Perceptions of Motivational Interviewing

Abstract

The practice of motivational interviewing (MI) has gained acceptance as an effective approach to support behavior change in various therapeutic contexts. In recent years, MI has been extended to clients within less traditional therapeutic settings including prisons and probation departments. Despite the known strengths of MI for positively affecting behavioral change in therapeutic contexts, the extent to which probation officers are able to effectively utilize MI remains unknown. The current study utilizes self-report responses from 485 probationers to assess the internal consistency and factor structure of the Client Evaluation of Motivational Interviewing (CEMI) as a tool for gathering feedback on MI fidelity in probation. Confirmatory factor analysis found two distinct MI factors to exist among this client base—technical and relational aspects of MI. Results suggest the CEMI is an effective tool to determine community corrections clients’ self-perceptions of probation officer’s use of MI-consistent techniques.

Keywords

motivational interviewing therapeutic alliance probation

Motivational Interviewing (MI) is increasingly recognized as an empirically supported treatment approach effective in assisting with the modification of a broad spectrum of antisocial behaviors. When implementing MI as part of a therapeutic approach, therapists or “treatment advocates” use a conversational approach with clients to promote prosocial change by guiding the individual to recognize and intrinsically value the need for behavior change. During conversations, the treatment advocate assists the client with resolving any identifiable resistance or ambivalence toward the prosocial behavior change that is needed (Miller & Rollnick, 2013). Robust findings to date suggest the techniques holds significant potential for application to a broad client base. Initially, MI was utilized with clients presenting with alcohol addiction issues (Miller, 1983; Miller & Rollnick, 1991; Walters, Clark, Gingerich, & Meltzer, 2007). Subsequently, studies have found MI can be successful in assisting clients with reducing substance use and risky sexual behaviors that could lead to the spread of HIV infection (Miller & Rose, 2009).

In recent years, the implementation of MI has been extended to clients within alternative treatment settings including probation departments. When supervising probation clients, probation officers are typically tasked with the dual and sometimes conflicting role of ensuring clients sentenced to probation comply with their court-ordered conditions, as well as assist these same clients in developing prosocial behaviors using a variety of treatment techniques (Clark, 2005; Clark, Walters, Gingerich, & Meltzer, 2006; Walters et al., 2007). When MI is used with a probationer client base, the supervising officer acts as a treatment advocate who aims to increase client motivation for change, and positively influence treatment outcomes in a community-based context (Martino, Ball, Nich, Frankforter, & Carroll, 2008).

Despite the known strengths of MI for positively affecting behavioral change in a number of therapeutic contexts, the extent to which probation officers are able to effectively utilize MI as a client-centered approach to communication remains unknown. Probation officers face unique challenges in their therapeutic efforts. During probation office appointments, officers are expected to both counsel their clients while also enforcing supervision conditions with the end goal of behavioral change. Additional challenges presented in the probation context include a less than ideal treatment setting in which “treatment sessions” occur, limitations on time allocated for officer–client interactions, and oftentimes limited resources to support both the therapeutic alliance as well as the client. Though probation officers utilizing MI tend to be thoroughly trained in MI and numerous other therapeutic skills, they are not licensed therapists who focus on treatment alone. All of these factors lead to questions as to whether MI is implemented with fidelity among probation populations given the challenging context.

This study examines a potential avenue for assessing the fidelity of MI implementation within the probation context. The fidelity of MI implementation in a traditional therapeutic setting is typically assessed by supervisors who directly observe client–treatment advocate interactions (Thigpen, Beauclair, Brown, & Guevara, 2012). Here, we test the utilization of a tool, the Client Evaluation of Motivational Interviewing (CEMI) developed by Madson, Bullock, Speed, and Hodges (2009), as a means for gathering feedback on MI treatment fidelity with probation clients. The aim of the CEMI is to serve as a tool that can measure client self-reported perceptions of their treatment advocate’s use of MI during office appointment interactions. If successfully validated with a probation population, the CEMI could become a valuable feedback mechanism for officers as part of their reflections on client interactions, as well as for jurisdictions that have rapidly moved toward the implementation of MI as they strive to implement evidence-based practices in their probation departments.

Literature Review

MI Principles and Processes

MI is a “collaborative communication style for strengthening a person’s own motivation and commitment to change” (Miller & Rollnick, 2013, p. 12). When MI is implemented, communication between an individual client and a treatment advocate is guided by a set of semi-structured interview techniques to increase the individual’s intrinsic level of motivation to change behavior (Miller & Rollnick, 2013). More specifically, the MI approach is unique in its recognition of the client’s ability to change and its tenet that the client has an unsurpassed knowledge of his or her inner values and drive or motivation. The emphasis during treatment is placed on the client as having a key role in determining the reasons and timing for his or her own behavioral change. The treatment advocate acts as a guide throughout the client’s journey of change—recommending and providing services or support to facilitate behavior change (Alexander, Van Benschoten, & Walters, 2008; Miller & Rollnick, 2013).

The key principles in the MI approach encompass four main elements: collaboration, compassion, acceptance, and evocation (Miller & Rollnick, 2013). MI is purported to be more effective if the client and the treatment advocate are collaboratively working toward the same goal—strengthening the clients’ own motivation to change (Hettema, Steele, & Miller, 2005; Miller & Rollnick, 2013). Both relational and technical components of MI are important to facilitating change talk while decreasing discussion about sustaining current behavioral patterns (Miller & Rollnick, 2013; Miller & Rose, 2009). Specifically, the treatment advocate must recognize and accept the perspective of the client, and display empathy for the challenges that he or she faces when determining whether a change in his or her behavior is necessary. This is the relational component of MI. It is vital for the treatment advocate to help the client progress at his or her own pace to ensure sustainable change. Once the client informs the treatment advocate that he or she is ready to move forward with change, the pair can work together to evoke a plan for change. This is the technical component of MI.

Four main processes will occur during the course of a client’s treatment planning with the assistance of the advocate using MI. Each process requires the treatment advocate to utilize a particular skill and/or technique. The process of engaging occurs as the client and the advocate begin to establish their relationship allowing for the advocate to have the ability to begin providing “information and advice to the client with permission” (Miller & Rollnick, 2013, p. 36). With the assistance of his or her advocate, the client determines which goal(s) he or she is ready to target. This process of focusing also supports the client in exploring the disparity between his or her goals and behaviors (Miller & Rollnick, 2013). Next, the evoking process prepares the client for behavior change and emphasizes the importance of enhancing “change talk” and minimizing “sustain talk” by examining the positive outcomes that thinking and behavior change may have on his or her life. Change talk is the communication from the client that indicates the “desire, ability, reasons, and need” to change his or her behavior (Miller & Rollnick, 2013, p. 166). Sustain talk is communication that reaffirms the client’s commitment to his or her current ways of thinking about his or her behavior. Finally, the planning process builds on each of the three prior processes and serves to support the client in determining a clear pathway for implementing changes. Miller and Rollnick (2013) argued that the elements of MI are interrelated, and without one element, the overall approach will fail to meet the standards of the technique itself.

Evidence Supporting MI Effectiveness

Since the development of MI, a large body of research has emerged examining the effects of MI with diverse populations in a variety of situational contexts from clinical control trials (Burke, Arkowitz, & Menchola, 2003; Hettema et al., 2005) to offenders incarcerated in prison (Anstiss, Polaschek, & Wilson, 2011). Largely, this body of research supports the use of MI as an effective stand-alone intervention or as an effective supplement to other treatment programs that reduces substance abuse and other problem behaviors (Burke et al., 2003; Hettema et al., 2005; Lundahl, Kunz, Brownell, Tollefson, & Burke, 2010; Vasilaki, Hosier, & Cox, 2006). A meta-analysis of 72 studies examining the efficacy of MI found 53% of studies reported a significant effect of MI in affecting behavior change across all outcome variables measured (Hettema et al., 2005). Robust effects were found regardless of whether MI was used as a stand-alone intervention or used in addition to other treatment programs. Interestingly, MI was found to have a larger impact on behavioral change more proximal to the onset of treatment with a decline in effects later in treatment. Meta-analytic results also demonstrated more robust effects within samples that were comprised primarily of ethnic minorities (Hettema et al., 2005). A more recent meta-analysis of 119 MI efficacy/effectiveness studies by Lundahl and colleagues (2010) also reported significant effect sizes favoring MI interventions. Similar to Hettema and colleagues, Lundahl and colleagues (2010) found the impact of MI differed by ethnicity with more positive outcomes among clients who were ethnic minorities.

Overwhelmingly, the extant literature on MI interventions supports the use of MI as an effective means for reducing problem behaviors. As a result of this strong evidence, the use of MI techniques is increasingly extended to treatment that encompasses offenders in the criminal justice system (McMurran, 2009). Researchers and practitioners alike have recognized the importance of an offenders’ motivation to change in the criminal desistence process, both for incarcerated offenders and for offenders on probation (Anstiss et al., 2011; Clark, 2005; Clark et al., 2006; Farbring & Johnson, 2008). The majority of research to date has been conducted with MI used as part of clinical trials and in therapeutic settings (Hettema et al., 2005; Lundahl et al., 2010) with only limited research on the use of MI in the correctional context.

Extension of MI into Corrections

The client-centered focus of MI allows for its adoption across a variety of treatment contexts. Given the strength of existing evidence demonstrating MI facilitates behavioral change among clients with substance abuse issues (Miller & Rollnick, 2013), and the high incidence of substance use and abuse within the offender population, many probation jurisdictions have adopted MI as the foundation of their approach to client communication. Clark and colleagues (2006) argued that the use of MI by probation officers benefits both the officer and the client. When using MI, the officer is encouraged to move beyond a security-minded role and recommit to a supportive, therapeutic role when working with probation clients. Proponents for the use of MI within a probation setting (i.e., Clark et al., 2006) emphasize that the approach of MI takes the officer’s dual role in stride by allowing officers to address violations of probation conditions and convey sanctions to probationers while continuing to use a motivational style of interaction (see also Clark, 2005; Walters et al., 2007). Furthermore, by maintaining the supportive role with probationers, officers are better able to handle client resistance, guide probationers into change talk, and “place the responsibility for behavior change on the offender” (Clark et al., 2006, p. 38).

Previous literature suggests the use of MI within a probation setting would be a promising therapeutic approach (Farbring & Johnson, 2008). Within the probation context, MI is applied as a technique of speaking with offenders about their necessary behavior change (Walters et al., 2007) with an underlying assumption that client’s motivation is not a fixed trait. With the appropriate supportive environment, rehabilitation efforts, and interaction with change agents (e.g., probation officers), probation officers have the ability to increase their client’s level of motivation toward achieving prosocial behavior change.

Studies reviewing the efficacy of MI in probation populations have shown MI to increase the likelihood of positive outcomes with some variation based on client characteristics and treatment setting (Anstiss et al., 2011; Harper & Hardy, 2000; McMurran, 2009). Based on this early evidence, MI is increasingly viewed as having the potential to become an evidence-based practice in community corrections. Proponents argue MI holds the potential to increase motivation levels among probationers resulting in increased compliance with court conditions and preparation for treatment programs (Clark et al., 2006; Hartzler & Espinosa, 2011; Walters, Vader, Nguyen, Harris, & Eells, 2010). Research demonstrating the positive impact of MI has also emphasized that offender behavior change takes time; however, studies show that some change can occur even in relatively brief and multi-focused interactions (Anstiss et al., 2011; Walters et al., 2010).

Fidelity of MI Implementation

Along with the encouraging the use of MI, proponents emphasize the need to evaluate the fidelity of MI as delivered by treatment advocates. Despite the development of numerous tools to assess the delivery of MI (see, for example, Motivation Interviewing Assessment: Supervisory Tools for Enhancing Proficiency by Martino et al., 2008, and Motivational Interviewing Treatment Integrity [MITI] by Moyers, Martin, Manuel, Hendrickson, & Miller, 2005), the majority of these tools rely upon direct observation of treatment advocate–client sessions to measure MI fidelity in practice (Madson & Campbell, 2006; Miller & Rose, 2009).

The Motivational Interviewing Skill Code (MISC) is one tool developed by Miller and Mount (2001) in support of observational evaluation of treatment fidelity. Based on the MI Process Code (Barsky & Coleman, 2001), the MISC aids MI trainers and supervisors in identifying provider skills as MI-consistent (see Moyers et al., 2005). Another tool frequently used to assess the fidelity of MI is the MITI scale. Developed by Moyers and colleagues (2005), this tool was also designed to assess therapist competence in the use of MI techniques, with the purpose of being a more focused and reliable assessment tool than the MISC. The MITI was designed to be used primarily by clinical supervisors as a means of measuring clinicians’ foundational competence and change in competence post-training (Moyers et al., 2005) and has been noted as being superior to other measures of MI fidelity (Pierson et al., 2007).

Despite an advancement of MI assessment tools, a number of criticisms of the existing tools remain. First, the tools remain lengthy, complex, and some questions regarding the psychometric properties of the assessment tools exist (Madson et al., 2009; Madson & Campbell, 2006). Second, numerous calls for the evaluation of client perceptions of the treatment advocate’s MI behaviors exist (Moyers & Martin, 2006; Orwin, 2000), yet the focus of these tools typically is on evaluating the behaviors of the treatment advocate (Madson & Campbell, 2006; Moyers & Martin, 2006; Orwin, 2000). Critics suggest that if the evaluation of MI implementation focuses on the treatment advocate’s delivery technique based on client perceptions of the treatment advocate’s use of MI, results could lead to improved identification of gaps in MI delivery, enhanced training of treatment advocates, and modifications in MI delivery (Madson et al., 2009). Such feedback would be especially valuable in the dynamic probation environment. Not only do probation supervisors find themselves overloaded with daily tasks and constantly “putting out fires,” the sheer number of line staff delivering MI under each supervisor presents a significant challenge to complete regular observation of probation officer–client interactions (see Armstrong, 2012). Finally, researchers argue fidelity of MI implementation should be based upon an assessment of two factors during the observed session—the technical aspect and the relational aspect—as discussed earlier (see, for example, Miller & Rose, 2009).

The CEMI Scale

In response to the needs identified above, Madson and colleagues (2009) developed the CEMI scale to measure client perceptions of the use of MI by treatment advocates. After extensive test development, 35 items were included in the original version of the CEMI, which aimed to evaluate a treatment advocate’s “display of the spirit of principles of MI from a client’s perspective” (Madson et al., 2009, p. 7). A high rating of treatment advocates on the CEMI indicates that the advocate is interacting with his or her client in an MI-consistent manner (Madson et al., 2009). In 2013, Madson and colleagues sought to evaluate the psychometric properties of the 35-item CEMI with a predominately White, male population in an inpatient psychiatric setting. Results from exploratory factor analysis suggested the retention of a relational and technical two-factor solution comprised of 16 items from the original CEMI. Confirmatory factor analysis (CFA) subsequently supported these findings (Madson et al., 2013).

In an effort to validate the 16-item CEMI with a more diverse population, Madson, Mohn, Schumacher, and Landry (2015) evaluated the psychometric properties of the CEMI using a predominately African American female sample who were participating in a lifestyle intervention. Results indicate modest support for two-factor structure CEMI and relatively low reliability estimates. However, the predominately African American female participants from the current study viewed their treatment sessions as more MI-consistent than the original predominately White male sample. There was also less variability in the 2015 participants’ perceptions of MI compared with the 2013 sample. While Madson and colleagues (Madson et al., 2015; Madson et al., 2013) have taken strides to validate the CEMI, to our knowledge, this tool has not been validated with a probation sample despite its potential for widespread application in the probation context and ongoing need for assessment within jurisdictions that have already implemented an MI approach with their clients.

The Current Study

The current study seeks to assess the internal consistency and factor structure of the CEMI Scale in a probation client base. Specifically, this article examines whether the two primary theoretical factors of MI—technical and relation aspects—are evident in a self-report survey with probation clients. The survey queries the extent to which the client’s current probation officer engages in MI as part of the officer–client communication and exhibits MI-consistent communication behaviors as intended within the jurisdiction. Based on the findings of Madson and colleagues’ (2013) evaluation of the CEMI in a psychiatric setting, we anticipate responses will demonstrate the two distinct MI factors as identifiable among this client base.

Method

Participants

Data for this study were collected from probation clients in a large probation department who attended an in-person office appointment during the evaluation period. All officers in the probation department received MI training and were encouraged to utilize an MI approach during interactions with their clients. At the conclusion of a probationer’s office visit, research staff approached the client to present him or her with an opportunity to complete an anonymous survey. The client was informed that the survey required less than 15 min to complete and that the survey was voluntary. Less than 5% of the probationers approached about the survey refused to participate. Probationers who declined participation typically indicated that they needed to depart the office immediately to arrive punctually at work. Approximately, 25 participants who attempted to complete the survey failed to respond to three or more of the CEMI indicators on the survey (the focus of this study) and were consequently excluded from the analysis. A total of 485 probationers who completed usable surveys form the basis of the participants in this study.

Limited demographic data were collected to protect the identity of the probation clients. Of the personal data collected, the majority of the clients indicated this was their first probation sentence (57.3%). Sentence lengths ranged from 1 to 168 months, with the most common length of sentence reported to be 12 months (13%), followed by 36 months (8.1%), 24 months (7.6%), and 60 months (7.6%). The participants were equally likely to be supervised by the same probation officer across the duration of their probation terms (49.9%) as having changed supervising officer (50.1%). Furthermore, the majority of participants (71.1%) were employed at the time of the survey.

Materials

CEMI

As discussed above, the CEMI is a self-report tool developed by Madson and colleagues (2009) to assess a client’s perceptions of his or her treatment advocate’s utilization of MI at the conclusion of his or her interaction. As Madson and colleagues discuss, the original CEMI was validated using both content and face validity approaches, along with the completion of a card sort activity. The end result of these validation efforts was the creation of the 35-item scale. Subsequently, Madson and colleagues (2013) validated the CEMI with a psychiatric inpatient population, finding 16 items most efficiently reflected the two primary dimensions of MI as discussed by Miller and Rose (2009).

In this survey, we include the CEMI indicators validated by Madson and colleagues (2013), which represent the technical and relational dimensions of MI with the exception of one indicator (“Changed the topic when you became upset about changing your behavior”), which was deemed inappropriate given the nature of the probation context. Thus, the survey included a total of 15 indicators for the CEMI–Probation version. Each item was measured on a 4-point Likert-type scale with responses ranging from 1 (never) to 4 (a great deal). Univariate statistics for the 15 CEMI items are displayed in Table 1. See Appendix for inter-item correlations.

Table 1:

Client Evaluation of Motivational Interviewing Descriptive Statistics

During your recent office visit, how much did your probation officer. . .	M	SD
1. Focus on your weakness	2.61	1.16
2. Help you talk about changing your behavior	2.85	1.04
3. Act as a partner in your behavior change	2.89	1.10
4. Help you discuss your need to change your behavior	2.64	1.19
5. Make you feel distrustful of him or her	1.62	1.04
6. Help you examine the pros and cons of changing your behavior	2.88	1.10
7. Help you feel hopeful about changing your behavior	2.83	1.15
8. Argue with you to change your behavior	1.48	0.91
9. Push you forward when you become unwilling to talk about an issue further	2.16	1.18
10. Act as an authority on your life	2.14	1.16
11. Tell you what to do	1.96	1.09
12. Argue with you about needing to be 100% ready to change your behavior	1.76	1.11
13. Show you that he or she believes in your ability to change your behavior	3.05	1.08
14. Help you feel confident in your ability to change your behavior	3.12	1.04
15. Help you recognize the need to change your behavior	3.05	1.07

Procedure

Researchers administered the self-reported survey to voluntary participants using paper and pencil format in small groups. Surveys were administered in a private classroom or conference room without the presence of probation staff. Due to the location of the rooms, the supervising probation officer was not aware of whether his or her client(s) agreed or declined participation. To further insure the anonymity of participants, demographic data were not collected as part of the survey, and date of survey completion was not recorded. The survey and its administration process was approved by an appropriate institutional review board. The survey focused on the extent to which the participating probation client perceived his or her supervising probation officer utilized techniques of MI during his or her office interactions.

Analytical Approach

Prior studies as well as the theoretical basis of MI suggests two primary components exist, relational and technical, which are correlated with each other. Here, we utilize a CFA approach to determine the theoretical model fit to these data. The CFA model was estimated using robust weighted least squares (WLSMV), which is the appropriate estimator in Mplus Version 7 for analyses involving binary or ordered categorical indicator variables (Byrne, 2012; Raykov & Marcoulides, 2006). Furthermore, 13 of the 15 CEMI items in the CFA model were considered to have distributions that were significantly non-normal (Warner, 2013). WLSMV has been shown to report accurate test statistics, parameter estimates, and standard error with normal and non-normal distributions of latent responses so the skewed indicators do not pose a problem for the analysis (Byrne, 2012; Kline, 2011). This analytical approach reflects the methodology used by Madson and colleagues (2013) in their validation of the CEMI within a psychiatric setting. All 485 cases were included in the CFA described.

Results

CFA

CFA examined the fit of the theoretical model (i.e., the relational and technical components of MI) to the probationer’s perceptions of MI used by his or her probation officer. As shown in Table 2, results indicated a relatively strong fit of the data to the theoretical model. Similar to items in Madson et al.’s (2015) analysis of the CEMI scale, items adhered to the technical and relational factor dimensions of MI. Nine items loaded most robustly on the technical factor including items that targeted the probation officer’s discussion of behavior change in a manner consistent with MI theory (Miller & Rollnick, 2013; Miller & Rose, 2009). The items that loaded most robustly included indications that the supervising probation officer is perceived by the client to “act as a partner in behavior change,” “help you recognize the need to change your behavior,” and “help you feel confident in your ability to change your behavior.” Cronbach’s alpha demonstrated that internal consistency among the nine items in the technical factor was very strong (α = .939).

Table 2:

Client Evaluation of Motivational Interviewing Factor Loadings: Confirmatory Factor Analysis

During your recent office visit, how much did your probation officer. . .	Technical factor loadings	Relational factor loadings
1. Focus on your weakness	.63*
2. Help you talk about changing your behavior	.88*
3. Act as a partner in your behavior change	.86*
4. Help you discuss your need to change your behavior	.76*
5. Make you feel distrustful of him or her		.60*
6. Help you examine the pros and cons of changing your behavior	.85*
7. Help you feel hopeful about changing your behavior	.79*
8. Argue with you to change your behavior		.73*
9. Push you forward when you become unwilling to talk about an issue further		.80*
10. Act as an authority on your life		.77*
11. Tell you what to do		.63*
12. Argue with you about needing to be 100% ready to change your behavior		.82*
13. Show you that he or she believes in your ability to change your behavior	.93*
14. Help you feel confident in your ability to change your behavior	.95*
15. Help you recognize the need to change your behavior	.94*

p < .001.

The indicators in the relational factor measured the behaviors of the probation officer during interactions with probationers (Miller & Rose, 2009). Six items loaded most robustly on the relational factor including “argue with you to change your behavior,” “tell you what to do,” and “argue with you about needing to be 100% ready to change your behavior,” indicating the highest factor loadings. The internal consistency of the items on the relational factor was deemed satisfactory using Cronbach’s alpha (α = .787).

Model fit was assessed using multiple techniques to provide a complete evaluation of proposed model fit (Raykov & Marcoulides, 2006). First, a chi-square index indicated an adequate model fit, χ²(89) = 526.73, p < .01. In addition, the root mean square error of approximation (RMSEA) index value of .098 (p < .001, 90% confidence interval [CI] = [.90, .107]) indicated the model reasonably approximates the data analyzed. Finally, the comparative fit index (CFI) and the Tucker–Lewis fit index (TLI) were used as a final model fit comparison. Both the CFI and TLI range from 0 to 1, with values closer to 1 indicating a well-fitting model. Here, results indicated the proposed model structure fit the data well (CFI = .977, TLI = .973). The correlation between the technical and relational factor was .250.

Discussion

The use of the CEMI to evaluate the fidelity of MI when implemented in other areas of practice is a necessary extension of the literature. This study creates a foundation of knowledge for further assessing the utilization of MI in correctional settings using the CEMI. Here we find evidence that the CEMI can reliably capture the two primary factors of MI as outlined by Miller and Rollnick (2013) in a non-psychiatric-based client population. According to the results from the CFA, the CEMI is effective at determining community corrections clients’ self-perceptions of probation officer’s use of MI-consistent techniques.

Results indicated the dispersion of the CEMI items on the two separate factors that parallel the technical and relational factors targeted by the 16-item CEMI scale evaluated by Madson and colleagues (2013), which was the basis for the CEMI used in this sample. The items that load primarily on the two distinct factors make intuitive sense with the explanations of the technical and relational factors given by Miller and Rose (2009). The item “Push you forward when you become unwilling to talk about an issue further” was the one item whose loading was the most closely contested between the two factors. In Madson et al. (2015), this particular item was found to have a negative loading with the MI Relationship factor and deemed to not be theoretically supported. Madson and colleagues further determined from anecdotal accounts that participants tended to be confused between the intent of this item and “change the topic when you became unwilling to talk about an issue further.” Their suggestion was a modification of the instrument wording. Unfortunately, this finding was published subsequent to our own study to implement the revised wording. Future studies should consider including Madson et al.’s suggestion of the revised item “make you talk about something you didn’t want to discuss.” In conclusion, evidence that the community corrections sample fit the two-factor model structure was clear. Moreover, the correlation between the technical and relational factors in this study was positive and strong.

The results of CEMI as a valid tool for assessing the fidelity of MI implementation within a probation context are promising; however, the study is not without its limitations. First, to ensure the clients felt comfortable in assessing their relationships with officer, we did not collect any significant personal identifiers. This limitation of the data precludes any examination of individual differences in client perceptions. Future research should attempt to collect additional data that would allow for the exploration of the relationships between client characteristics and perceptions of MI techniques within a probation context with a particular attention to differences between ethnic subgroups. This information would allow for a determination of whether MI effects are more robust in subpopulations of offenders as suggested by Madson and colleagues (2015) among others.

Next, this study focused on a validation of a specific instrument within a probation context without any comparison of the CEMI responses with other approaches to assessing the fidelity of MI. Future research could include a comprehensive assessment of MI use in a probation setting using both an observational assessment by a trained clinician and the CEMI to compare results of the two assessment modalities. An observational checklist or scale for the use of MI practice in a probation setting could be developed from this approach as an added quality assurance method to confirm both the officer’s skills and the client’s perceptions of the technical and relational factors assessed using the CEMI.

The promising finding that the CEMI is a valid tool for assessing the fidelity of MI in a probation context has a clear and definitive next step. Importantly, researchers should engage in evaluations to examine the relationship between MI perceptions as measured with the CEMI and its connection to client’s likelihood of recidivism. These results will be critical for determining the extent to which MI should be considered an evidence-based practice with probation clients. Implications of these findings for probation officers are significant. Initial evidence that the CEMI–Probation version can be used as a valid tool for measuring client perceptions of MI as used by probation officers sets the stage for cost-effective and efficient feedback to officers, opens the door for use as a training tool, and might facilitate quality assurance processes conducted by probation supervisors. Such tools are particularly important given the broad span of control that probation officers have with their caseload, and probation supervisors have over probation officers especially within jurisdictions emphasizing evidence-based practices (Armstrong, 2012). The CEMI could be used in a proactive manner to determine a need for booster training of probation staff and/or targeted mentorship in support of individual officers who are struggling with effectively employing the technique.

Footnotes

Appendix

Correlation Matrix

CEMI indicator	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1. Focus on your weakness	1
2. Help you talk about changing	.52	1
3. Act as a change partner	.51	.76	1
4. Help you discuss need to change	.34	.67	.61	1
5. Make you feel distrustful	.09	.13	.12	−.03	1
6. Help examine pros and cons of change	.40	.67	.77	.58	.11	1
7. Help you feel hopeful about changing	.37	.61	.69	.62	−.07	.72	1
8. Argue with you to change	.10	.11	.11	.15	.43	.11	.09	1
9. Push you forward to talk	.28	.40	.38	.39	.21	.37	.33	.32	1
10. Act as an authority on your life	.18	.30	.24	.31	.28	.25	.24	.32	.46	1
11. Tell you what to do	.11	.15	.08	.15	.28	.07	.09	.38	.33	.51	1
12. Argue with you about need to change	.18	.19	.19	.20	.28	.18	.18	.54	.37	.47	.45	1
13. Show you he or she believes you can change	.38	.60	.66	.55	−.03	.65	.64	.10	.37	.21	.14	.20	1
14. Help you feel confident to change	.40	.61	.67	.52	−.00	.68	.67	.10	.36	.24	.10	.17	.88	1
15. Help you recognize the need to change	.41	.66	.68	.60	.05	.69	.65	.17	.41	.28	.15	.25	.84	.82	1

Gaylene S. Armstrong is professor and chair in the Department of Criminal Justice and Criminology at Sam Houston State University. Her research focuses on advancing knowledge on evidence-based policy and practice in corrections.

Cassandra Atkin-Plunk is an assistant professor in the School of Criminology and Criminal Justice at Florida Atlantic University. Her research interests span institutional and community corrections, focusing primarily on evidence-based practices, program evaluations, and reentry and reintegration.

Nancy R. Gartner is a doctoral student in the Department of Criminal Justice and Criminology at Sam Houston State University. Her research interests focus on the U.S. correctional system, specifically the challenges of reentry and the aging correctional population.

References

Alexander

Van Benschoten

S. W.

Walters

S. T.

(2008). Motivational interviewing training in criminal justice: Development of a model plan. Federal Probation, 72(2), 61-66.

Anstiss

Polaschek

D. L. L.

Wilson

(2011). A brief motivational interviewing intervention with prisoners: When you lead a horse to water, can it drink for itself? Psychology, Crime & Law, 17, 689-710.

Armstrong

(2012). Factors to consider for optimal span of control in community supervision evidence based practice environments. Criminal Justice Policy Review, 23, 427-446.

Barsky

Coleman

(2001). Evaluating skill acquisition in motivational interviewing: The development of an instrument to measure practice skills. Journal of Drug Education, 31, 69-82.

Burke

B. L.

Arkowitz

Menchola

(2003). The efficacy of motivational interviewing: A meta-analysis of controlled clinical trials. Journal of Consulting and Clinical Psychology, 71, 843-861.

Byrne

B. M.

(2012). Structural equation modeling with Mplus: Basic concepts, applications, and programming. New York, NY: Taylor & Francis.

Clark

M. D.

(2005). Motivational interviewing for probation staff: Increasing the readiness to change. Federal Probation, 69(2), 22-28.

Clark

M. D.

Walters

Gingerich

Meltzer

(2006). Motivational interviewing for probation officers: Tipping the balance toward change. Federal Probation, 70(1), 38-44.

Farbring

C. A.

Johnson

W. R.

(2008). Motivational interviewing in the correctional system: An attempt to implement motivational interviewing in criminal justice. In Miller

W. R.

Rollnick

(Eds.), Motivational interviewing: Preparing people for change (2nd ed., pp. 304-323). New York, NY: Guilford Press.

10.

Harper

Hardy

(2000). An evaluation of motivational interviewing as a method of intervention with clients in a probation setting. British Journal of Social Work, 30, 393-400.

11.

Hartzler

Espinosa

E. M.

(2011). Moving criminal justice organizations toward adoption of evidence-based practice via advanced workshop training in motivational interviewing: A research note. Criminal Justice Policy Review, 22, 235-253.

12.

Hettema

Steele

Miller

W. R.

(2005). Motivational interviewing. Annual Review of Clinical Psychology, 1, 91-111.

13.

Kline

R. B.

(2011). Principles and practice of structural equation modeling (3rd ed.). New York, NY: Guilford Press.

14.

Lundahl

B. W.

Kunz

Brownell

Tollefson

Burke

B. L.

(2010). Meta-analysis of motivational interviewing: Twenty-five years of research. Research on Social Work Practice, 20, 137-160.

15.

Madson

M. B.

Bullock

E. E.

Speed

A. C.

Hodges

S. A.

(2009). Development of the client evaluation of motivational interviewing. Motivational Interviewing Network of Trainers Bulletin, 15(1), 6-8.

16.

Madson

M. B.

Campbell

T. C.

(2006). Measures of fidelity in motivational enhancement: A systematic review. Journal of Substance Abuse Treatment, 31, 67-73.

17.

Madson

M. B.

Mohn

R. S.

Schumacher

J. A.

Landry

A. S.

(2015). Measuring client experiences of motivational interviewing during a lifestyle intervention. Measurement and Evaluation in Counseling Development, 48, 140-151.

18.

Madson

M. B.

Mohn

R. S.

Zuckoff

Schumacher

J. A.

Kogan

Hutchison

Stein

(2013). Measuring client perceptions of motivational interviewing: Factor analysis of the client evaluation of Motivational Interviewing Scale. Journal of Substance Abuse Treatment, 44, 330-335.

19.

Martino

Ball

S. A.

Nich

Frankforter

T. L.

Carroll

K. M.

(2008). Community program therapist adherence and competence in motivational enhancement therapy. Drug and Alcohol Dependence, 96, 37-48.

20.

McMurran

(2009). Motivational interviewing with offenders: A systematic review. Legal and Criminal Psychology, 14, 83-100.

21.

Miller

W. R.

(1983). Motivational interviewing with problem drinkers. Behavioural Psychotherapy, 11, 147-172.

22.

Miller

W. R.

Mount

K. A.

(2001). A small study of training in motivational interviewing: Does one workshop change clinician and client behavior? Behavioral and Cognitive Psychotherapy, 29, 457-471.

23.

Miller

W. R.

Rollnick

(1991). Motivational interviewing: Preparing people to change addictive behavior. New York, NY: Guilford Press.

24.

Miller

W. R.

Rollnick

(2013). Motivational interviewing: Helping people change. New York, NY: Guilford Press.

25.

Miller

W. R.

Rose

G. S.

(2009). Toward a theory of motivational interviewing. American Psychologist, 64, 527-537.

26.

Moyers

T. B.

Martin

(2006). Therapist influence on client language during motivational interviewing sessions: Support for a potential causal mechanism. Journal of Substance Abuse Treatment, 30, 245-251.

27.

Moyers

T. B.

Martin

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, 19-26.

28.

Orwin

R. G.

(2000). Assessing program fidelity in substance abuse health services research. Addiction, 95(Suppl. 3), S309-S327.

29.

Pierson

H. M.

Hayes

S. C.

Gifford

E. V.

Roget

Padilla

Bissett

. . . Fisher

(2007). An examination of the motivational interviewing treatment integrity code. Journal of Substance Abuse Treatment, 32, 11-17.

30.

Raykov

Marcoulides

G. A.

(2006). A first course in structural equation modeling (2nd ed.). New York, NY: Psychology Press.

31.

Thigpen

M. L.

Beauclair

T. J.

Brown

Guevara

(2012). Motivational interviewing in corrections: A comprehensive guide to implementing MI in corrections. Washington, DC: National Institute of Corrections.

32.

Vasilaki

E. I.

Hosier

S. G.

Cox

W. M.

(2006). The efficacy of motivational interviewing as a brief intervention for excessive drinking: A meta-analytic review. Alcohol and Alcoholism, 41, 328-335.

33.

Walters

S. T.

Clark

M. D.

Gingerich

Meltzer

M. L.

(2007). A guide for probation and parole: Motivating offenders to change (NIC Accession Number 022253). Washington, DC: National Institute of Corrections.

34.

Walters

S. T.

Vader

A. M.

Nguyen

Harris

T. R.

Eells

(2010). Motivational interviewing as a supervision strategy in probation: A randomized effectiveness trial. Journal of Offender Rehabilitation, 49, 309-323.

35.

Warner

R. M.

(2013). Applied statistics. Thousand Oaks, CA: SAGE.