Institute, 2000). In his sample of 179
30
college drinkers, Caldwell (2002) found that there are varying levels of drinking
and readiness to change drinking and that even drinkers below traditional binge
levels experience negative consequences of drinking. Given the diversity of ways
in which college students consume and think about their consumption of alcohol,
Caldwell argues that they should not be characterized in a dichotomous (binge/not
binge) way. Taken along with findings by Masterman and Kelly (2003), which
found that adolescent drinkers are quite diverse and that some interventions work
for some at-risk drinkers but not for others, these findings suggest that prevention
and intervention efforts should not be the same for all young drinkers, but rather
should be tailored to the individual’s particular drinking and risk profiles. Such a
flexibility and idiographic orientation is definitional for MI. To that end, a number
of interventions that make use of MI have been shown to be effective in reducing
alcohol consumption in the college population (Borsari & Carey, 2000; Marlatt et
al., 1998, Baer, Kivlahan, Blume et al., 2001). The prevalence, consequences, and
diversity of heavy drinking in the college population as well as the efficacy and
flexibility of MI in that context provide evidence that continued study of MI will be
an appropriate and potentially fruitful arena in which to explore the variables
associated with the change process.
The Assessment of Cognitions: ATSS
Our primary cognitive assessment tool was the Articulated Thoughts in
Simulated Situations paradigm as developed by Davison, Robins, & Johnson
(ATSS; 1983). ATSS presents participants with “multi-segmented, imaginal
31
vignettes or scenarios via audiotape. During the course of an imagined situation,
individuals think out loud at specified points interspersed between brief segments
of the story.” (Davison & Best, 2004) ATSS stresses that the participant immerse
him or herself in the scenario, i.e., the participant is asked to articulate thoughts as
if he/she were really in the situation and to avoid statements like “If this happened
to me, I would”, etc. Because it allows for the participant to respond with any and
all thoughts that have occurred to her, ATSS seems to have particularly high
potential when “little is known of the cognitive terrain of interest.” (Davison,
Vogel, & Coffman, 1997, p.955) ATSS allowed us to construct specific scenarios
that are relevant to the participants and, via a coding scheme, provided access to
our participants’ cognitions in response to the conditions. In this way, ATSS has
the potential to provide insight into cognitive processes as they happen.
ATSS has been used in studies of a wide variety of topics, including marital
conflict (Eckhardt, Barbour, & Davison (1998), hypertension and “Type A”
behavior patterns (Weinstein, Davison, De Quattro, & Allen, 1986), depression
(White, Davison, Haaga, & White, 1992), smoking relapse (Haaga, 1987), and
thoughts about hate crimes (Rayburn & Davison, (2002), amongst others. Studies
on psychometric properties are relatively few. This is partly due to the fact that
ATSS is conceived as a paradigm, flexible and adaptable to each experimental
purpose. As such, as opposed to a paper-and-pencil measure, the ATSS scenario
and coding scheme are often unique for each study in which it is used. However,
there is some research indicating that ATSS does indeed do what it purports to do.
32
Davison, Vogel, & Coffman (1997) surveyed the ATSS literature at the
time and found that there is evidence for its face, concurrent, predictive, and
construct validity. Davison, Feldman, & Osborn (1984) found good concurrent
validity between participants’ scores on the Fear of Negative Evaluation Scale
(Watson & Friend, 1969), a questionnaire to measure irrational beliefs (Jones,
1968), and anxiety statements in the ATSS data. Davison and Zighelboim (1987)
conducted an additional study with actual patients that replicated this finding.
With an analog assessment tool like ATSS, psychometric issues are often
brought up in response to the coding of data rather than data themselves (Heyman
& Slep, 2003). In the case of ATSS, validity and reliability, especially inter-rater
reliability, are contingent on the coding scheme selected and the training of coders.
To the degree that the coding scheme has relevance to the topic being discussed and
inter-rater reliability is acceptable, ATSS has the potential to yield valid and
reliable data. Importantly, the results from our previous study show that ATSS can
be used in the context of this study in a reliable and valid way.
Significance of the proposed study
This study assessed the cognitions and behavior of persons exposed to
differing therapy strategies and styles of therapeutic communication. In so doing,
we sought to identify and distinguish between therapeutic processes that encourage
motivation to change a particular behavior and those that do not. We aimed to do so
through an analysis of the effects of therapist style (warm/Rogerian vs.
confrontational/educational), the effects of the decisional balance exercise (present
33
vs. absent), and the effects of the interaction of therapist style and decisional
balance on client cognitions and drinking behavior.
The current study aimed to replicate and expand our previous study and to
contribute to our understanding of MI, the relationship between therapist and client
behaviors, and substance abuse treatment. ATSS may be helpful in revealing more
about the cognitive experiences of those participating in MI and may help to
identify efficient and powerful cognitive and behavioral change mechanisms that
could be refined and then perhaps applied to other types of therapy as well as to
public health initiatives.
34
Chapter 2: Method
Experimental Design
The study makes use of a 2 X 2 factorial design. Participants were screened
and, if selected, were randomly assigned to one of four ATSS conditions: The first,
featuring a combination of a warm/Rogerian communication style and the use of a
decisional balance exercise, is meant to simulate the approach used in MI. The
second condition features the use of only the warm/Rogerian communication style
without the use of the decisional balance exercise in the context of a general
discussion about alcohol use. The third condition features the use of a
confrontational/educational communication style (often used in non-MI alcohol
counseling) coupled with the decisional balance exercise. The fourth condition
features the use of a confrontational/educational communication style in the context
of a general discussion of alcohol similar to that in condition two
(see table 2 (a copy of table 1) below).
Table 2. The four experimental conditions in the 2X2 experimental design
Decisional Balance NO Decisional
Balance
Warm-Rogerian Condition One (MI) Condition Two
Confrontational/Educational Condition Three Condition Four
Participants were assessed with paper-and-pencil measures at pre-ATSS and
post-ATSS as well as at a follow-up visit one-month (30 days) after the initial
35
assessment session. Participants’ ATSS responses were coded and the ATSS data
and paper-and-pencil data were analyzed.
Participants
315 potential participants were selected and recruited from the University of
Southern California (USC) undergraduate subject pool. A brief self-report measure
that assesses drinking habits (Frequency-Quantity Questionnaire) was included in
the pre-measure packet distributed to students enrolled in the subject pool.
Students who met our criteria for binge drinkers on this questionnaire (described
below) were contacted and recruited to participate in the study. 155 participants
signed up for the study, but because our design required that all students qualify as
binge drinkers in the 30 days leading up to the day they came to the lab for their
first assessment and because many of the 155 people no longer met this criterion
when they arrived later in the semester, 42 participants were no longer qualified by
the time they tried to schedule a time and so were dropped. 10 participants skipped
or failed to schedule their follow-up appointment and 14 participants were dropped
due to large amounts of missing data (due either to malfunctioning audio
equipment during ATSS or because they skipped paper-and-pencil measures),
leaving a final total of 89 participants included in the analyses. The sample was
made up of 70 percent female and 30 percent male students and the self-reported
average age and standard deviation for the sample was 19.5 (1.4) years. Self-
reported ethnicity for the sample was 62 percent Caucasian/white, 17 percent
Asian/Asian-American, 11 percent latino/a, 3 percent African/American, 2 percent
36
Middle Eastern, and 5 percent “other”. All participating students received extra
course credit for their involvement in the study. Strict confidentiality was
maintained throughout the experimental process.
Measures
Screening Measure:
1) Frequency-Quantity Questionnaire. Numerous versions of Frequency-
Quantity questionnaires have been constructed to assess patterns of alcohol
consumption. Originally developed by Cahalan and Cisin (1968), the
version used for the current study assesses the respondents’ report of the
number of times per week and per month alcohol is consumed as well as the
quantity of drinks usually consumed, the maximum number of drinks
consumed on one occasion, and occasions of binge drinking (defined as 4 or
more drinks for women and 5 or more drinks for men). The cut-off for
consideration as a binge drinker and inclusion in the study was two episodes
of binge drinking within the past month (30 days). Please see appendix A.
Questionnaire Dependent Variables: Pre-ATSS and post-ATSS and Follow-up
measure packet:
2) Alcohol Timeline Followback (TLFB; given at pre-ATSS and follow-up
only). Developed by Sobell and Sobell (1992, 1995), this self-report
measure was designed to assess alcohol use over a variety of time periods
(for our study, we used a 30 day version). The measure asks participants to
37
retrospectively recall their use of alcohol by filling in the number of drinks
consumed on a calendar provided to them. Sobell and Sobell (1995) have
shown high concurrent validity and reliability for this method of
assessment. For an example of instructions and a calendar, see appendix B.
3) Readiness to Change Questionnaire (RTCQ). This self-report measure
(Rollnick, Heather, Gold, & Hall, 1992) was developed as a method to
classify participants into one of the first three stages of the Transtheoretical
Model of Change (Prochaska & DiClemente, 1986): precontemplation,
contemplation, or action. The authors found a clear factor structure for the
12 item scale and they found acceptable internal consistency and test-retest
reliability. Please refer to appendix C.
4) Alcohol Stages of Change Ruler (ASCR). The Readiness ruler, also
referred to as the ASCR, was originally developed by Miller as an
additional attempt to measure stage of change/readiness to change as a
continuous variable with respect to a variety of substances (CASAA, 1995).
It was adapted more recently to focus primarily on alcohol and was found to
correlate strongly with the stage of change as indicated by the RTCQ at
three points in time during a study (mean r=.707, p<.001) (LaBrie et al.,
2006). Please refer to appendix D.
5) Frequency/Quantity Intention to Drink (F/Q Intention). Based on the
same retrospective frequency/quantity measure described above, this
measure was developed for this study and asked about the amount the
38
participant intended to drink on four dimensions in the upcoming 30 day
period. Please refer to appendix E.
6) Overall Impressions of Pros and Cons of Drinking (OIDA; given at
post-ATSS and follow-up only). This measure, developed by the
experimenter, was designed to directly assess participants’ subjective
impression of the overall “balance” of the pros and cons of drinking
alcohol. Please refer to appendix F.
7) Overall Impressions of Pros and Cons of Quitting Drinking (OIQA;
given at post-ATSS and follow-up only).. This measure, developed by the
experimenter, was designed to directly assess participants’ subjective
impression of the overall “balance” of the pros and cons of quitting drinking
alcohol. Please refer to appendix G.
8) Experimental Check (ECS; given at post-ATSS only). This is a brief
self-report measure commonly used in ATSS research to assess the
participant’s experience with the ATSS procedure. Please refer to appendix
H.
Other Pre-ATSS Questionnaires
9) Demographics Questionnaire (pre-ATSS packet only). Personal
information variables including gender, age, and race/ethnicity. Please refer
to appendix I.
10) Therapeutic Reactance Scale (TRC; pre-ATSS packet only). Developed
by Dowd, Milne, and Wise (1991), this measure aims to assess
39
psychological reactance conceived of as a trait as defined by Brehm and
Brehm (1981). Please refer to appendix J.
Procedure
Participants were recruited via the Frequency-Quantity questionnaire (see
appendix A) placed in the USC undergraduate psychology subject pool
questionnaire packet. The questionnaire asked students for information on drinking
habits and asked the participants to provide contact information and consent to be
contacted by the experimenter. The participants who met study qualifications for
binge drinkers were contacted by a HIPAA-certified research assistant (RA) to
schedule an appointment time. Please refer to appendix A for rules and a script for
contacting qualified participants.
The study was conducted in two sessions for each participant. In the first
session, participants began by being asked to read and sign an informed consent
form. Second, they were asked to complete the demographics questionnaire and
the pre-ATSS paper-and-pencil measure packet (the demographics questionnaire,
TLFB, TRS, ASCR, RTCQ, F/Q Intention). Third, the experimenter provided a
brief overview of and instructions for the ATSS procedure and addressed any
questions or concerns that the participant expressed. Fourth, following the
instructions, the participant listened to a brief instruction and practice tape
explaining again the goals of ATSS and providing a practice example. This practice
tape was listened and responded to under the supervision of the experimenter and
the participant received feedback to ensure that he/she was participating properly.
40
At this time the participant was given another opportunity to voice questions or
concerns.
During step five, the ATSS procedure, each participant listened to the
cassette for one of the four conditions. Please note that each scenario was reviewed
and evaluated by a group of experienced clinicians to ensure external validity.
Please see appendix K for the measure. In each of the four scenarios, the
participant was asked to imagine him or herself engaging with a counselor in a
discussion of his/her drinking behavior.
Condition One (the Motivational Interviewing) scenario. The counselor
voice featured in the scenario embodied the warm, empathic, non-judgmental, and
open-ended questioning style that is the hallmark of MI. In an additional attempt to
ensure external validity, the segments were structured in the manner of the MI
decisional balance exercise, e.g., in certain segments the participant was asked to
think about the pros of drinking and the cons of drinking. Segments prompted the
participant to respond with cognitions about the pros and cons of drinking as well
as cognitions about the counselor and the process as a whole. Please refer to
appendix L for the scenario script.
Condition two scenario. The counselor voice featured in the scenario
embodied the warm, empathic, non-judgmental, and open-ended questioning style
that is the hallmark of MI. Unlike condition one, in this scenario the counselor did
not make use of the decisional balance exercise and instead asked the participant to
engage in a general conversation about drinking. Segments prompted the
41
participant to respond with cognitions about the drinking questions and topics
raised by the counselor as well as cognitions about the counselor and the process as
a whole. Please refer to appendix M for the scenario script.
Condition three scenario. The counselor voice featured in this scenario was
more stern, less warm, and took a more direct, confrontational, and educational
approach compared to the scenario in conditions one and two, e.g., let the student
know that he/she qualifies as a “problem drinker” and advises that he/she change
immediately. Like the condition one scenario, the segments were structured in the
manner of the decisional balance exercise, e.g., in certain segments the participant
was asked to think about the pros of drinking and the cons of drinking. Segments
prompted the participant to respond with cognitions about the pros and cons of
drinking as well as cognitions about the counselor and the process as a whole.
Please refer to appendix N for the scenario script.
Condition four scenario. Like scenario three, the counselor voice featured in
this scenario was more stern, less warm, and took a more direct, confrontational,
and educational approach compared to the scenario in conditions one and two, e.g.,
let the early student know that he/she qualifies as a “problem drinker” and advises
that he/she change immediately. Like condition two, in this scenario the counselor
did not make use of the decisional balance exercise and instead asked the
participant to engage in a general conversation about drinking. Segments prompted
the participant to respond with cognitions about the drinking questions and topics
42
raised by the counselor as well as cognitions about the counselor and the process as
a whole. Please refer to appendix O for the scenario script.
Once the ATSS scenario section was completed, at step six participants
filled out the post-ATSS paper-and pencil measure packet (ASCR, RTCQ, F/Q
Intention, OIDA, OIQA, ECS). During step seven, the experimenter removed the
ATSS scenario tapes and played for the participant a three minute “pleasant
scenario” tape in order to bring about a more neutral or positive mood state in the
case that the ATSS scenario had generated negative affect. See appendix P for the
pleasant scenario script.
Debriefing (step eight)
All participants received a debriefing in which they were reassured that the
scenario was not real and that the information shared is confidential. The
participant had the opportunity to voice concerns or questions to the experimenter
at this time. The experimenter also provided referral information to the USC
student counseling service for interested participants.
Follow-up (step nine)
All participants were contacted by the experiment team approximately two
weeks after the experiment in order to schedule a follow-up assessment meeting for
a day one month (30 days) after the initial assessment session. At this follow-up
session, the participants completed the follow-up packet which included a TLFB
assessment of drinking behavior in the one month (30 day) time period since the
43
first assessment session as well as paper-and-pencil measures (ASCR, RTCQ, F/Q
Intention, OIDA, OIQA).
ATSS Dependent Variables: The Coding Format
The coding format for this study was designed to assess a number of
cognitive variables. Research assistants were trained for approximately ten hours
before coding began. The coding format was as follows and examples are provided
following the below description:
1. Incidence of cognitions indicating a positive orientation to change (change
cognitions), adapted from the Motivational Interviewing Skills Code 2.0
(CASAA, 2003). These include statements of the following cognitions:
a. Ability to change
b. Commitment to Change
c. Desire to Change
d. Need To Change
e. Reasons to change
f. Taking Steps Towards Change
2. Cognitions indicating a generally cooperative attitude towards the therapist
and therapy situation (therapy cooperation cognitions), adapted from the
Client Resistance Code (CRC) (Kavanagh, Gabrielson, & Chamberlain,
1982)). These include statements reflecting the following types of
cognitions:
44
a. Nonresistant-all statements that are neutral, cooperative, or
following the direction set by the therapist
b. Facilitative-short utterances indicating attention or agreement
3. Incidence of cognitions indicating resistance to change (change resistance
cognitions), also adapted from the Motivational Interviewing Skills Code
2.0 (CASAA, 2003). These include statements that reflect the following
cognitions:
a. Inability to change
b. Commitment not to Change
c. Desire not to Change
d. Lack of Need to Change or Need To Not Change
e. Reasons not to change
f. Taking Steps Away from Change
4. Cognitions indicating a generally resistant attitude (therapy resistance
cognitions) towards the therapist or therapy, also adapted from the Client
Resistance Code (CRC) (Kavanagh, Gabrielson, & Chamberlain, 1982)).
These include statements reflecting the following types of cognitions:
a. Negative attitude cognitions-unwillingness/inability to cooperate
with counselor suggestions. Includes statements of hopelessness,
disagreement, defeat, blaming others
b. Challenge/Confront cognitions-challenging the reputation, skills,
and knowledge of the therapist
45
c. Own agenda-bringing up new/other topics
d. Not tracking-inattention, not responding, disqualifying a previous
statement
Following are some prototypical examples of the coding scheme:
1. Change cognition: “I really need to stop drinking,” would be coded as
one (1) count of desire to change.
2. Therapy cooperative cognition: “Yeah, I agree with what the therapist is
saying,” would be coded as one (1) nonresistant statement.
3. Change resistance cognition: “I don’t see why I need to stop drinking, I
don’t have a problem,” would be coded as one (1) for reason not to change.
4. Therapy resistance cognitions: “I just don’t agree with what this guy is
saying. It isn’t my fault I drink anyway,” would be coded as two (2)
separate therapy resistance cognitions.
Please see appendix Q for the complete coding manual, featuring more detailed
descriptions and examples of the coding format.
Raters were all undergraduate students and the group consisted of four
females and two males. Ethnicity for the raters was as follows: Four were
white/Caucasian, one was Asian-American, and one was Middle Eastern .
Inter-rater reliability was assessed with intra-class coefficients (ICCs;
McGraw & Wong, 1996) for each pair of raters for each code. Reliability checks
were conducted with each quartile of the participant data in an attempt to maintain
reliability and avoid drift. We were able to obtain highly acceptable levels of inter-
46
rater agreement as measured by the ICCs for each of our ATSS variables as
follows: change cognitions=.93, resistance to change cognitions=.88, therapy
cooperation cognitions=.87, and therapy resistance cognitions=.93. The scores for
each pair of raters were then averaged to generate each participant’s final score for
each ATSS variable
47
Chapter 3: Results
Preliminary Analyses
Means and standard deviations were calculated for each dependent variable
for each condition before preliminary analyses were conducted. Please see table 3
for details.
Table 3. Means and stand deviations for each dependent variable grouped by
condition
Condition One
(N=23)
Mean(SD)
Condition Two
(N=24)
Mean(SD)
Condition Three
N=21
Mean(SD)
Condition Four
N=21
Mean(SD)
Pre-ATSS
Measures
ASCR1 4.4(3) 2.9(2.1) 4.1(2.8)
4(2.6)
RTCQ1.SB 2.1 (.92) 1.8(.8) 2(.92)
2(.89)
TRS 70.1(7.6) 68.7(8.6) 72(7.9)
68.5(8.4)
FQI1.1 9.3(2.9) 7.8(3.5) 8.8(2.7)
8.1(3.2)
FQI1.2 3.8(1.6) 4.8(1.4) 4.7(2.1)
4.9(1.8)
FQI1.3 6.8(3.2) 7.9(2.7) 7.9(3.1)
8.5(3.3)
FQ1.4 2.3(2.8 3.9(2.9) 4.1(3.7)
3.3(2.4)
TL1TOT 43.2(19.8) 43.8(18.5) 49.6(26.7)
40.2(18.1)
TL1.MAX 7.5(3) 8.8(2.8) 8.5(3.1)
7.9(2.6)
TL1.DD 11(4.4) 8.8(3.3) 9.5(3.9)
9.1(5)
TL1BNG 5.3(3.2) 5.5(2.8) 6.8(3.2)
4.9(2.3)
TL1.AVDD 4.2(1.8) 5.2(1.6) 5.2(1.7)
5(1.8)
ATSS
Measures
Change cognitions 7.3(3.7) 5.5(3.5) 5.8(4.3)
7.7(6.6)
Resistance to Change
Cognitions
10.1(5) 7.8(5.2) 8.1(3.6)
14.5(6.2)
Therapy Cooperation
Cognitions
3.6(3) 4.4(3.9) 4.1(3.7)
3.2(3.3)
Therapy Resistance
cognitions
3.1(3.6) 4.1(5) 2.5(2.9)
3.4(4.2)
48
(Table 3 continued)
Post-ATSS
measures
ASCR.2 4.3(2.8) 3.1(2.2) 4.3(2.6)
4.2(2.8)
RTCQ2.SB 2.3(.93) 1.9(.9) 1.9(.83)
1.8(.87)
FQ2.1 8.8(3.2) 7.9(3.5) 8.6(2.7)
8(3.5)
FQ2.2 3.6(1.5) 4.7(1.2) 4.7(2)
4.8(1.8)
FQ2.3 6(2.8) 7.7(2.8) 7.6(3)
8.1(3.5)
FQ2.4 2.1(2.8) 3.5(2.6) 4(3.7)
3.6(2.9)
OVD2 25.2(38.6) 27.5(33.4) 27(42.9)
8.4(52.1)
OVQ2 -7.8(46.8) -14.9(39.6) -28(60.6)
-14.7(51.1)
REAL 7.8(2) 8.8(1.6) 8.1(1.9)
7.8(2.1)
WARM 7.5(2.4) 9.6(2.1) 5.7(2.7)
6.3(2.5)
Follow-Up
Measures
ASCR3 5.4(3.2) 3.4(2.8) 4(3)
3.9(2.8)
RTCQ3.SB 2.1(.95) 1.6(.8) 1.8(.9)
1.9(.91)
FQ3.1 7.6(3.2) 7.9(3.6) 8.6(2.5)
8.1(2.8)
FQ3.2 3.8(2) 4.7(1.3)
4.7(1.9) 4.8(2.2)
FQ3.3 6.7(3.2) 7.2(2.2)
8(3.5) 8.4(3.1)
FQ3.4 3.1(3.5) 3.7(3)
4.7(3.4) 4.1(2.9)
OVD3 15.7(37.6) 26.3(35.2)
32.1(45.1) 4.9(47.9)
OVQ3 -7.4(52.7) -17.4(37.4)
-22.9(52.9) 3.5(55.2)
TL2TOT 35.4(24.4) 39(25.9)
43.1(25.1) 34.7(16.1)
TL2.MAX 6.7(3.1) 6.8(2.8)
7.9(3.2) 7.9(3.5)
TL2.DD 8.5(3.9) 8.3(4.1)
8.8(3.5) 8.5(4.1)
TL2BNG 4.1(3.8) 5.3(3.9)
5.8(3.7) 4.1(2.1)
TL2.AVDD 4(1.9) 4.3(1.8)
4.9(1.8) 4.4(1.7)
Note.
ASCR1=Alcohol Stages of Change Ruler, RTCQ1.SB=Readiness to Change
Questionnaire, TRA=Therapeutic Reactance Scale, FQ1.1=F/Q Intention item 1,
F/Q1.2 Intention item 2, FQ1.3=F/Q Intention item 3, FQ1.4=F/Q Intention item 4,
TL1TOT=Timeline Followback total number of drinks, TL1.MAX= Timeline
Followback max. drinks on one occasion, TL1.DD=Timeline Followback drinking
days, TL1BNG= Timeline Followback number of binge episodes, TL1.AVDD=
Timeline Followback average drinks per drinking occasion
49
(Table 3 continued)
ASCR2=Alcohol Stages of Change Ruler, RTCQ2.SB=Readiness to Change
Questionnaire, FQ2.1=F/Q Intention item 1, F/Q2.2=F/Q Intention item 2,
FQ2.3=F/Q Intention item 3, FQ2.4=F/Q Intention item 4, OVD2=Overall
Impressions of Drinking, OVQ2=Overall Impressions of Quitting,
REAL=realism/vividness scale of experimental check, WARM=warm/Rogerian
scale of experimental check
ASCR3=Alcohol Stages of Change Ruler, RTCQ3.SB=Readiness to Change
Questionnaire, FQ3.1=F/Q Intention item 1, F/Q3.2=F/Q Intention item 2,
FQ3.3=F/Q Intention item 3, FQ3.4=F/Q Intention item 4, OVD3=Overall
Impressions of Drinking, OVQ3=Overall Impressions of Quitting,
TL2TOT=Timeline Followback total number of drinks, TL2.MAX= Timeline
Followback max. drinks on one occasion, TL2.DD=Timeline Followback drinking
days, TL2BNG= Timeline Followback number of binge episodes,
TL2.AVDD=Timeline Followback average drinks per drinking occasion
Outliers and the Normality assumptions
Several preliminary analyses were conducted to determine whether there
were any missing data, whether there were any univariate or multivariate outliers,
and whether there were any violations of the assumptions required for the statistical
methods used in our primary analyses (Wilcox, 2003). Once conducted, these
preliminary analyses determined whether any adjustments to the data were
required, e.g., winsorizing, removing extreme outliers, or transformation.
Paper-and-pencil measures
First, the paper-and-pencil measures were screened for missing data. There
were no missing data. Second, the paper-and-pencil measures were screened for
univariate and multivariate outliers. Several univariate outliers were found on
various measures (one each on the Therapeutic Reactance Scale, frequency/quantity
intention to drink at follow-up questions one, three, and four, binge episodes for
50
the first administration of the timeline follow-back, and total drinks, maximum
number of drinks on one occasion, and average number of drinks per drinking day
for the second administration of the timeline followback. Two univariate outliers
were found for total number of drinks on the first administration of the timeline
followback. These values were all winsorized, i.e., made equivalent to its closest
neighbor (Guttman, 1973). By using Mahalanobis distance with p < .001, no cases
were identified as multivariate outliers with p < .001 (Tabachnick and Fidell,
1996).
Finally, the paper-and-pencil measures were screened for unacceptable
violations of skewness. The following variables had slightly-to-moderately
unacceptable violations of skewness: intention to drink at pre-ATSS question four
(standardized skewness=3.2), intention to drink at post-ATSS question four
(standardized skewness=3.12), intention to drink at follow-up question two
(standardized skewness=3.22), intention to drink at follow-up question three
(standardized skewness=3.5), total drinks for the first administration of the timeline
followback (standardized skewness=4.1), total drinks for the second administration
of the timeline followback (standardized skewness=4.6), maximum number of
drinks per occasion for the first administration of the timeline followback
(standardized skewness=3.3), number of binge episodes for the first administration
of the timeline followback (standardized skewness=3.6), and number of binge
episodes for the second administration of the timeline followback (standardized
skewness=4). Each of these variables was transformed using the least extreme
51
possible transformation: a square-root transformation (Tabachnick and Fidell,
1996). In all cases, this transformation worked to normalize the distributions and
thus make the variables amenable to analysis using the proposed statistical
methods.
ATSS dependent variables
First, the coded ATSS variables were screened for missing data. There
were no missing data. Second, ATSS variables were screened for univariate and
multivariate outliers. Resistance to change cognitions and therapy resistance
cognitions each had one case of a univariate outlier and therapy cooperation
conditions had two outliers. These values were winsorized (Guttman, 1973). By
using Mahalanobis distance with p < .001, no cases were identified as multivariate
outliers.
Finally, the ATSS measures were screened for unacceptable violations of
skewness. The following variables had slightly-to-moderately unacceptable
violations of skewness: change cognitions (standardized skewness=3.4), therapy
cooperation cognitions (standardized skewness=3.03), and therapy resistance
cognitions (standardized skewness=6). Again, each of these variables was
transformed using a square-root transformation (Tabachnick and Fidell, 1996) and
in all cases this worked to normalize the distributions and thus make the variables
amenable to analysis using the proposed statistical methods.
52
Screening for Initial Group Differences
Participants were recruited from the USC undergraduate subject pool from
the Spring 2005 (semester one), Fall 2006 (semester two), and Spring 2006
(semester three) semesters. ANOVA, MANOVA, and Chi-square analyses were
conducted to ensure that participants from the different semesters did not differ
significantly in terms of demographic characteristics and in terms of the initial
drinking measures. No significant differences were found on all variables (please
see table 4 for a summary), providing justification for considering these
populations to be equivalent and, therefore, to be combined into one final group of
89 participants.
ANOVA and Chi-square analyses were also conducted on the demographics
and screening measure for each of the four conditions to ensure that groups within
the sample did not differ significantly on these variables. No significant
differences were found (please see table 5 for a summary) and so we were able to
conclude that our groups were equivalent for these variables.
53
Table 4. Demographic differences between semester one, two, and three
Note. “Other” category in race variable may include African-American, middle eastern, native
American, and those who endorse more than one race category, e.g., asian-american and
white/Caucasian
TL1.tot=timeline followback 1 total drinks consumed by participant, TL1.max=timeline followback
1 maximum number of drinks on a single occasion, TL1.dd=timeline followback 1 number of days
on which participant consumed at least one drink, TL1.bng=timeline followback 1 number of binge
drinking episodes, TL1.avdd=timeline followback 1 average number of drinks consumed for all
days on which participant consumed at least one drink
Table 5. Demographic differences among the experimental groups
Variable
Condition 1
mean/sd
Condition 2
mean/sd
Condition 3
mean/sd
Condition 4
mean/sd Stat Value
Sig.
level
Age
19.7/1.7 19.4/1.2 19.3/1.1 19.8/1.7
ANOVA F=.659 0.580
TL1.tot,
TL1.max,
TL1.dd,
TL1.bng,
TL1.avdd
6.4/4.9 6.8/5.0 5.8/4.4 7.0/4.7
MANOVA
(wilk’s
lambda) F=1.2 0.272
Variable
Condition 1
male/female
ratio
Condition 2
male/female
ratio
Condition 3
male/female
ratio
Condition 4
male/female
ratio Stat Value
Sig.
level
Sex
6/17 9/15 4/17 8/13
Chi-square 2.64 0.45
Variable
Condition 1
race %
Condition 2
race %
Condition 3
Race %
Condition 4
race % Stat Value
Sig.
level
Race
White=52.1
As-Am=26.1
Latino/a=8.7
Other=10.1
White=70.9
As-Am=8.3
Latino/a=12.5
Other=8.3
White=61.9
As-Am=9.5
Latino/a=14.3
Other=14.3
White=61.9
As-Am=23.8
Latino=9.5
Other=4.8 Chi-square 17.33 0.299
Variable
Semester 1
mean/sd
Semester 2
mean/sd
Semester 3
mean/sd Stat Value
Sig.
level
Age
19.6/1.7 19.3/1.2 19.9/1.2
ANOVA F=1.0 0.371
TL1.tot,
TL1.max,
TL1.dd,
TL1.bng,
TL1.avdd
6.4/4.7 7.1/4.5 7.1/4.5
MANOVA
(Wilk’s lambda) F=1.3 0.229
Variable
Semester One
male/female
ratio
Semester Two
male/female
ratio
Semester Two
male/female
ratio Stat Value
Sig.
level
Sex
12/23 10/24 5/15
Chi-square .542 0.76
Variable
Semester 1 race
%
Semester 2
race %
Semester 3
race % Stat Value
Sig.
level
Race
White=54.2
As-Am=28.6
Latino/a=8.6
Other=8.6
White=67.6
As-Am=5.9
Latino/a=11.8
Other=14.7
White=65
As-Am=15
Latino/a=15
Other=5 Chi-square 11.23 0.34
54
(Table 5 continued)
Note.“Other” category in race variable may include African-American, middle
eastern, native American, and those who endorse more than one race category, e.g.,
asian-american and white/Caucasian
TL1.tot=timeline followback 1 total drinks consumed by participant,
TL1.max=timeline followback 1 maximum number of drinks on a single occasion,
TL1.dd=timeline followback 1 number of days on which participant consumed at
least one drink, TL1.bng=timeline followback 1 number of binge drinking
episodes, TL1.avdd=timeline followback 1 average number of drinks consumed for
all days on which participant consumed at least one drink
Word Count
We assessed the number of words articulated by each subject during the
ATSS scenario as an index of ability and willingness to verbalize. The groups in
each condition did not differ from one another in terms of word count,
F(1,88)=.556, p=.65, providing support for the assumption that participants in all
groups had an equivalent ability and willingness to verbalize their cognitions (and
that any between group differences in ATSS dependent variables are not just an
artifact of between group differences in verbalization).
Inter-item Correlations
Pearson’s correlations were computed for four grouping of variables: first,
the pre-ATSS dependent variables, second, the ATSS dependent variables, third,
the post-ATSS dependent variables, and fourth, the follow-up dependent variables.
Overall, inter-measure correlations were in the expected directions and strengths.
For example, the readiness to change measures (the RTCQ and the ASCR) were
highly correlated, the drinking measures were correlated with each other as well as
55
with the intention to drink (greater intention to drink was associated with more
drinking) and readiness to change measures (lower readiness to change was
associated with more drinking), and the overall impressions of drinking and
quitting measures were correlated with several drinking variables (more positive
impressions of drinking was associated with more drinking) and with readiness to
change measures (a more positive impression of drinking was associated with
lower readiness to change). Please refer to tables 6,7,8, and 9 for all of the
correlations.
56
Table 6. Pre-ATSS Paper-and-Pencil Measures Inter-item Correlations
TRS1.
T
ASCR
1
RTCQ
1.SB
FQI1.1 FQI1.2 FQI1.3
FQ1.4
TR
TL1TO
TTR
TL1.M
AX
TL1.D
D
TL1B
NGTR
ASCR
1
-0.086
RTCQ
1.SB
.099 .53**
FQI1.1
.063 -.112 -.059
FQI1.2
-.043 -.236* -.100 -.119
FQI1.3
.174 -.254* -.125 .027 .74**
FQ1.4
TR
.068 -.33** -.241* .245* .69** .71**
TL1TO
TTR
.039 -.204 -.134 .46** .36** .40** .49**
TL1.M
AX
.130 -.101 .046 -.061 .54** .64** .38** .48**
TL1.D
D
.057 -.099 -.142 .65** -.213* -.061 .127 .69** -.064
TL1B
NGTR
-.035 -.222* -.157 .36** .33** .28** .50** .86** .270* .52**
TL1.A
VDD
-.024 -.057 .078 -.267* .75** .67** .44** .36** .78** -.36** .33**
Note. **. Correlation is significant at the 0.01 level (2-tailed).*. Correlation is
significant at the 0.05 level (2-tailed).
ASCR1=Alcohol Stages of Change Ruler, RTCQ1.SB=Readiness to Change
Questionnaire, FQ1.1=F/Q Intention item 1, F/Q Intention item 2, FQ1.3=F/Q
Intention item 3, FQ1.4TR=F/Q Intention item 4, TL1TOTTR=Timeline
Followback total number of drinks, TL1.MAX= Timeline Followback max. drinks
on one occasion, TL1.DD=Timeline Followback drinking days, TL1BNGTR=
Timeline Followback number of binge episodes, TL1.AVDD= Timeline
Followback average drinks per drinking occasion
57
Table 7. ATSS Dependent Variable Correlations
CCOGS.AT CHRES.AT TCCOG.AT TRCOG.AT
Pearson
Correlation
1 .154 -.185 -.111
Sig. (2-tailed)
. .150 .082 .302
CCOGS.AT
N
89 89 89 89
Pearson
Correlation
.154 1 -.354(**) .094
Sig. (2-tailed)
.150 . .001 .381
CHRES.AT
N
89 89 89 89
Pearson
Correlation
-.185 -.354(**) 1 -.456(**)
Sig. (2-tailed)
.082 .001 . .000
TCCOG.AT
N
89 89 89 89
Pearson
Correlation
-.111 .094 -.456(**) 1
Sig. (2-tailed)
.302 .381 .000 .
TRCOG.AT
N
89 89 89 89
Note. ** Correlation is significant at the 0.01 level (2-tailed).
CCOGS=change cognitions, CHRES=change resistance cognitions,
TCCOGS=therapy cooperation cognitions, TRCOGS=therapy resistance cognitions
58
Table 8. Post-ATSS Paper-and-Pencil Measures Inter-item Correlations
Variable ASCR2 RTCQ2.SB FQ2.1 FQ2.2 FQ2.3 FQ2.4TR OVD2 OVQ2 REAL
RTCQ2.SB .414**
FQ2.1 -.268* -.083
FQ2.2 -.235* -.044 -.018
FQ2.3 -.270* -.016 .172 .775**
FQ2.4TR -.332** -.119 .370** .735** .759**
OVD2 -.274** -.306** .274** .101 .187 .328**
OVQ2 .222* .337** -.320** -.133 -.187 -.316** -.540**
REAL -.212* -.039 -.017 -.139 -.158 -.172 -.123 .020
WARM .015 .076 -.023 -.179 -.109 -.144 -.053 .152 .281**
Note. **. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
ASCR2=Alcohol Stages of Change Ruler, RTCQ2.SB=Readiness to Change
Questionnaire, FQ2.1=F/Q Intention item 1, F/Q2.2=F/Q Intention item 2,
FQ2.3=F/Q Intention item 3, FQ2.4TR=F/Q Intention item 4, OVD2=Overall
Impressions of Drinking, OVQ2=Overall Impressions of Quitting, Real=Realistic
Engaged Index from Experimental Check, Warm=Nonspecifics index from
Experimental Check
59
Table 9. Follow-Up Paper-and-Pencil Measures Inter-item Correlations
ASCR
3
RTCQ
3.SB
FQ3.1 FQ3.2
TR
FQ3.3
TR
FQ3.4 OVD3 OVQ3 TL2T
OTTR
TL2.
MAX
TL2.D
D
TL2B
NGTR
RTCQ
3.SB
.6
**
FQ3.1
-.47
**
-.34
**
FQ3.2
TR
-.245
*
-.14 .255
*
FQ3.3
TR
-.309
**
-.17 .374
**
.760
**
FQ3.4
-.336
**
-.28
**
.576
**
.724
**
.776
**
OVD3
-.341
**
-.35
**
.360
**
.116 .102 .234
*
OVQ3
.273
**
.276
**
-.334
**
-.174 -.072 -.252
*
-.662
**
TL2T
OTTR
-.292
**
-.128 .608
**
.500
**
.574
**
.615
**
.254
*
-.245
*
TL2.
MAX
-.206 -.079 .386
**
.528
**
.692
**
.528
**
.067 -.063 .693
**
TL2.D
D
-.207 -.109 .641
**
.065 .206 .260
*
.265
*
-.233
*
.763
**
.363
**
TL2B
NGTR
-.342
**
-.202 .543
**
.502
**
.499
**
.594
**
.270
*
-.318
**
.897
**
.554
**
.609
**
TL2.A
VDD
-.227
*
-.102 .248
*
.754
**
.712
**
.697
**
.063 -.102 .668
**
.698
**
.061 .668
**
Note. **. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
ASCR3=Alcohol Stages of Change Ruler, RTCQ3.SB=Readiness to Change
Questionnaire, FQ3.1=F/Q Intention item 1, F/Q3.2=F/Q Intention item 2,
FQ3.3TR=F/Q Intention item 3, FQ3.4=F/Q Intention item 4, OVD3=Overall
Impressions of Drinking, OVQ3=Overall Impressions of Quitting,
60
(Table 9 continued)
TL2TOTTR=Timeline Followback total number of drinks, TL2.MAX= Timeline
Followback max. drinks on one occasion, TL2.DD=Timeline Followback drinking
days, TL2BNGTR= Timeline Followback number of binge episodes,
TL2.AVDD=Timeline Followback average drinks per drinking occasion
Primary Analyses
Demographic data for the sample are detailed in table 10. Data for the
entire sample were combined given that no statistically significant differences
across cells or semesters were detected.
Table 10. Demographic data for the sample.
Variable
Sex (%) female/male 70/30
Race (%) white 62
asian-american 17
latino 11
middle eastern 2
african american 3
other 5
Age mean/std. dev. 19.5/1.4
Main Effects Hypotheses: Therapist style
Our first set of analyses was conducted on the data to determine whether
there were main effects for therapist style on the dependent variables as predicted
by our hypotheses.
61
ATSS Dependent Variables hypotheses
We obtained mixed results for our hypotheses about main effects on ATSS
dependent variables. Each of these variables was analyzed with a univariate
ANOVA and, contrary to our hypothesis, no main effects were found for therapist
style on change cognitions, therapy cooperation cognitions, or therapy resistance
cognitions. However, we did find a main effect for therapist style on change
resistance cognitions, F (1,85)=4.71,p<.05. To assess the details and direction of
the main effect found, an independent samples T-test was then conducted
comparing the therapy style groups (warm/Rogerian vs.
confrontational/educational), t(87) = -2.005, p < .05 (effect size, d=.41). This effect
is in the direction hypothesized, i.e., the warm/Rogerian condition was associated
with significantly fewer change resistance cognitions than the
confrontational/educational therapy style.
Pre-ATSS, Post-ATSS, and Follow-Up Dependent variables hypotheses
We conducted several univariate and repeated measures analyses to
examine hypothesized main effects for therapist style on paper and pencil measures
(the drinking measures, the ASCR, the RTCQ, the F/Q Intention, the OIDA, the
OIQA, and items seven and eight on the experimental check) and also obtained
mixed results.
Using repeated measures analyses, no main effects were found on the
ASCR or the frequency/quantity intention to drink measure. Using univariate
ANOVAs, no main effects were found for either the OIDA or the OIQA.
62
However, we did find a main effect for therapist style on the RTCQ that was
consistent with one of our hypotheses. Using a repeated measures analysis, an
interaction between time and therapist style was significant using the criterion of
Wilk’s lambda, F (2,86)=3.7,p<.05. Follow-up analyses revealed a within subjects
quadratic contrast that was significant, F(1,87)=4.63,p<.05, indicating that each
therapist condition displayed a different within subjects pattern over time. Please
refer to figure 1 for a graph of these patterns. A series of paired sample t-tests was
then conducted to verify the significance of these patterns (please see table 11).
These tests showed that the warm/Rogerian therapist style was associated with a
strong trend (p=.06, d=.28) to increased readiness to change immediately after the
ATSS scenario and then a significant decrease (p<.05, d=.30) at follow-up that
returned the participants to a readiness to change score statistically equivalent to the
score they received before the ATSS scenario. These paired t-tests also confirmed
that this pattern was significantly different from the pattern displayed by the
participants in the confrontational/educational condition; for these participants, they
showed a strong trend (p=.05, d=.30) to decrease immediately after the ATSS
scenario and to then maintain that lower level of readiness to change at follow-up.
While not all of these t-tests were significant, at the very least the significant
findings show that the confrontational/educational therapist condition was
associated with a drop in readiness to change from initial levels that was
maintained over time whereas the warm/Rogerian therapist condition was
63
associated with a tendency towards a jump or increase in readiness to change that
returned to baseline over time.
Figure 1. Plot of significant quadratic within subjects contrast for therapist style as
measured by the Readiness to Change Questionnaire.
Estimated Marginal Means of MEASURE_1
TIME
3 2 1
Estimated Marginal Means
2.2
2.1
2.0
1.9
1.8
1.7
THERP
1.00
2.00
Note. THERP 1=warm/Rogerian therapist style, THERP
2=confrontational/educational therapist style; Time 1=RTCQ at pre-ATSS, Time
2=RTCQ at post-ATSS Time 3=RTCQ at follow-up
64
Table 11. Paired Samples T-tests associated with significant within subjects
quadratic contrast for the readiness to change questionnaire
Conditions t df Effect
Sizes
1/2
RTCQ1.1 - RTCQ2.1 -1.930 46 .28
++
RTCQ2.1 - RTCQ3.1 2.046 46 .30*
RTCQ1.1 - RTCQ3.1 .843 46 .12
n/s
RTCQ1.2 - RTCQ2.2 2.011 41 .31
+
RTCQ2.2 - RTCQ3.2 .388 41 .06
n/s
RTCQ1.2 - RTCQ3.2 1.937 41 .30
++
Note. 1)
ns
= not significant,
*
= p < .05;
+
=p=.05,
++
=p<.07
2) Effect size is represented in terms of Cohen’s d
RTCQ1.1=Readiness to Change Questionnaire (RTCQ) at pre-ATSS for
warm/Rogerian therapist condition,
RTCQ2.1=RTCQ at post-ATSS for warm/Rogerian therapist condition,
RTCQ3.1=RTCQ at follow-up for warm/Rogerian therapist condition,
RTCQ1.2=RTCQ at pre-ATSS for confrontational/educational therapist condition,
RTCQ2.2=RTCQ at post-ATSS for confrontational/educational therapist condition
RTCQ3.2=RTCQ at follow-up for confrontational/educational therapist condition
While four out of five of the drinking variables showed no effect for
therapist style, there was a non-significant trend for one variable: maximum
number of drinks consumed on one occasion. For this variable, a univariate
ANOVA, F(1,84)=3.462,p<.07, and subsequent t-test, t(87)=-1.705,p<.1, showed
that the confrontational/educational therapist style was associated with a higher
number of maximum drinks at follow-up (once maximum drinks at the initial
assessment were covaried out) than the warm/Rogerian therapist style. This
statistically non-significant trend is consistent with our hypotheses.
65
Additionally, two non-significant trends consistent with hypotheses were
identified in the experimental check items. For experimental check question 7 (“To
what degree did you feel the experience with the counselor influenced you to think
about changing your drinking?”), a univariate ANOVA, F(1,85)=1.688,p<.1, and
subsequent t-test, t(87)=1.667,p<.1, showed that the warm/Rogerian therapist was
associated with a trend towards higher scores on this item. For experimental check
question 8 (“To what degree did you feel the experience with the counselor
influenced you to think about NOT changing your drinking?”), a univariate
ANOVA, F(1,85)=2.946,p<.1 and subsequent t-test, t(1,87)=-1.733,p<.9, showed
that the confrontational/educational therapist style was associated with a trend
towards higher scores on this item.
Main Effects Hypotheses: Decisional Balance
Our second set of analyses was conducted on the data to determine whether
there were main effects for the decisional balance on the dependent variables as
predicted by our hypotheses.
ATSS dependent variables
Change cognitions, therapy cooperation cognitions, and therapy resistance
cognitions were each analyzed with univariate ANOVAs and we found no main
effect for decisional balance on these variables. A non-significant trend,
F(1,85)=3.516,p<.07, was detected for change resistance cognitions and an
exploration of the marginal means indicates that the conditions using the decisional
balance (marginal mean=9.101) were associated with fewer change resistance
66
cognitions than the conditions without the decisional balance (marginal
mean=11.137). This trend was contrary to our hypotheses and to the findings in
our previous study.
Pre and Post Dependent variables hypotheses
None of the hypothesized main effects for decisional balance on pre-ATSS,
post-ATSS, and follow-up paper and pencil measures (the drinking variables, the
ASCR, the RTCQ, the F/Q Intention, the OIDA, the OIQA, and items seven and
eight on the ECS) was supported by the data.
Interaction Effects Hypotheses: Therapist style X Decisional Balance
To remind the reader, we had several hypotheses regarding interaction
effects of therapist style and decisional balance. For sake of terminological ease for
the remaining results and discussion sections, table 12 (a copy of table 1) appears
below with more brief labels for the four conditions.
Table 12.The four experimental conditions in the 2X2 experimental design
Decisional Balance NO Decisional
Balance
Warm-Rogerian Warm/DB+ (1) Warm/DB- (2)
Confrontational/Educational Confront/DB+ (3) Confront/DB- (4)
67
ATSS dependent variables hypotheses
One significant interaction effect was found for one of the ATSS variables:
resistance to change cognitions, F(1,85)=14.53, p<.001. T-tests were conducted for
each condition to determine the nature of this effect. The fourth condition
(confront/DB-) was associated with significantly more resistance to change
cognitions than the other three conditions (none of which was significantly
different from the others). The t-tests and associated effect sizes are summarized in
table 13. This effect lends support to one aspect of our hypothesis that the fourth
condition (confront/DB-) would be associated with a unique effect on ATSS
variables (we hypothesized that this condition would indeed be associated with
more change resistance cognitions than all of the other conditions), but the other
aspects of this hypothesis regarding differing levels of change cognitions, therapy
cooperation cognitions, and therapy resistance cognitions were not supported. Our
hypotheses regarding an interaction effect for the condition meant to simulate MI
(Warm/DB+) were not supported by the data.
68
Table 13. Interaction effect statistics for therapist style X decisional balance for
change resistance cognitions
Conditions t df Effect
Sizes
1/2
Warm/DB+ v. Warm/DB- 1.594 45 .46
ns
Warm/DB+ v. Confront/DB+ 1.543 42 .46
ns
Warm/DB+ v. Confront/DB- -2.579 42 .73
*
Warm/DB- v. Confront/DB+ -.237 43 .07
ns
Warm/DB- v. Confront/DB- -3.971 43 1.02
***
Confront/DB+ v. Confront/DB- -.4.093 40 1.07
***
Note. 1)
ns
= not significant,
*
= p < .05;
**
= p < .01;
***
= p < .001
2) Effect size is represented in terms of Cohen’s d
Condition 1=warm/Rogerian with decisional balance, 2=warm/Rogerian without
decisional balance, 3=confrontational/educational with decisional balance,
4=confrontational/educational without decisional balance
Pre and Post Dependent Variables Hypotheses
The results also do not support our hypotheses regarding unique interaction
effects for the Warm/DB+ condition or for the Confront/DB- condition for any of
the dependent variables.
Additional Analyses
Checks on the validity of the ATSS scenarios
We employed several single item measures in the post-ATSS measure
packet (collectively known as the “experimental check scale” or “ECS” in
69
appendix G) as a way to analyze whether the participants perceived the ATSS
scenarios as we intended. There were two sets of theoretically related measures
that were correlated. The first set of measures inquired as to whether the
participants found the ATSS scenarios to be vivid, realistic, and whether they were
able to really imagine themselves in the scenario during the experiment (items 1, 2,
and 3). This set of experimental checks serves to indicate whether the participants
were engaged in the task and whether they found the scenario to be appropriately
realistic (especially important given the analogue nature of the ATSS process).
These three theoretically-related questions were correlated (please see table 14) and
were therefore collapsed into a single “realistic and engaged” index. When
combined into this one subscale, the measure of internal consistency, or alpha, was
determined to be .64. No group differences were found on this “realistic and
engaged” index, F(3,85)=1.545, p=.209. Moreover, the mean and standard
deviation for all subjects was 8.16 (1.92) out of a possible 12, indicating a moderate
level of perceived realism, vividness, and engagement that was consistent across all
scenarios.
70
Table 14. Inter-item correlations for the participant experimental check scale
(ECS)
ECS1 ECS2 ECS3 ECS4 ECS5 ECS6
Pearson
Correlation
1 .365(**) .290(**) .353(**) .209(*) .343(**)
Sig. (2-
tailed)
. .000 .006 .001 .050 .001
ECS1
N
89 89 89 89 89 89
Pearson
Correlation
.365(**) 1 .489(**) .100 .089 .189
Sig. (2-
tailed)
.000 . .000 .353 .408 .076
ECS2
N
89 89 89 89 89 89
Pearson
Correlation
.290(**) .489(**) 1 .119 .096 .239(*)
Sig. (2-
tailed)
.006 .000 . .266 .373 .024
ECS3
N
89 89 89 89 89 89
Pearson
Correlation
.353(**) .100 .119 1 .797(**) .683(**)
Sig. (2-
tailed)
.001 .353 .266 . .000 .000
ECS4
N
89 89 89 89 89 89
Pearson
Correlation
.209(*) .089 .096 .797(**) 1 .674(**)
Sig. (2-
tailed)
.050 .408 .373 .000 . .000
ECS5
N
89 89 89 89 89 89
Pearson
Correlation
.343(**) .189 .239(*) .683(**) .674(**) 1
Sig. (2-
tailed)
.001 .076 .024 .000 .000 .
ECS6
N
89 89 89 89 89 89
Note. ** Correlation significant at .01 level (2-tailed), *Correlation is significant at
.05 level (2-tailed).
The second set of questions (4,5, and 6) inquire about to the extent to which
participants found the therapist in the scenario to exhibit the non-specifics of
therapy (empathy, warmth, and genuineness) as described originally by Carl
Rogers (1951, 1961) These theoretically-related variables were also correlated
(again, see table 14) and therefore collapsed into one measure of non-specifics. The
71
alpha for this scale was determined to be .88. We applied a univariate ANOVA to
the non-specifics variable we derived in order to determine whether the scenarios
we intended to reflect the warm/Rogerian therapist condition were in fact perceived
by the participants as more warm/Rogerian, i.e., rated more highly on the non-
specifics scale, than the scenarios designed to reflect the
confrontational/educational therapist condition. We found that there was indeed a
significant difference, F(1,85)=22.48,p<.001. We then performed a t-test and
confirmed that the warm/Rogerian therapist condition was rated more highly on the
non-specifics scale, t(87) = 4.741, p < .001, with an effect size of Cohen’s d=.9.
We also distributed a similar set of single-item measures before we began the study
to three experienced clinicians as a way to analyze whether clinical experts would
perceive the ATSS scenarios as we intended (please refer to appendix H). The first
two questions inquired as to whether the experts found the ATSS scenarios to be
vivid and realistic (items 1 and 2, respectively). We found that the experts
considered the scenarios to be moderately-to-very realistic (mean (std. dev.)=3.37
(.57) on a four point scale) and moderately vivid (mean (std. dev.)=3.54 (.58) on a
four point scale). Univariate ANOVAs revealed no differences on these variables
by condition (Realistic: F(3,20)=.426,p=.74, Vivid: F(3,20)=1.1,p=.37), indicating
that all conditions were perceived by the experts as equally vivid and realistic.
The second set of questions (3,4, and 5) inquired about to the extent to
which the experts found the therapist in the scenario to exhibit the non-specifics of
therapy (warmth, empathy, and genuineness). These variables were correlated
72
(please see table 15) and therefore collapsed into one measure of non-specifics. The
alpha for this scale was determined to be .82. We conducted a univariate ANOVA
and found a significant effect for therapist style, F(1,22)=50.571, p.<.001. A
follow-up t-test found significant group differences between the warm/Rogerian
and confrontational/educational therapist conditions in the expected direction (the
warm/Rogerian therapist style was higher than the confrontational/educational
therapist style), t(22)=7.11,p<.001, with an effect size, d=1.6.
Table 15. Inter-item correlations for expert experimental check scale
Correlations
Q3 Q4 Q5
Pearson
Correlation
1 .912(**) .435(*)
Sig. (2-tailed)
. .000 .033
Q3
N
24 24 24
Pearson
Correlation
.912(**) 1 .406(*)
Sig. (2-tailed)
.000 . .049
Q4
N
24 24 24
Pearson
Correlation
.435(*) .406(*) 1
Sig. (2-tailed)
.033 .049 .
Q5
N
24 24 24
Note. ** Correlation is significant at the 0.01 level (2-tailed).
* Correlation is significant at the 0.05 level (2-tailed).
Post-Hoc Exploration of the Therapeutic Reactance Scale
While there were no hypotheses regarding the relationships between
variables without regard to experimental condition, the data collected generated
some interesting post-hoc questions about the TRS.
73
While the TRS that was administered in the pre-ATSS packet did not have
any significant effects as a covariate in our primary analyses, we explored the
relationship between this trait measure of reactance and potentially related
variables. Our first exploration was meant to determine whether the trait reactance
measured by the TRS might have moderated or helped to predict the degree to
which participants articulated thoughts about change and about the therapist as
measured by ATSS variables. None of the regression analyses indicated that the
TRS was predictive of any of the ATSS variables. We were also curious as to
whether trait reactance was associated with the degree to which participants were
able to engage in the ATSS procedure (the “realistic and engaged” index from the
ECS) and the degree to which participants found the various therapist styles
empathic, warm, and genuine (the warm/Rogerian index from the ECS). No
relationship was found in either case. However, we did find that the TRS was
correlated with the ECS item 8 (“To what degree did you feel the experience with
the counselor influenced you to think about NOT changing your drinking?”) and a
linear regression analysis was conducted to evaluate the prediction of this item
from scores on the TRS. The following regression equation was found to be
significant:
Predicted score on ECS 8=.032 (trait reactance) - .274
The 95% confidence interval for the slope, .006-.059, does not include zero and
therefore trait reactance is significantly related to the score on ECS 8. The R-
74
squared value was .063, meaning that approximately 6% of the variance in ECS can
be accounted for by the score on the TRS.
75
Chapter 4: Discussion
Our results can be summarized as follows. First, we have evidence that
therapist communication style has an impact on client cognitions reflecting
resistance to change their drinking and on client readiness to change and we have
tentative evidence that therapist communication style can affect client drinking
behavior. Second, we have evidence that particular combinations of therapist
communication style and the decisional balance exercise can have an impact on a
client’s cognitions related to resistance to change. Third, we have evidence that the
ATSS methodology used in this study was a useful and valid way to assess the
cognitions of people engaged in a psychotherapeutic interaction. Fourth, we have
evidence that a client’s level of trait reactance may predict client response to the
therapist, independent of therapy communication style. We turn now to a
discussion of each of these findings.
Consistent with our hypotheses, the warm/Rogerian therapist
communication style was associated with significantly fewer change resistance
cognitions than the confrontational/educational therapist communication style.
This result replicates one of the findings from our earlier study (Grodin et al., 2004)
and is consistent with the work of Patterson and Forgatch (1985), which found that
clients’ verbal non-compliant behavior increased more in sessions when a therapist
engaged in a direct/confront style of therapy. We also found that the combination
of confrontational/educational therapist style without the decisional balance
(condition four) was associated with more resistance to change cognitions than any
76
of the other combinations. From this we gather that a conversation about drinking
with a therapist focused on confronting and directing will result in clients who
think more about their inability, lack of need, lack of desire, reasons not, and
commitment against changing: the exact opposite of the intended effect of an
intervention.
This is significant in the sense that, while a confrontational/educational
approach has often been used in the context of counseling problem drinkers and in
past overtly confrontational approaches like Synanon, these data on change
resistance cognitions suggest that such an approach may generate more thoughts
negatively oriented towards making a change in drinking than one using a more
warm/Rogerian style and/or the decisional balance exercise.
It was not possible to determine whether change resistance cognitions
mediate a relationship between therapist communication style and future behavior
change given that we were not able to detect significant effects for our measures of
drinking behavior at follow-up. Other research, though, has shown this
relationship, e.g., the work of Miller et al. (1993) showing that an empathic, non-
confrontational therapist approach yields less client resistance in-session and that
such an approach predicted better outcomes at one year in a group of problem
drinkers than did a directive-confrontational approach. Our findings are consistent
with that study’s findings about in-session communication. On the other hand, it
may be that generating change resistance cognitions in session does not necessarily
have a negative impact on behavior change.
77
The warm/Rogerian therapist style, though, was also associated with an
increase in readiness to change immediately after the simulated therapy session
while the confrontational/educational therapist style was associated with a decrease
in readiness to change. While this increase associated with the warm/Rogerian
therapist style returned to baseline at follow-up, the decrease in readiness to change
associated with the confrontational/educational style was maintained, i.e., once they
decreased readiness to change, they stayed at that lower level of readiness to
change thirty days later. In other words, at the very least a
confrontational/educational therapist style influences people to be less ready to
change than they were at baseline while a warm/Rogerian style helps to maintain
and possibly increase baseline levels of readiness to change.
Three non-significant trends in our data also illuminated the impact of
therapist style. Participants’ responses to ECS questions 7 and 8 provided evidence
that clients may be more influenced to think about changing their drinking and,
conversely, less influenced to think about not changing their drinking by the
warm/Rogerian therapist style. In addition, the non-significant trend showing that
the warm/Rogerian therapist style was associated with a lower number of
maximum drinks on a single occasion at follow-up shows that there is a possibility
that therapist style can have an impact on actual drinking behavior and is again
consistent with the work of Miller et al. (1993).
While it should be noted that our study featured an analog of a single, brief
session which in reality might be only a part of a longer and more comprehensive
78
intervention, when taken all together and when placed in the context of the existing
literature on therapist behavior, our data show that therapist communication style is
a matter of significant consequence for both in session behavior and for client
behavior outside of the session. It appears that the less a therapist confronts and
educates and the more a therapist expresses empathy, warmth, and genuineness, the
greater the chance that the client will not generate thoughts resisting change and the
greater the chance that the client will begin to consider or take action to start
making a change.
The only finding related to the decisional balance was a statistically non-
significant trend showing that the conditions using the decisional balance were
associated with fewer change resistance cognitions than those without the
decisional balance. This trend was inconsistent with our hypothesis and with our
previous study (Grodin et al., 2004). Based on our earlier findings, we expected
that the conditions featuring the decisional balance (which specifically asks
participants to articulate change resistance and change cognitions) would have
elicited more change resistance cognitions than the general conversation about
drinking featured in the other conditions. This was not the case. This finding is
difficult to interpret. It may indicate that something in the non-decisional balance
scenarios evoked a particularly large amount of change resistance cognitions, but
barring any particular evidence for an explanation and given that this was a
statistically non-significant trend, we will make no further interpretation.
79
As with the previous study, we were able to construct reasonably realistic
and engaging ATSS scenarios according to both the participants and expert
clinicians and we were able to portray different therapist communication styles
effectively. In addition to the fact that we have generated significant findings in
both studies, these data on ATSS show that it is a useful and unique way to
examine psychotherapy process and outcome.
Finally, we identified an interesting potential moderating variable through
our use of the TRS. We found that those high in trait reactance were more likely to
report that, regardless of therapist communication style, the counselor influenced
them to not change their drinking. This finding is consistent with the work of other
clinical researchers who have specifically explored the degree to which trait-like
reactance can be an important active ingredient in psychotherapy. For example,
Beutler, Engle et al. (1991) showed that those with high levels of trait-like
reactance as measured by a sub-scale of the MMPI improved more from a
supportive, self-directed (as opposed to therapist-directed) therapy approach than
those rated low in reactance (who improved more in a group cognitive therapy
approach). This was consistent with their theory that highly reactant clients would
rebel against specific instructions given by a therapist. Shoham-Salomon, Avner,
& Neeman (1989) found that those high in reactance, as measured by tone of voice,
benefited more from paradoxical interventions than those low in reactance. This
finding was consistent with their theory that highly reactant individuals would
benefit from paradoxical interventions because such interventions count on the
80
client to rebel against instructions from the therapist. In rebelling, the client
ignores or acts in opposition to seemingly counterintuitive instructions from the
therapist and thereby engages in more desirable and healthy behavior. While MI
views reactance as more of a state-like variable resulting from poor therapist-client
interaction (Miller & Rollnick, 2002), the findings from the reactance literature and
our data on the TRS suggest that a trait-like conceptualization of reactance deserves
attention from MI researchers and clinicians because it can moderate the influence
that therapy has on different clients.
Limitations
This study was only able to find support for some hypotheses and was only
able to replicate selected results from our previous project. This may be due to
somewhat limited statistical power given that our sample size was slightly smaller
than intended and smaller than the sample in the previous study. The fact that we
were able to generate some significant findings and several trends (including one
trend for the impact of therapist style on drinking behavior), though, indicates that
with a larger sample size, we might have had more power to detect a variety of
effects of therapist style and the decisional balance on our other dependent
variables, like the measures of drinking behavior.
Two other limitations of the study have to do with the facts that the ATSS
paradigm is analog in nature and that our sample was not made up of people who
are actually seeking or participating in psychotherapy for substance abuse. While
the scenarios were pilot-tested, carefully designed, reviewed and evaluated by a
81
group of experienced clinicians, the degree to which the scenarios are externally
valid, i.e., that they are accurate representations of the real-life experience of
talking with a counselor and that they are meaningful for the participant is
somewhat unclear. As in the previous study, though, we were able to obtain
moderately strong ratings of how “realistic and engaging” the ATSS experience
was for both the participants and for a panel of outside clinical experts.
Future Directions
We can make two recommendations to improve the quality and relevance of
an ATSS study of this type in the future: first, we recommend conducting this study
with a clinical population so as to increase the likelihood that the findings are
generalizable to actual therapy clients and, second, we recommend using emerging
technologies such as digital audio or virtual reality in order to possibly enhance the
validity of the ATSS experience. Using a clinical population and new technology
to enhance the imaginal and involving nature of the ATSS paradigm may yield data
that are even more ecologically valid than the current study.
Implications for MI and Conclusions
Our findings have implications for MI from a research, training, and clinical
perspective and for clinical work in general. As we conclude, we reiterate and
highlight a few conclusions suggested by our findings.
This study has shown the importance of therapist communication style for
MI for psychotherapy in general. In addition to increases in readiness to change
and to a lesser extent in decreases on a drinking variable, we were able to see how
82
a warm/Rogerian therapist communication style impacts the way clients think
about change in session. This has implications for the training of clinicians. Given
that different communication styles can have a systematic impact on process and
outcome variables as shown by our data and several other studies cited, the use of a
given communication style should not be left as an arbitrary choice for individual
clinicians. While there are certainly exceptions to every rule, to the extent it is
possible, trainers should focus on cultivating qualities such as empathy, warmth,
and genuineness in trainees aside from instructing them in the use of particular
therapy techniques like, for example, the decisional balance or cognitive
restructuring. Whether it is possible to train someone to improve these qualities is
a topic for further research, but one that is certainly worth investigating.
It is notable also that the increase in readiness to change we found disappeared
rather quickly and the only effect that we found on drinking behavior was a modest
trend on one variable. Therefore, any influence that an intervention using a
warm/Rogerian therapist communication style may have will likely have to be
reinforced by additional sessions or some other kind of intervention or follow-up to
stabilize these gains. Consistent with the recent work of McCambridge and Strang
(2005), which found that the effects of a single-session MI intervention dissipated
over the course of a year, perhaps one session is just not enough for long term
cognitive or behavioral change.
Finally, our finding on trait reactance fills out the picture of client resistance
in MI (and in other therapies for that matter). Client reactance may not be only a
83
state-like result of the therapist-client interaction, as the developers of MI have
argued, but rather a combination of reactance generated by the therapeutic
interaction and a more general tendency on the part of the client to react against
others. For MI practitioners and other clinicians, perhaps including some measure
like the TRS will help clinicians to identify those clients who are more likely to
exhibit reactance so that they can tailor their interventions in some way to account
for this, e.g., by increasing awareness of needing to “roll with resistance” or by
making great efforts to express validation of the client’s point of view. MI
researchers and trainers could benefit from an examination of the literature on
client trait reactance. At this point, the precise nature and implications of client
trait reactance are not yet determined and certainly worthy of further study.
The current study shows that the therapeutic communication style
advocated by MI is an active ingredient with its own effects aside from any
particular technique like the decisional balance. Further efforts are warranted to
understand these effects more precisely and to understand whether these effects are
maintained over time. In addition, questions remain about how in-session
cognitions and behaviors relate to behavior outside the therapy, which specific
techniques work best in combination with the warm/Rogerian communication style,
and how much or how many sessions of MI are necessary to maintain any effects
that are generated. Finally, building an understanding of whether and how trait-like
variables like reactance operate will be important for filling out our picture of the
process and outcome variables important to MI. We are just now beginning to
84
learn about the ways in which MI works and continued effort in this area has great
potential to help researchers and clinicians to understand and improve MI and
psychotherapy in general.
85
References
Agostinelli, G., Brown, J. M., & Miller, W. R. (1995) Effects of normative
feedback on consumption among heavy drinking college students. Journal of
Drug Education, 25 (1), 31-40.
Amrhein, P.C., Miller, W.R., Yahne, C.E., Palmer, M., Fulcher, L. (2003)
Client commitment language during motivational interviewing predicts drug
use outcomes. Journal of Consulting and Clinical Psychology, 71(5), 862-878.
Baer, J. S., Kivlahan, D. K., Blume, A. W., McKnight, P., & Marlatt, G. A.
(2001) Brief intervention for heavy drinking college students: Four-year follow-
up and natural history. American Journal of Public Health, 91(8), 1310-1316.
Baer, J. S., Kivlahan, D. R., & Donovan, D. M. (1999) Integrating skills
training and motivational therapies: Implications for the treatment of substance
dependence. Journal of Substance Abuse Treatment, 17 (1-2), 15-23.
Bandura, A. (1977) Self-efficacy: Toward a unifying theory of behavioral
change. Psychological Review, 84, 191-215.
Bandura, A. (1997) Self-efficacy: The exercise of control. New York: Freeman
Bem, D. J. (1967) Self-perception: An alternative interpretation of cognitive
dissonance phenomena. Psychological Review, 74, 183-200.
Beck, A.T. (1979) Cognitive theory of depression. New York: Guilford Press.
Beck, J.S. (1995) Cognitive therapy: basics and beyond. New York: Guilford
Press
Bien, T. H., Miller, W. R., & Boroughs, J. M. (1993) Motivational interviewing
with alcohol outpatients. Behavioural and Cognitive Psychotherapy, 21, 347-
356.
Bien, T. H., Miller, W. R., & Tonigan, J. S. (1993) Brief interventions for
alcohol problems: A review. Addiction, 88, 315-336.
Borsari, B., & Carey, K. B. (2000) Effects of a brief motivational intervention
with college student drinkers. Journal of Consulting and Clinical Psychology,
68(4), 728-733.
Brehm, S. S., & Brehm, J. W. (1981) Psychological reactance: A theory of
freedom and control. New York: Academic Press
86
Budd, R. J.; Rollnick, S. (1996) The structure of the Readiness to Change
Questionnaire: A test of Prochaska & DiClemente's transtheoretical
model. British Journal of Health Psychology 1 (4), 365-376.
Beutler, L. E., Engle, D., Mohr, D., & Daldrup, R. J. (1991). Predictors of
differential response to cognitive, experiential, and self-directed
psychotherapeutic procedures. Journal of consulting and clinical psychology,
59(2), 333-340
Burke, B., Arkowitz, H., & Menchola, M. (2002). The efficacy of motivational
interviewing: A meta-analysis of controlled clinical trials. In: Miller, W. R.&
Rollnick, S. Motivational interviewing: Preparing people for change (2nd ed.);
p. 217-250. New York: Guilford Press
Burns and Nolen Hoeksema (1992) Therapeutic empathy and recovery from
depression in cognitive-behavioral therapy: A Structural Equation Model.
Journal of Consulting and Clinical Psychology, 60(3), 441-449
Caldwell, P.E. (2002) Drinking levels, related problems and readiness to
change in a college sample. Alcoholism Treatment Quarterly, 20(2), 1-15
Carbonari, J. P., & DiClemente, C. C. (2000) Using transtheoretical model
profiles to differentiate levels of alcohol abstinence success. Journal of
Consulting and Clinical Psychology, 68(5), 810-817.
Carroll, K. M., Ball, S. A., Nich, C., Martino, S., Frankforter, T. L., &
Farentinos, C. et al. (2006). Motivational interviewing to improve treatment
engagement and outcome in individuals seeking treatment for substance abuse:
A multisite effectiveness study. Drug and alcohol dependence, 81(3), 301-312.
CASAA (Center on Alcoholism, Substance Abuse, and Addictions) (1995)
Alcohol Stages of Change Ruler (ASCR) aka Readiness Ruler. Retrieved April
12, 2003 from CASAA website: http://casaa.unm.edu/inst/inst.html
CASAA (Center on Alcoholism, Substance Abuse, and Addictions) (2003)
Motivational Interviewing Skills Code 2.0 (MISC 2.0). Retrieved April 12, 2003
from: CASAA website, http://casaa.unm.edu/tandc.html
Chamberlain, J. & Haaga, D. A. (1999) Convergent validity of cognitive
assessment methods. Behavior Modification, 23(2), 294-315.
Chamberlain, P., Patterson, G., Reid, J., Kavanagh, K., & Forgatch, M. (1984)
Observation of client resistance. Behavior Therapy, 15, 144-155.
87
Cisin, I.H.; Cahalan, D. (1968) Comparison of abstainers and heavy drinkers in
a national survey. Psychiatric Research Reports 24, 10-22.
CORE Institute (2000) 2000 Statistics on Alcohol and Other Drug Use on
American Campuses. Retrieved April 12, 2003 from CORE Institute Website:
http://www.siu.edu/departments/coreinst /public_html/recent.html
Davison, G. C. & Best, J. L. (2004). Think-aloud techniques. In W. O'Donohue,
J. E. Fisher, & S. C. Hayes (Eds.). Empirically-supported techniques of
cognitive behavior therapy: A step-by-step guide for clinicians. New York:
Wiley.
Davison, G. C., Feldman, P. M., & Osborn, C. E. (1984). Articulated thoughts,
irrational beliefs, and fear of negative evaluation. Cognitive Therapy and
Research, 8, 349-362.
Davison, G. C., Navarre, S. G., & Vogel, R. S. (1995). The articulated thoughts
in simulated situations paradigm: A think-aloud approach to cognitive
assessment. Current Directions in Psychological Science, 4, 29-33.
Davison, G. C., Robins, C., & Johnson, M. K. (1983). Articulated thoughts
during simulated situations: A paradigm for studying cognition in emotion and
behavior. Cognitive Therapy and Research, 7, 17-40.
Davison, G. C., Vogel, R. S., & Coffman, S. G. (1997). Think-aloud
approaches to cognitive assessment and the articulated thoughts in simulated
situations paradigm. Journal of Consulting and Clinical Psychology, 65, 950-
958.
Davison, G. C.; Williams, M. E.; Nezami, E. et al. (1991) Relaxation, reduction
in angry articulated thoughts, and improvements in borderline hypertension and
heart rate. Journal of Behavioral Medicine, 14(5), 453-468.
Davison, G. C., & Zighelboim, V. (1987). Irrational beliefs in the articulated
thoughts of college students with social anxiety. Journal of Rational-Emotive
Therapy, 5, 238-254.
DiClemente, C. C., & Hughes, S. O. (1990) Stages of change profiles in
outpatient alcoholism treatment. Journal of Substance Abuse, 2, 217-235.
DiClemente, C. C., & Prochaska, J. O. (1998) Toward a comprehensive,
transtheoretical model of change: Stages of change and addictive behaviors. In:
W. R. Miller, & N. Heather (Eds.) Treating addictive behaviors (2nd ed,), 3-24.
New York: Plenum Press
88
Draycott, Simon; Dabbs, Alan (1998) Cognitive dissonance 2: A theoretical
grounding of motivational interviewing. British Journal of Clinical Psychology,
37(3), 355-364.
Dunn, C., DeRoo, L., & Rivara, F. P. (2001) The use of brief interventions
adapted from motivational interviewing across behavioral domains: A
systematic review. Addiction, 96(12), 1725-1742.
Dunn, E. C., Neighbors, C., & Larimer, M. E. (2006). Motivational
enhancement therapy and self-help treatment for binge eaters. Psychology of
Addictive Behaviors, 20(1), 44-52.
Eckhardt, C. I., Barbour, K. A., & Davison, G. C. (1998). Articulated thoughts
of maritally violent and nonviolent men during anger arousal. Journal of
Consulting and Clinical Psychology, 66, 259-269.
Ellis, A., and Dryden, W. (1997) The Practice of Rational Emotive Behavior
Therapy (2
nd
ed.). New York: US Springer Publishing Co.
Festinger, L. (1957) A Theory of Cognitive Dissonance. Stanford, CA: Stanford
University Press
Greenberg, L.S, Watson, J.C., Elliot, R., & Bohart, A.C. (2001). Empathy.
Psychotherapy: Theory, Research, Practice, Training, 38(4), 380-384
Grodin, J., Davison, G., Earleywine, M. (2004) Utilizing the Articulated
Thoughts in Simulated Situations Paradigm to Understand the Mechanisms of
Action for Motivational Interviewing. Unpublished master’s thesis.
Guttman, L. (1973). Premium and protection of several procedures for dealing
with outliers when sample sizes are moderate or large. Technometrics, 15, 385-
404
Haaga, D. A. (1987). Smoking schemata revealed in articulated thoughts
predicts early relapse from smoking cessation. Paper presented at the 21
st
Annual Convention of the Association for Advancement of Behavior Therapy,
Boston.
Haaga, D. A. F., Davison, G. C., Williams, M. E., Dolezal, S. L., Haleblian, J.,
et. al. (1994). Mode-specific impact of relaxation training for hypertensive men
with type A behavior pattern. Behavior Therapy, 25, 209-223.
Heather, N., Rollnick, S., & Bell, A. (1993) Predictive validity of the Readiness
to Change Questionnaire. Addiction, 88, 1667-1677.
89
Heather, N., Rollnick, S., Bell, A., & Richmond, R. (1996) Effects of brief
counseling among male heavy drinkers identified on general hospital wards.
Drug and Alcohol Review, 15(1), 29-38.
Heyman, R.E. & Slep, A.M. (2003) Analogue behavioral observation. In:
Haynes, S. N. and Heiby, E. (Eds.). Behavioral Assessment. New York:
Wiley
Janis, I. L., & Mann, L. (1977) Decision-making: A psychological analysis of
conflict, choice, and commitment. New York: Free Press .
Johnston, L. D., O'Malley, P. M., & Bachman, J. G. (2000). Monitoring the
future national results on adolescent drug use: 1975-1999. Vol. II: College
students and young adults. Bethesda, MD: National Institute on Drug Abuse.
Jones, R. G. (1968). A factored measure of Ellis’s irrational belief system with
personality and maladjustment correlates. Unpublished doctoral dissertation,
Texas Technological College.
Kavanagh, K., Gabrielson, P., & Chamberlain, P. (1982) Manual for coding
client resistance. Unpublished Instrument, 207 E. 5
th
Avenue, OR 97401
LaBrie, J. W., Pedersen, E. R., Earleywine, M., & Olsen, H. (2006). Reducing
heavy drinking in college males with the decisional balance: Analyzing an
element of motivational interviewing. Addictive Behaviors, 31(2), 254-263.
Luborksy, L., McLellan, A.T., Woody, G.E., O’Brien, C.P., & Auerbach, A.
(1985) Therapist success and its determinants. Archives of General Psychiatry,
42, 602-611
Markland, D., Ryan, R. M., Tobin, V. J., & Rollnick, S. (2005). Motivational
interviewing and self-determination theory. Journal of Social & Clinical
Psychology, 24(6), 811-831.
Marlatt, G. A., Baer, J. S., Kivlahan, D. R., Dimeff, L. A., Larimer, M. E.,
Quigley, L. A., Somers, J. M., & Williams, E. (1998) Screening and brief
intervention for high-risk college student drinkers: Results from a two-year
follow-up assessment. Journal of Consulting and Clinical Psychology, 66(4),
604-615.
Masterman, P., Kelly, A., (2003) Reaching adolescents who drink harmfully:
Fitting intervention to developmental reality. Journal of Substance Abuse
Treatment, 24, 347-355.
90
McCambridge, J., & Strang, J. (2005). Deterioration over time in effect of
motivational interviewing in reducing drug consumption and related risk among
young people. Addiction, 100(4), 470-478.
McCambridge, J., & Strang, J. (2004) The efficacy of single-session
motivational interviewing in reducing drug consumption and perceptions of
drug-related risk and harm among young people: results from a multi-site
cluster randomized trial. Addiction, 99, 39-52.
Migneault, J. P., Pallonen, U.E., Velicer, W. F.( 1997) Decisional balance and
stage of change for adolescent drinking. Addictive Behaviors, 22(3), 339-351.
Migneault, J. P., Velicer, W. F., Prochaska, J. O. et al. (1999) Decisional
balance for immoderate drinking in college students. Substance Use & Misuse,
34(10), 1325-1346.
Miller, W.R (1983) Motivational interviewing with problem drinkers.
Behavioural Psychotherapy, 11, 147-172.
Miller, W. R. (1994). Motivational Interviewing: III. On the ethics of
motivational intervention. Behavioral and Cognitive Psychotherapy, 22, 111-
123.
Miller, W.R. (2001) The use of brief interventions adapted from motivational
interviewing across behavioral domains: A systematic review: Reply.
Addiction, 96(12), 1774-1775
Miller, W.R. (2002). Motivational Interviewing: Findings from clinical trials.
Motivational Interview website, http://motivationalinterview.org/library/
outcomes_files/frames.htm
Miller, W.R. (n.d.). Values Card Sort. Motivational Interview Website,
http://motivationalinterview.org/library/valuescardsort.pdf
Miller, W. R. & Rollnick, S. (1991) Motivational interviewing: Preparing
people to change addictive behavior. New York: Guilford Press
Miller, W. R. & Rollnick, S. (2002) Motivational interviewing: Preparing
people for change (2nd ed.).New York: Guilford Press
Miller, W. R., Benefield, R. G., & Tonigan, J. S. (1993) Enhancing motivation
for change in problem drinking: A controlled comparison of two therapist
styles. Journal of Consulting and Clinical Psychology, 61(3), 455-461.
91
Miller, W.R., Taylor, C.A., & West, J.C. (1980). Focused versus broad
spectrum behavior therapy for problem drinkers. Journal of Consulting and
Clinical Psychology, 48, 590-601.
Miller, W.R., Yahne, C., Tonigan, J.S. (2003) Motivational Interviewing in
Drug Abuse Services: A Randomized Trial. Journal of Consulting and Clinical
Psychology, 71(4), 754-763.
Moyers, T. B., Martin, T., Manuel, J. K., Hendrickson, S. M. L., & Miller, W.
R. (2005). Assessing competence in the use of motivational interviewing.
Journal of substance abuse treatment, 28(1), 19-26.
Mullins, S., Suarez, M., Ondersma, S., Page, M. (2004) The impact of
motivational interviewing on substance abuse treatment retention: A
randomized control trial of women involved with child welfare. Journal of
Substance Abuse Treatment, 27, 51-58.
Murphy, J., Benson, T., Vuchinich, R. et al. (2004) A comparison of
personalized feedback for college student drinkers delivered with and without a
motivational interview. Journal of Studies on Alcohol, 65, 200-203
National Institute on Alcohol Abuse and Alcoholism, U.S. Department of
Health and Human Services. (1995). Motivational enhancement therapy
manual: A clinical research guide for therapists treating individuals with
alcohol abuse and dependence (Project MATCH Monograph Series, NIH
Publication No. 94-3723). Rockville, MD
Patterson, G. R., & Forgatch, M. S. (1985) Therapist behavior as a determinant
for client noncompliance: A paradox for the behavior modifier. Journal of
Consulting and Clinical Psychology, 53, 846-851.
Perkins, H. W., Meilman, P. W., Leichliter, J. S. et al. (1999) Misperceptions of
the norms for the frequency of alcohol and other drug use on college
campuses. Journal of American College Health, 47(6), 253-258.
Plummer, B. A., Velicer, W. F., Redding, C. A. et al. (2001) Stage of change,
decisional balance and temptations for smoking. Measurement and validation in
a large, school-based population of adolescents. Addictive Behaviors, 26(4),
551-571.
Prochaska, James O.; DiClemente, Carlo C. (1986). Toward a comprehensive
model of change. In: Miller, William R.; Ed; Heather, Nick; Ed; Treating
addictive behaviors: Processes of change, 3-27; New York: Plenum Press
92
Prochaska, J. O., Velicer, W. F., Rossi, J. S., Goldstein, M. G., Marcus, B. H.,
Rakowski, W., Fiore, C., Harlow, L. L., Redding, C. A., Rosenbloom, D., &
Rossi, S. R. (1994) Stages of change and decisional balance for twelve problem
behaviors. Health Psychology, 13(1), 39-46.
Project MATCH Research Group (1997) Matching alcoholism treatments to
client heterogeneity: Project MATCH Post-treatment drinking
outcomes. Journal of Studies on Alcohol, 58(1), 7-29.
Rayburn, N.R., & Davison, G. C. (2002). Articulated thoughts about anti-gay
hate crimes. Cognitive Therapy and Research, 26(4), 431-447.
Rogers, C. R. (1951) Client-centered therapy. Boston: Houghton-Mifflin
Rogers, C. R. (1961) On becoming a person. Boston: Houghton Mifflin
Rokeach, M. (1973) The Nature of Human Values. New York: Free Press
Rokeach, M. (Ed.). (1979) Understanding Human Values. New York:
Macmillan
Rollnick, S. (2001) The use of brief interventions adapted from motivational
interviewing across behavioral domains: A systematic review: Comment.
Addiction, 96(12), 1769-1770
Rollnick, S., Heather, N., & Bell, A. (1992) Negotiating behaviour change in
medical settings: The development of brief motivational interviewing. Journal
of Mental Health, 1, 25-39.
Rollnick, S., Heather, N., Gold, R., & Hall, W. (1992) Development of a short
‘readiness to change’ questionnaire for use in brief opportunistic intervention
among excessive drinkers. British Journal of Addiction, 87, 743-754
Rollnick, S., & Miller, W. R. (1995) What is motivational interviewing?
Behavioural and Cognitive Psychotherapy, 23(4), 325-334.
Rossi, S. R.; Greene, G. W.; Rossi, J. S. et al. Validation of decisional balance
and situational temptations measures for dietary fat reduction in a large school-
based population of adolescents. Eating Behaviors, 2(1), 1-18.
Sanchez, F. (2000). A values-based intervention for alcohol problems. Doctoral
dissertation, University of New Mexico, Albuquerque.
93
Share, D., McCrady, B., Epstein, E. (2004) Stage of change and decisional
balance for women seeking alcohol treatment. Addictive Behaviors, 29, 525-
535.
Shoham-Soloman, V., Avner, R., & Neeman, R. (1989). You're changed if
you do and changed if you don't: Mechanisms underlying paradoxical
interventions. Journal of Consulting and Clinical Psychology, 57, 590-598.
Sobell, L.C.; Maisto, S.A.; Sobell, M.B.; and Cooper, A.M. (1979) Reliability
of alcohol abusers' self-reports of drinking behavior. Behaviour Research and
Therapy, 17, 157-160.
Sobell, L.C., & Sobell, M.B., (1992). Timeline followback: A technique for
assessing self-reported alcohol consumption. In R.Z. Litten & J. Allen (Eds.),
Measuring alcohol consumption: Psychosocial and biological methods (41-72).
New Jersey: Humana Press.
Sobell, L.C., and Sobell, M.B. (1995) Alcohol Timeline Followback Users'
Manual. Toronto, Canada: Addiction Research Foundation.
Stein, L. A. R., Colby, S. M., Barnett, N. P., Monti, P. M., Golembeske, C., &
Lebeau-Craven, R. et al. (2006). Enhancing substance abuse treatment
engagement in incarcerated adolescents. Psychological Services, 3(1), 25-34.
Strang, J. & McCambridge, J. (2004) Can the practitioner correctly predict
outcome in motivational interviewing? Journal of Substance Abuse Treatment,
27, 83-88.
Tait, R. & Hulse, G. (2003) A systematic review of the effectiveness of brief
interventions with substance using adolescents by type of drug. Drug and
Alcohol Review, 22, 337-346.
Vik, P.W., Culbertson, K.A., Sellers, K. (2000). Readiness to change drinking
among heavy-drinking college students. Journal of Studies on Alcohol, 61, 674-
680.
Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety.
Journal of Consulting and Clinical Psychology, 33, 87-104.
Wechsler, H., Davenport, A., Dowdall, G. et al. (1994) Health and behavioral
consequences of binge drinking in college: A national survey of students at 140
campuses. JAMA: Journal of the American Medical Association, 272(21),
1672-1677.
94
Wechsler, H., Lee, J.E., Kuo, M., and Lee, H. (2000) College Binge Drinking in
the 1990s: A Continuing Problem. Journal of American College Health, 48,
199-210.
Weinstein, K. A., Davison, G. C., DeQuattro, V., & Allen, J. W. (1986). Type A
behavior and cognitions: Is hostility the bad actor? Paper presented at the 94
th
Annual Convention of the American Psychological Association. Washington,
D.C.
White, J., Davison, G. C., Haaga, D. A. F., & White, K. (1992). Cognitive bias
in the articulated thoughts of depressed and nondepressed psychiatric patients.
Journal of Nervous and Mental Disease, 180, 77-81
Wilcox, R. (2003) Applying Contemporary Statistical Techniques. USA:
Elsevier Science.
Williams Marian E., Davison Gerald C., Nezami Elahe, & DeQuattro Vincent
L. (1992). Articulated thoughts of Type A and B individuals in response to
social criticism. Cognitive Therapy & Research, 16(1), 19-30
95
Appendices
Appendix A
Frequency/Quantity Screening and Follow-Up Measure
All of these questions ask about your drinking during the PAST MONTH (30
days). For each question please write the correct number on the space provided.
How many times per week do you drink alcoholic beverages (0-7)? _____
How many times per month do you drink alcoholic beverages (0-7)? _____
For the following questions, one standard drink of alcohol equals one beer (12 oz),
one mixed drink or 1 shot (1.2 oz) of liquor or 1 glass of wine (5.0 oz)
Again, in the LAST 30 DAYS:
How many standard drinks do you usually consume when you have alcohol? ____
What’s the LARGEST number of standard drinks you’ve consumed on a single
occasion? _____
How many times per month do you drink more than 4 drinks on a single occasion?
_____
How many times per month do you drink more than 5 drinks on a single occasion?
_____
Contact authorization
I (circle one) (give/do not give) my permission to be contacted regarding my potential
participation in a study by e-mail and/or by phone in order to schedule a particular
experiment at a convenient time. The purpose of obtaining the student ID number is to verify
your name, e-mail address and phone number. No other information will be obtained from
the university records and your student ID number will not be released to people outside the
research team.
_______________________ __________________ __________________
(name – please print) (e-mail address) (phone number)
________________________ (student ID number)
96
NOTE: Participant Contact Rules (will not appear on Frequency/Quantity
questionnaire given to potential participants):
Contact limits. Participants will be contacted no more than once by phone and
once by e-mail. If they decline to respond or participate, no further contact will be
made.
Contact script. “You qualify for a special study the Psychology Department is
conducting right now and we are calling to see if you are interested in participating.
If you are interested we can help you schedule your participation at your
convenience. Your participation is completely at your discretion and you have no
obligation to participate.”
97
Appendix B
Alcohol Timeline Followback Instructions and Example Calendar
To help us evaluate your drinking, we need to get an idea of what your alcohol use was like
in the past ____ days. To do this, we would like you to fill out the attached calendar.
Filling out the calendar is not hard!
Try to be as accurate as possible.

We recognize you won’t have perfect recall. That’s OKAY.

WHAT TO FILL IN
• The idea is to put a number in for each day on the calendar.
• On days when you did not drink, you should write a ”0”.
• On days when you did drink, you should write in the total number of drinks you had.
• We want you to record your drinking on the calendar using Standard Drinks. For example, if
you had 6 beers, write the number 6 for that day. If you drank two or more different kinds of
alcoholic beverage in a day such as 2 beers and 3 glasses of wine, you would write the
number 5 for that day.
It’s important that something is written for every day, even if it is a
“0”.

YOUR BEST ESTIMATE
• We realize it isn’t easy to recall things with 100% accuracy.
• If you are not sure whether you drank 7 or 11 drinks or whether you drank on a Thursday or a
Friday, give it your best guess! What is important is that 7 or 11 drinks is very different from
1 or 2 drinks or 25 drinks. The goal is to get a sense of how frequently you drank, how much
you drank, and your patterns of drinking.

HELPFUL HINTS
• If you have an appointment book you can use it to help you recall your drinking.
• Holidays such as Thanksgiving and Christmas are marked on the calendar to help you
better recall your drinking. Also, think about how much you drank on personal holidays &
events such as birthdays, vacations, or parties.
• If you have regular drinking patterns you can use these to help you recall your drinking.
98
For example, you may have a daily or weekend/weekday pattern, or drink more in the
summer or on trips, or you may drink on Wednesdays after playing sports.

COMPLETING THE CALENDAR
• A blank calendar is attached. Write in the number of Standard Drinks that you had each
day
• The time period we are talking about on the calendar is
from ________________________ to _______________________
• In estimating your drinking, be as accurate as possible.
• DOUBLE CHECK THAT ALL DAYS ARE FILLED IN BEFORE RETURNING THE CALENDAR.
• Before you start, look at the SAMPLE CALENDAR AND STANDARD DRINK CHART on the
next page.
Instructions for Filling Out the Timeline Alcohol Use Calendar

SAMPLE CALENDAR
2000 SUN MON TUES WED THURS FRI SAT
1
8
2
0
S
3
7
4
Labor Day
0
5
3
6
8
7
1
8
0
9
11
E
10
2
11
2
12
0
13
3
14
5
15
14
16
4
P
17
2
18
0
19
0
20
0
21
0
22
2
23
13
T
24
0
25
0
26
6
27
0
28
0
29
0
30
2
U. S. STANDARD DRINK CONVERSION CHART
One Standard Drink Is Equal To
12 oz of BEER (5%)
5 oz of WINE (10% – 12%)
3 oz of FORTIFIED WINE (16% – 18%)
99
1 1.2 oz of HARD LIQUOR (86 proof – 100 proof; 43% – 50%)
WINE: 1 Bottle
25 oz/750 ml = 5 standard drinks
40 oz/1.5 liter = 8 standard drinks
25 oz fortified = 8 1/3 standard drinks
HARD LIQUOR: 1 Bottle
12 oz (mickey) = 8 standard drinks
26 oz = 17 1/3 standard drinks
40 oz = 26 2/3 standard drinks
100
Appendix C
Readiness to Change Questionnaire
The following questionnaire is designed to identify how you personally feel about
your drinking right now. Please read each of the questions below carefully, and
then decide whether you agree or disagree with the statements. Please circle the
answer of your choice to each question. Your answers are completely private and
confidential.
Strongly Disagree Unsure Agree Strongly
Disagree Agree
1. I don’t think I drink too much 1 2 3 4 5
2. I am trying to drink less than I used to 1 2 3 4 5
3. I enjoy my drinking but sometimes I drink too much 1 2 3 4 5
4. Sometimes I think I should cut down on my drinking 1 2 3 4 5
5. It’s a waste of time thinking about my drinking 1 2 3 4 5
6. I have just recently changed my drinking habits 1 2 3 4 5
7. Anyone can talk about wanting to do something
about drinking, but I’m actually doing
something about it 1 2 3 4 5
8. I am at a stage when I should think about
drinking less alcohol 1 2 3 4 5
9. My drinking is a problem sometimes 1 2 3 4 5
10. There is no need for me to think about changing
my drinking 1 2 3 4 5
11. I am actually changing my drinking habits right now 1 2 3 4 5
12. Drinking less alcohol would be pointless for me 1 2 3 4 5
101
Appendix D
Alcohol Stages of Change Ruler
On the Ruler Below, please circle the number that best describes how you feel right
now:
0------1---------2---------3---------4---------5---------6---------7---------8-------9------10
Never think Sometimes I I have I am already My drinking
About my think about decided trying to has changed.
Drinking drinking less to drink less cut back on I now drink
my drinking less than
before
102
Appendix E
F/Q Intention
All of these questions ask about how much alcohol you INTEND TO DRINK
OVER THE NEXT MONTH (30 days). For each question please write the
correct number on the space provided.
For the following questions, one standard drink of alcohol equals 1 beer, 1
mixed drink or 1 shot (1.5 oz) of liquor or 1 glass of wine (4.0 oz)
1. How many days do you intend to drink alcoholic beverages (0-30)?
_____
2. How many standard drinks do you intend to consume on average
when you drink alcohol?
_____
3. What’s the LARGEST number of standard drinks you intend to
consume on a single occasion?
_____
4. How many times do you intend to drink more than 5 drinks on
a single occasion?
_____
Subject # _______
103
Appendix F
Overall Impression of Drinking Alcohol
For this exercise, we’d like to get an idea of your overall opinion of the role of
alcohol in your life. Like the other questionnaires in this packet, your answers are
confidential and there is no right or wrong answer….
We want to see how the advantages (the positive consequences it causes you
physically, mentally, professionally, personally and with other people) compare to
the disadvantages (the negative consequences it causes you physically, mentally,
professionally, personally and with other people).
Out of 100 percentage points, please assign a number to the sum total of the
advantages (or “pros”) and the sum total of the disadvantages (or “cons”)…here are
three examples to clarify what we mean:
1. For you personally, if the advantages of drinking very much outweigh
the disadvantages of drinking, you might indicate that the balance is 95
pro drinking and 5 percent con drinking.
2. For you personally, if the advantages and disadvantages are about the
same, you might indicate that it is 50/50.
3. For you personally, if drinking is mostly but not totally disadvantageous
for you, you might indicate 25 pro drinking and 75 con drinking.
Note that all of these add up to 100 total points and you can enter in any
combination of two numbers you like in order to indicate your impression
of alcohol drinking in your life…as long as they add up to a total of 100.
Please use the above only as examples and make sure to indicate your own
answer with your own numbers in the space provided below.
What is the balance of the sum total of the advantages/pros and sum
total of the disadvantages/cons of drinking alcohol for you right now at
this very moment:
Overall Advantages/Pros of Drinking: ______
Overall Disadvantages/Cons of Drinking: ______
Total Must Add Up To: 100 %
104
Appendix G
Overall Impression of Quitting Alcohol
For this exercise, we’d like to get an idea of your overall opinion of the issue of
QUITTING alcohol consumption in your life. Like the other questionnaires in this
packet, your answers are confidential and there is no right or wrong answer….
We want to see how the advantages of quitting (the positive consequences it would
cause for you physically, mentally, professionally, personally and with other
people) compare to the disadvantages of quitting (the negative consequences it
would cause you physically, mentally, professionally, personally and with other
people). If you no longer consume alcohol, please indicate the pros and cons
quitting has caused for you.
Out of 100 percentage points, please assign a number to the sum total of the
advantages (or “pros”) of quitting and the sum total of the disadvantages (or
“cons”) of quitting…here are three examples to clarify what we mean:
1. For you personally, if the advantages of quitting drinking very much
outweigh the disadvantages of quitting drinking, you might indicate that the
balance is 95 pro drinking and 5 percent con drinking.
2. For you personally, if the advantages of quitting and disadvantages of
quitting are about the same, you might indicate that it is 50/50.
3. For you personally, if quitting drinking is mostly but not totally
disadvantageous for you, you might indicate 25 pro drinking and 75 con
drinking.
Note that all of these add up to 100 total points and you can enter in any
combination of two numbers you like in order to indicate your impression
of quitting alcohol drinking…as long as they add up to a total of 100.
Please use the above only as examples and make sure to indicate your own
answer with your own numbers in the space provided below.
What is the balance of the sum total of the advantages/pros and sum
total of the disadvantages/cons of quitting drinking alcohol for you
right now at this very moment:
Overall Advantages/Pros of Quitting Drinking: ______
Overall Disadvantages/Cons of Quitting Drinking: ______
Total Must Add Up To: 100 %
105
Appendix H
Experimental Check for Subjects
1=Not at all
2=Somewhat
3=Moderately so
4=Very much so
To what degree did you find the scenario realistic? 1 2 3 4
To what degree did you find the scenario vivid? 1 2 3 4
To what degree were you able to imagine
yourself really in the scenario? 1 2 3 4
To what degree did you feel the therapist in the scenario
had empathy for you? 1 2 3 4
To what degree did you feel the therapist in the scenario
was warm? 1 2 3 4
To what degree did you feel the therapist in the scenario
was being genuine? 1 2 3 4
To what degree did you feel the experience with
the counselor influenced you to think about
changing your drinking? 1 2 3 4
To what degree did you feel the experience with
the counselor influenced you to think about
NOT changing your drinking? 1 2 3 4
106
Appendix I
Demographics Questionnaire
Directions: This form requests personal information. The information you provide
is completely confidential and will not be used to personally identify you.
Date: _________________
1. Subject Pool Participant #:____________
2. Age: ____
3.
4. Gender: Male ____ Female ____
2. Race/Ethnicity: African-American/Black ____
Asian-American/Pacific Islander ____
Caucasian/White ____
Latino/Latina ____
Middle Eastern ____
Native American ____
Other (please specify) _________________
107
Appendix J
Therapeutic Reactance Scale (TRS)
Please circle the response that best describes you.
Strongly Disagree Agree Strongly
Disagree Agree
1. If I receive a lukewarm dish at a restaurant, I make an
attempt to let that be known. 1 2 3 4
2. I resent authority figures who try to tell me what to do. 1 2 3 4
3. I find that I often have to question authority. 1 2 3 4
4. I enjoy seeing someone else do something that neither of us is supposed to do. 1 2 3 4
5. I have a strong desire to maintain my personal freedom. 1 2 3 4
6. I enjoy playing “devil’s advocate” whenever I can. 1 2 3 4
7. In discussions, I am easily persuaded by others. 1 2 3 4
8. Nothing turns me on as much as a good argument! 1 2 3 4
9. It would be better to have more freedom to do what I want on a job. 1 2 3 4
10. If I am told what to do, I often do the opposite. 1 2 3 4
11. I am sometimes afraid to disagree with others. 1 2 3 4
12. It really bothers me when police officers tell people what to do. 1 2 3 4
13. It does not upset me to change my plans because someone in
the group wants to do something else. 1 2 3 4
14. I don’t mind other people telling me what to do. 1 2 3 4
15. I enjoy debates with other people. 1 2 3 4
16. If someone asks a favor of me, I will think twice about
what this person is really after. 1 2 3 4
17. I am not very tolerant of others’ attempts to persuade me. 1 2 3 4
18. I often follow the suggestions of others. 1 2 3 4
19. I am relatively opinionated. 1 2 3 4
20. It is important to me to be in a powerful position relative to others. 1 2 3 4
21. I am very open to solutions to my problems from others. 1 2 3 4
22. I enjoy “showing up” people who think they are right. 1 2 3 4
23. I consider myself more competitive than cooperative. 1 2 3 4
24. I don’t mind doing something for someone else even when
I don’t know why I’m doing it. 1 2 3 4
25. I usually go along with others’ advice. 1 2 3 4
26. I feel it is better to stand up for what I believe than to be silent. 1 2 3 4
27. I am very stubborn and set in my ways. 1 2 3 4
28. It is very important for me to get along well with the people I work with. 1 2 3 4
108
Appendix K
Experimental Check-Experts on the scenario
1=Not at all
2=Somewhat
3=Moderately so
4=Very much so
To what degree did you find the scenario realistic? 1 2 3 4
To what degree did you find the scenario vivid? 1 2 3 4
To what degree did you feel the therapist in the scenario
was empathic? 1 2 3 4
To what degree did you feel the therapist in the scenario
was warm? 1 2 3 4
To what degree did you feel the therapist in the scenario
was being genuine? 1 2 3 4
109
Appendix L
Condition One ATSS Script
(Warm/Rogerian with DB: MI)
Segment One:
You have agreed for class credit to share some information with a counselor about
your drinking habits. The counselor, whom you immediately feel comfortable
with, has assured you of the confidentiality of your conversation. He seems
genuinely curious-he’s patient and doesn’t seem to be judgmental.
Segment Two:
The counselor greets you as you sit down:
T: Good morning. Thank you for coming in-I’m glad you were able to make it
today. Today I was hoping to gather some information from you on your thoughts
about drinking. This is not a test, but more of a chance to learn some things about
you and drinking alcohol. Maybe we could start by you telling me what are some
of the reasons why you like drinking? Just feel free to mention anything
whatsoever that you like about drinking-social reasons, physical reasons, anything
at all.
Segment Three:
The counselor seems very patient, just gathering the information and more
concerned that you are comfortable than with reacting positively or negatively to
your thoughts. While he pauses to write these down you have a moment to reflect
on what this is like for you. You may think about drinking in this way a lot or this
may be the first time you have thought about drinking in this way.
Segment Four:
T: Ok, next what I’d like to hear about are some of the not-so-good things about
drinking. Again, feel free to mention whatever you think of.
Segment Five:
As you list these things, the counselor listens closely. You and he continue to talk-
sometimes he repeats back to you the things you are saying, sometimes asks you to
elaborate, other times nods, all the while being fairly laid-back and showing that he
is listening and getting what you are saying. You gather your thoughts about what
it is like talking with this particular person about this subject matter.
110
Segment Six:
The counselor continues:
T: I understand how everyone has different thoughts and feelings about drinking
and I appreciate you sharing this information with me. I’m curious what you think
might be good if you changed your drinking (however much you do) in some way-
maybe decreased it. I’m not telling you to do this, but rather I’m wondering if you
were to decide to do this, what you think might be positive about doing so?
Segment Seven:
As you and he continue to chat, you notice that he continues to listen and repeat
back some of the things you say, showing that he “gets” that there are good and bad
things about drinking. It may or may not be surprising that the counselor has
brought the idea of stopping drinking. You think about what he brings up and also
about what you think of him in general.
Segment Eight:
T: Ok, so now we’ll move on if it is alright with you to talk about what you would
not like about changing your drinking. Anything about what you would miss, or
what it would be like to change that comes to mind?
Segment Nine:
After carefully noting the things you have said, the counselor sets his writing
material down and congratulates you for the work you have just done. He reassures
you that the information is for your use and that what you choose to think about it
or do in response to this visit is entirely up to you. He knows that it can be hard for
some people to think about things in this way, but that it can be a helpful way to
gather ones thoughts and feelings. As he leads you to the exit, you reflect on
whatever thoughts and feelings you just experienced about the therapist, talking
about drinking, what you might do, if anything, and so on.
111
Appendix M
Condition Two ATSS Script
(Warm/Rogerian without Decisional Balance)
Segment One:
You have agreed for class credit to share some information with a counselor about
your drinking habits. The counselor, whom you immediately feel comfortable
with, has assured you of the confidentiality of your conversation. He seems
genuinely curious-he’s patient and doesn’t seem to be judgmental.
Segment Two:
The counselor greets you as you sit down:
T: Good morning. Thank you for coming in- I’m glad you were able to make it
today. Today I was hoping to gather some information from you on your thoughts
about drinking. This is not a test, but more of a chance to learn some things about
you and drinking alcohol. Maybe we could start by you telling me some of your
thoughts about when you started drinking.
Segment Three:
The counselor seems very patient, just gathering the information and more
concerned that you are comfortable than with reacting positively or negatively to
your thoughts. While he pauses to write these down you have a moment to reflect
on what this is like for you and what you think about the counselor. Also, you may
think about drinking a lot or this may be the first time you have thought about
drinking.
Segment Four:
T: Ok, I’m curious to know if you have ever thought about if you drink a normal
amount, less than other people, or more than other people.
Segment Five:
As you respond, the counselor listens closely. You and he continue to talk-
sometimes he repeats back to you the things you are saying, sometimes asks you to
elaborate, other times nods, all the while being fairly laid-back and showing that he
is listening and getting what you are saying. You gather your thoughts about what
it is like talking with this particular person about this subject matter.
112
Segment Six:
The counselor continues:
T: I understand how everyone has different thoughts and feelings about drinking
and I appreciate you sharing this information with me. Have you noticed anything
about your drinking-has it changed over the years, or even recently?
Segment Seven:
As you and he continue to chat, you notice that he continues to listen and repeat
back some of the things you say, showing that he “gets” that everyone has different
thoughts about their drinking. You think about what he brings up and also about
what you think of him in general.
Segment Eight:
T: Ok, so now we’ll move on if it is alright with you to talk about if you have
thought about changing anything about your drinking? Like the amount you drink,
or what you drink, or if you feel you would be able to make a change if you
decided to. As I’ve mentioned, people think about these topics very differently, so
whatever comes to mind, if anything, would be interesting to hear.
Segment Nine:
After carefully noting the things you have said, the counselor sets his writing
material down and congratulates you for the work you have just done. He reassures
you that the information is for your use and that what you choose to think about it
or do in response to this visit is entirely up to you. He knows that it can be hard for
some people to think about things in this way, but that it can be a helpful way to
gather ones thoughts and feelings. As he leads you to the exit, you reflect on
whatever thoughts and feelings you just experienced about the therapist, talking
about drinking, what you might do, if anything, and so on.
113
Appendix N
Condition Three ATSS Script
(Confrontational/Educational with Decisional Balance)
Segment One:
You have agreed for class credit to share some information with a counselor about
your drinking habits. The counselor has assured you of the confidentiality of your
conversation. He has looked over a form you filled out that details how much you
drink and when you drink.
Segment Two:
The counselor greets you as you sit down:
T: Good morning. Thank you for coming in- I’m glad you were able to make it
today. Today I was hoping to gather some information from you on your thoughts
about drinking and to tell you about some of the problems with drinking a lot of
alcohol. On your response to the questionnaire you indicated that you drink an
amount of alcohol that would qualify you as a binge drinker. Why do you drink so
much?
Segment Three:
The counselor takes a pause while he writes some things down so that he can
remember them later. The counselor seems to have reacted to everything you say
with great energy. He appears to be judging your responses and formulating his
own ideas. You may think about drinking a lot or this may be the first time you
have thought about it. As he writes you have a moment to reflect on what this is
like for you.
Segment Four:
Therapist: A lot of people say good things about drinking, despite the fact that the
amount of drinking-which you told me about in the paperwork we gathered just
before-is really just harmful and bad in the long run. I’m sure you see the
disadvantages in drinking-what are they?
Segment Five:
As you list these things and he writes them down, you are given another moment to
gather your thoughts about what it is like talking with this particular person about
this subject matter.
114
Segment Six:
The counselor continues:
Ok. I told you a bit before about how changing alcohol drinking is the goal. What
you really need to do is to not drink at all. Even after you have stopped you can’t
really go back and even have one drink. I know you aren’t thinking about this much
right now even though you need to. Imagine that you did stop-what would be good
about it?
Segment Seven:
Again, the counselor seems to be interested in communicating his agenda to you
more than anything else. It may or may not be surprising that the counselor has
brought up the idea of stopping drinking. You think about what he brings up and
also about what you think of him in general.
Segment Eight:
T: Ok, so if you did stop drinking completely, what do you think would be bad
about it? In other words, despite the advantages of not drinking and avoiding all
the risks associated, what would be the disadvantages if you stopped drinking?
Segment Nine:
After carefully noting the things you have said, the counselor sets his writing
material down and turns directly towards you. He repeats his statements earlier
about the unquestionable problems with drinking alcohol at all. He wishes you
luck and tells you that he hopes you strongly consider these issues. As he leads you
to the exit, you reflect on whatever thoughts and feelings you just experienced
about the therapist, talking about drinking, what you might do, if anything, and so
on.
115
Appendix O
Condition Four ATSS Script
(CE without Decisional Balance)
Segment One:
You have agreed for class credit to share some information with a counselor about
your drinking habits. The counselor has assured you of the confidentiality of your
conversation. He has looked over a form you filled out that details how much you
drink and when you drink.
Segment Two:
The counselor greets you as you sit down:
T: Good morning. Thank you for coming in- I’m glad you were able to make it
today. Today I was hoping to gather some information from you on your thoughts
about drinking and to tell you about some of the problems with drinking a lot of
alcohol. On your response to the questionnaire you indicated that you drink an
amount of alcohol that would qualify you as a binge drinker. When did you start
drinking?
Segment Three:
The counselor takes a pause while he writes some things down so that he can
remember them later. The counselor seems to have reacted to everything you say
with great energy. He appears to be judging your responses and formulating his
own ideas. You may think about drinking a lot or this may be the first time you
have thought about it. As he writes you have a moment to reflect on what this is
like for you.
Segment Four:
The counselor continues:
A lot of people say good things about drinking, despite the fact that the amount of
drinking-which you told me about in the paperwork we gathered just before-is
really just harmful and bad in the long run. What do you think about the amount
you drink compared to others?
Segment Five:
As you list these things and he writes them down, you are given another moment to
gather your thoughts about what it is like talking with this particular person about
this subject matter.
116
Segment Six:
The counselor continues:
Ok. I told you a bit before about how changing alcohol drinking is the goal. Even
after you have stopped you can’t really go back and even have one drink. I know
you aren’t thinking about this much right now even though you need to. Have you
always engaged in this kind of drinking or has it changed over time?
Segment Seven:
Again, the counselor seems to be interested in communicating his agenda to you
more than anything else. It may or may not be surprising that the counselor has
brought up the idea of stopping drinking. You think about what he brings up and
also about what you think of him in general.
Segment Eight:
T: Ok, so if you did stop drinking completely, what would have to change in your
life? And could you do what you need to do to change your life and cease to drink
any alcohol?
Segment Nine:
After carefully noting the things you have said, the counselor sets his writing
material down and turns directly towards you. He repeats his statements earlier
about the unquestionable problems with drinking alcohol at all. He wishes you
luck and tells you that he hopes you strongly consider these issues. As he leads you
to the exit, you reflect on whatever thoughts and feelings you just experienced
about the therapist, talking about drinking, what you might do, if anything, and so
on.
117
Appendix P
ATSS Pleasant Scenario Transcript
In this three-minute scenario, the participant is asked to imagine himself sitting on
a tropical beach and to reflect upon the scene.
Segment One:
You have just arrived at the beach on a beautiful Saturday morning. The water is a
deep blue and the sand is warm under your feet. A few birds fly around in the blue
sky above.
Segment Two:
You set down your towel and bag and look around. The beach is not crowded, but
there are some surfers in the water and some children playing soccer down the
beach.
Segment Three:
The sun is really warm on your back and you lie back to enjoy the perfect weather.
A little breeze blows by and cools you off.
Segment Four:
A friend shows up with some cool drinks and snacks and you think about maybe
going swimming or walking down the beach or maybe taking a little nap.
Segment Five
As it gets late in the afternoon and the sun starts to set you decide to pack up your
stuff and get home. You stop at the edge of the beach and watch the sun set for a
moment before leaving an ideal day of rest at the beach.
118
Appendix Q
Coding Manual
This Coding manual is to be used to code the ATSS transcripts for:
Talking About Alcohol: Your Thoughts and Feelings
Please consult with Jed Grodin with any questions and please code as directed.
Meeting deadlines for coding is of the highest priority.
How to Code: What is a thought/utterance?
You will be provided with transcripts that are marked with a slash at the end
of each coding unit. Therefore, you will not have to determine how big to make a
coding unit. Simply look at each coding unit in each segment and mark it if it fits
into one of your assigned codes. Total up the count for each code for each segment
and then for each participant.
You can look to the statements around the coding unit to discern the meaning if it
seems very unclear (see examples below), but do not reach far for a meaning. It is
perfectly fine to not code something if you cannot discern the meaning. If it is not
possible to discern the meaning of an utterance, do not code it, but make a mark in
your coding log to bring up with the project coordinator at a later date. For
example, if someone makes a statement like “I want to get myself under control”
but it is not very clear that that statement is tied to drinking, then do not code it as a
change cognition, but do mark it for review at your coding meeting.
So you have an idea of how we break it down, here is an “inside” look of how we
are thinking: If a participant’s turn includes two statements, each of which can be
assigned a different code (as below), then both are coded as utterances. This would
include:
Two utterances in one sentence that would be given different codes:
I really have to stop smoking (change cog),
but I just don’t want to (resistance to change cog)
or two utterances that state different content (e.g., reasons for or against change):

I’d have a better chance of getting my children back if
I quit drinking (change cog)
and I’m sure I’d feel better, too (change cog)
but I would miss going out with my friends (resistance to change cog)
119
What to Code: The Coding Categories:
The following pages include detailed descriptions of each code. Please
code as appropriate in your SPSS file and keep notes in your coding log of any
questionable coding decisions for later review.
CHANGE COGNITIONS
These include statements of the following cognitions:
Ability to change
Commitment to Change
Desire to Change
Need to Change
Reasons to change
Taking Steps towards Change
CODE 1: Ability to change
I’m positive that I could quit.
Sure I can lose the weight – it’s just a matter of sticking to it.
Absolutely. I can quit whenever I want.
Very likely, I could do it if I tried.
I’m rather sure I can do it.
I’m pretty positive I can do it.
I can do it.
Yes, it’s possible (for me).
I could.
I think I have it in me.
Probably I can do it.
Pretty much, yes, I think I can.
I might be able to.
I guess I could.
I’m sort of good at sticking to things.
*Do not code statements such as ‘I think I could.’
*Do not code ‘if’ statements or statements in the future tense-
(i.e. ‘If I had a problem, I could’).
120
CODE 2: Commitment to Change
Commitment is generally expressed in present or future tense. Statements in past
tense usually do not reflect a present commitment to change.
I guarantee
I will
I promise
I vow
I shall
I give my
word
I assure
I dedicate
myself
I know
I am
devoted to
I pledge to
I agree to
I am prepared to
I intend to
I am ready to
I look forward to
I consent to
I plan to
I resolve to
I expect to
I concede to
I declare my
intention to
I favor
I endorse
I believe
I accept
I volunteer
I aim
I aspire
I propose
I am predisposed
I anticipate
I predict
I presume
I mean to
I foresee
I envisage
I assume
I bet
I hope to
I will risk
I will try
I think I will
I suppose I will
I imagine I will
I suspect I will
I contemplate
I guess I will
I wager
I will see (about)
CODE 3: Desire to Change
Absolutely – I want to get off drugs for
good.
I want to be clean and sober, period.
I’m sick of smoking. It disgusts me.
I really wish I could just cut down.
I’ve just about had it with cigarettes.
I’m very tired of being overweight.
I’d like to get free of dope.
I wish I could just snap my fingers and
lose 40 pounds.
I just want to wake up sober in the morning.
Mostly, I want to quit.
Yeah, probably I do need to eat better.
Part of me wants to exercise.
I guess I’d like to smoke less.
I sort of wish I hadn’t started using coke.
I’m a little bit tired of the drug scene.
*It would be better for me…
CODE 4: Need To Change
I should change/cut down on my
drinking
I definitely have to get off the street.
I can’t go on crashing like this!
I absolutely have to lose weight.
I really have to quit getting messed up
like this.
I need very much to be sober.
It’s really important for me to stop this
nonsense.
I can’t just keep on having these one-nighters.
I’ve got to do something about my drinking.
I have to clean up my act.
Mostly, I can’t keep on having these one-nighters.
Probably I need to do something about my
drinking.
It’s pretty important for me to clean up my act.
I guess I need to cut down.
I kind of have to clean up my act.
I probably need to do a little something about my
drinking
121
CODE 5: Reasons to change
I definitely can’t afford to get another
DWI
There’s no way I want to go back to jail
because of a urine test.
Something has to change or I’m going to
lose my job. (Vocal tone emphasizes
has to)
I’ll be in a lot of trouble if I turn in
another positive urine.
I really can’t afford to get another DWI.
I’m right on the brink of losing my job
and my retirement.
If I lose a lot of money again, my
husband is going to divorce me.
I don’t want to set the wrong example for my
kids.
It’s embarrassing not to remember what I did.
The reasons to quit are starting to pile up.
If I lose money again, my husband is probably
going to leave me.
It’s fairly embarrassing not to remember what I
did.
I’d be a little healthier if I exercised.
I guess I’d be healthier if I exercised.
I’m sort of embarrassed when I can’t remember
what I did.
*Drinking is bad for me.
CODE 6: Taking Steps Towards Change
No way I was going to an AA meeting
this week.
I totally cleaned the cigarettes out of my
house and car yesterday.
I was particularly careful to keep good
records this week.
I really worked on keeping good record
cards this week.
I was very conscious of staying away
from my drug-dealer friends this week.
I actually tested my blood sugar every
day this week.
I stayed away from casinos all week.
I began exercising on Wednesday.
I bought a jogging suit and a good pair
of shoes.
I was a little better about keeping food records
this week.
I did a few things to eat better.
I was somewhat more active this week.
I considered going to an AA meeting.
I thought about smoking only outside, not in the
house or car.
I almost joined a gym last week.
*Code change implied only within 6 months – do
not code statements with references to high
school
122
CHANGE RESISTANCE COGNITIONS
These include statements of the following cognitions:
CODE 7: Inability to change
I have a little trouble sticking to things.
My guess is that I couldn’t.
I might not be able to.
I don’t think I can.
No, I probably couldn’t do it.
I pretty much tried, and it didn’t work.
I just can’t keep off the weight.
I just don’t have it in me.
I couldn’t do it.
I’m pretty sure I can’t do it.
Very unlikely – I wouldn’t have much of a
chance.
I really don’t think I can.
It’s just impossible.
There is no way I could make it without
cigarettes.
I don’t stand a chance.
CODE 8: Commitment not to Change
Commitment to status quo (coded with a negative valence) can be expressed either as a lack of
commitment to change:
I don’t intend to change
I would predict I’m not going to quit.
Or as a statement of commitment to continue current behavior
I intend to keep smoking marijuana.
I bet I’ll still be drinking a year from now.
CODE 9: Desire not to Change
I kind of enjoy smoking.
I guess I’m not very motivated to
exercise.
A little of me would miss the booze.
Pretty much, yes, I like the drinking.
For the most part, I enjoy eating
whatever I like.
Probably I want to keep on smoking.
I just want people to mind their own
business and not hassle me about using
the drugs.
I do enjoy a good Scotch.
Why would I want to quit?
I really don’t want to quit. I really like pot.
I’m very happy the way I am.
I really like the whole ritual of doing it, you
know.
No way. I’m not interested in quitting.
Forget it.
I’m definitely no teetotaler.
123
CODE 10: Lack of Need to Change or Need To Not Change
I sort of have to drink.
I guess I don’t think I need to quit.
I need a little bit of dope to live.
Mostly, I have to drink.
I guess I need some of this excitement in
my life.
I kind of have to keep dealing drugs.
I can’t go without cocaine.
I won’t make it unless I have my pills.
I need to smoke.
I really can’t go without cocaine.
It would be very hard for me to get along without
my pills.
I very much need to smoke.
I definitely have to have my pain pills.
There’s no way I can go through that withdrawal
again. The last time was horrible.
I need cigarettes, period!
I don’t drink as much as/more than other people I
know
CODE 11: Reasons NOT to change
I guess it relaxes me some.
I’d kind of miss going to the casinos.
It’s sort of how I meet people.
Mostly, I just don’t see any benefits to
quitting.
Pretty much all my friends are gamblers.
I probably would have trouble sleeping
without it.
Why should I give up drinking? I just
don’t see how my drinking is a problem.
I have my reasons for what I do.
Gambling gives me something to do during the
day.
I get very relaxed and my problems go away.
It’s really the only way I have to meet people and
make friends.
I get very depressed when I don’t go to the pub.
I positively can’t get off heroin because my
boyfriend always wants me to fix with him.
There’s no way I can work the streets and not do
drugs. You’ve got to be high.
I definitely can’t stand the pain if I don’t have the
pills.
CODE 12: Taking Steps Away from Change
I thought about dropping out of Weight
Watchers.
I nearly went into a casino on Tuesday.
I got this close to buying a bottle.
I was kind of careless about monitoring
my blood sugar.
I ate a little too much.
I sort of slacked off on the exercise.
I didn’t take my Antabuse.
I went to the casino.
I bought a pack of cigarettes.
I really overdid it.
I was very bad with my diet this week.
I was pretty careless about taking my medication.
I totally blew it this week in watching my diet.
I completely slacked off on exercise.
I definitely messed up with keeping records
124
THERAPY COOPERATION COGNITIONS
These cognitions are only those about the therapist of therapy experience itself.
The include statements of the following cognitions:
CODE 13: Nonresistant:
All statements that are cooperative or following the direction set by the
therapist. Anything positive about the therapy/therapist. Ex) being
comfortable with the situation.
CODE 14: Facilitative:
Utterances indicating agreement with the therapist.
Examples:
“Yeah, I agree with what the therapist is saying,” would be coded as one (1)
cooperation cognition.
THERAPY RESISTANCE COGNITIONS
These cognitions are only those about the therapist or therapy experience itself.
These include statements of the following cognitions:
CODE 15: Negative attitude cognitions:
Anti-therapy statements, uncooperative with the therapy and generally
uncooperative.
Examples include: I am not enjoying this experience, I am uncomfortable in this
situation.
CODE 16: Challenge/ Confront cognitions:
Anti- therapist cognitions challenging the reputation, skills, and knowledge
of the therapist.
Example: He/She is not helping me at all, I hope he/she does not have real patients
because they are not doing a good job.
CODE 17: Own agenda:
Bringing up new/other topics directly in response to therapist statements
CODE 18: Not tracking:
Inattention (zoning out), not responding, cognitions that are completely off
topic all because the subject is not paying attention.
Examples: “I just don’t agree with what this guy is saying. It isn’t my fault I drink
anyway,” would be coded as two (2) separate therapy resistance cognitions.
125
Appendix R
Non-disclosure Agreement for research project:
TALKING ABOUT ALCOHOL: YOUR THOUGHTS AND FEELINGS
I hereby certify that I will not disclose any of the data gathered for this research
project to any persons or organizations or any other entity except as required by the
appropriate and applicable federal, state, local, and/or University of Southern
California laws and policies.
--------------------------- ---------------------------------------
(Print) Research Assistant Name Signature and Date

Abstract (if available)

Abstract Current evidence suggests that Motivational Interviewing (MI) can help to create behavior change in a variety of contexts, especially for those with alcohol use problems, but evidence is less clear on how and why MI seems to work. Using the Articulated Thoughts in Simulated Situations paradigm (ATSS), this study aimed to determine the mechanisms by which MI may modify the cognitions and behaviors of college-aged binge drinkers. Two mechanisms typical of an MI intervention were isolated, simulated, and analyzed: the presence of a warm/Rogerian therapist communication style and the use of a decisional balance exercise. By using a 2X2 design to compare these mechanisms to commonly used alternative therapeutic approaches, we sought to determine which mechanisms have the greatest impact on cognitions related to binge drinking and on intention to change, intention to drink, overall impressions of drinking, as well as actual changes in drinking behavior at a 30 day follow-up. As in a previous study, we found moderately strong evidence supporting the validity of the ATSS scenarios for our purposes. No significant effects were found on drinking behavior or any other variables at 30-day follow-up. However, as in the earlier study, the data show that, as compared to a more directive and confrontational communication style, the warm/Rogerian therapist style was associated with significantly fewer participant cognitions reflecting resistance to change during the ATSS scenario (a simulated therapy session) and with a temporary increase in readiness to change immediately after the scenario. We also found that, regardless of condition, higher levels of trait reactance increase the degree to which participants report that the therapist influenced them to think about not changing their drinking behavior.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

In-flight turbulence: an articulated thoughts in simulated situations (ATSS) paradigm investigation of air travelers ' experiences

PDF

Personalized normative feedback applied to undergraduates with problem drinking: a comparison with psychoeducation and an examination of cognitive-affective change mechanisms via the articulated ...

PDF

Measuring narcissistic cognitions in response to relationship threat using articulated thoughts in simulated situations

PDF

Exploring experiences of minority-stress and alcohol use among sexual minorities using articulated thoughts in simulated situations: a proof of concept study

PDF

Memory, gender, and rumination in depression: recall of simulated situations with articulated thoughts and the Rivermead behavioral memory test

PDF

The articulated thoughts of heterosexual male college students in reaction to anti-gay hate speech

PDF

A "second chance" program for vulnerable young adults: a program evaluation and a randomized controlled trial of adjunctive motivational interviewing to improve retention

PDF

Breaking it down to make it stronger: examining the role of source credibility and reference group specificity in the influence of personalized normative feedback on perceived alcohol use norms a...

Asset Metadata

Creator Grodin, Jed P. (author)

Core Title Assessing therapeutic change mechanisms in motivational interviewing using the articulated thoughts in simulated situations paradigm

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 09/29/2006

Defense Date 06/15/2006

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag articulated thoughts,cognitive assessment,college drinking,motivational interviewing,OAI-PMH Harvest,psychotherapy mechanisms of Change

Language English

Advisor Davison, Gerald C. (committee chair), Goodyear, Rodney K. (committee member), Huey, Stanley (committee member), Walsh, David A. (committee member)

Creator Email jgrodin@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m58

Unique identifier UC1113118

Identifier etd-Grodin-20060929 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-4862 (legacy record id),usctheses-m58 (legacy record id)

Legacy Identifier etd-Grodin-20060929.pdf

Dmrecord 4862

Document Type Dissertation

Rights Grodin, Jed P.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu