Introduction

Over the past 6 decades, the multicultural counseling (MC) movement has rendered far-reaching impacts on the fields of psychology and mental health, including counseling psychology, counselor education, and clinical psychology, both in Canada (Arthur and Collins 2010) and the U.S. (Abreu et al. 2000; Ponterotto 2008). Within this MC movement, the question surrounding ‘best practice’ in delivering multicultural counseling training (MCT) and curricula for counselors and psychologists constitutes one key focus of the MC literature – that is, the MCT research (Smith et al. 2006). While much empirical knowledge has been acquired through cumulative literature on MCT thus far, MC scholars and practitioners have pointed to a number of critical gaps and challenges facing this body of research (Pieterse et al. 2009; Smith and Trimble 2016).

For instance, at the practical front, there have been calls by MCT researchers and practitioners for multicultural education to be more effective in promoting and cultivating students’ multicultural counseling and therapy skills (Kuo 2012; Smith et al. 2006) and more proactive in incorporating social justice issues into the current MCT curriculum (e.g., Pieterse et al. 2009). As well, there has been an increasing call for current MCT curricula to transcend beyond the typical didactic, in-class, and single-course MC coursework (Priester et al. 2008); for example, with experientially-based, hands-on cultural training involving direct services to culturally diverse clients (Dickson and Jepsen 2007). At the research and methodological front, MC scholars have advocated for MCT research to: a) examine both the outcomes and the process associated with trainees’ development of multicultural counseling competencies (MCC) (Barden and Cashwell 2013) and b) incorporate longitudinal research designs and data collection to help address more in-depth and process-focused research questions about MCC (Sammons and Speight 2008), particularly as pertaining to how trainees learn and acquire actual multicultural skills (Sheu and Lent 2007).

While MCT and practice are particularly relevant to Canada, where the current study was undertaken, as a country of immigrants and diverse populations, a previous review has shown that current MC research and literature are surprisingly scarce (Kuo 2012). Further to that, Canadian MCT researchers Arthur and Januszkowski (2001) observed a decade and a half ago that there was a lack of research and empirical findings on if and how MCT is being implemented in counselor education, counseling psychology, and clinical psychology programs across Canada. Considering the above, this present research had two-pronged objectives: a) to describe a unique refugee-serving multicultural counseling practicum course for advanced-level clinical psychology Ph.D. students at the University of Windsor, Ontario; and b) to assess the effects this practicum-based model of MCT has on clinical/therapist trainees’ reported MC competence, self-efficacy, and skills by collecting both quantitative and qualitative data longitudinally.

For these objectives, in this article we first review literature on experientially-based MC education (i.e., cultural immersion and MC supervision and practica) and then briefly describe the design of this MC psychotherapy practicum. Following that, we investigated the outcomes and the process of trainees’ MC development within this practicum from three cohort groups: the 2010–2011, the 2011–2012, and the 2012–2013 academic years. To these ends, we examined the pre- to post-changes in trainees’ MC competency and self-efficacy, and also analyzed trainees’ weekly post-session critical incident journals (CIJ) with a Multi-Level Model (MLM) analysis. Lastly, we attempted to elucidate trainees’ nuanced cultural and clinical learning experiences by probing into and presenting exemplar narratives from trainees’ CIJs.

It is important to note that while this multicultural practicum was designed and implemented within a clinical psychology Ph.D. program, this practicum-based MC training model and the current study’s findings hold direct relevance for students, clinicians, and graduate training programs in other mental health disciplines, including counselor education, counseling psychology, social work, psychiatry, etc. This is because all of these disciplines have a practical training component and are concerned with how best to enhance clinicians’ ability to overcome cultural barriers in serving diverse client and patient populations. As well, by definition, multicultural counseling competencies are trans-theoretical and trans-disciplinary culturally-informed competencies expected of all mental health clinicians (e.g., American Counseling Association 2005; American Psychological Association 2003).

Experientially-Based Multicultural Counseling Training: Theories and Evidence

Prior research has touted experiential approaches to MCT as a superlative way of teaching and facilitating students’ learning of multicultural and intercultural knowledge and skills, since such instructional methods expose students to culturally-diverse populations or place them in cross-cultural environments (Kim and Lyons 2003). For example, a study of Canadian counselor trainees found that experiential MCT facilitated trainees’ development of ‘cultural empathy’ and enhanced their ability to integrate multicultural theory with practice (Arthur and Achenbach 2002). Similarly, in a sample of U.S. counselor trainees, Sammons and Speight (2008) found that a combined didactic-experiential/interactive MCT approach was the most effective way of bringing about attitudinal and behavioral changes in trainees.

Multicultural counseling or psychotherapy practica represent a distinctive type of experientially-based MCT, in which counselors-in-training provide direct psychological services and interventions to actual multiculturally diverse clients within a supervised clinical context (Kuo and Arcuri 2014). The need and benefit of incorporating MC practica to complement didactic MC coursework for counselling and clinical trainees have been advocated repeatedly over the years (e.g., Abreu et al. 2000; Smith et al. 2006). For example, in a sample of 516 students in U.S. counsellor education programs, Dickson and Jepsen (2007) found that trainees who had exposure to multicultural practica reported higher levels of MC relationship and MC awareness above and beyond the exposure to all other MC interventions, such as directed lectures, reading, guest speakers, simulation activities, role-play, etc. However, despite the apparent merits and appeals of MC practica, there is currently very little research on how to design, teach, and implement experientially-based MCT generally (Kim and Lyons 2003) and MCT practica specifically (Kuo and Arcuri 2014). For this reason, to the authors’ knowledge, the multicultural practicum outlined in the current study represents the first MCT model of this kind that involves actual, community-referred refugees with mental health needs across North America. In the following sections we provide a brief overview of this refugee-serving multicultural practicum housed at the University of Windsor. Interested readers are referred to Kuo and Arcuri (2014) and Kuo (2018) for more detailed discussion about this training MC practicum.

Description of the Current Refugee-Serving Multicultural Therapy Practicum

Student Trainee and Supervisor Characteristics

This refugee-serving multicultural therapy practicum is an 8-month long (September to April) in-house, course-based therapy training in a fully Canadian Psychological Association (CPA) accredited Adult Clinical Psychology Ph.D. program at the University of Windsor in Ontario, Canada. Each year a cohort of 4 to 5 advanced doctoral students, who are in their 4th or 5th year of the Clinical Psychology Ph.D. program, enrol in this cultural-clinical practicum. As a pre-requisite to the practicum, trainees are required to first complete a full-semester didactic MC course, which is grounded in the Tripartite Model of MCC as stipulated by Sue et al. (1992) and comprises cultural awareness, knowledge, and skills. This didactic-practical-experiential, cultural-clinical training sequence/model has been identified in the literature as the optimal approach to designing MC education for counselors and psychologists by multicultural scholars and educators (e.g., Abreu et al. 2000; Sammons and Speight 2008). In the current practicum, the supervisor/instructor (the first author) is a tenured faculty and a registered psychologist with expertise in multicultural psychology and cross-cultural psychology.

Community Partner and Refugee Client Characteristics

Grounded in the principle and the spirit of inter-professional collaboration (APA 2010; Kuo 2018), this practicum is built on the basis of community-university partnership. The refugee clients whom the therapy practicum serves are referred by a local immigrant and refugee settlement community agency – the Multicultural Council of Windsor and Essex. The therapist trainees and the practicum supervisor work in concert with case managers at the agency, who in turn provide the essential logistic supports for refugees in attending the therapy (e.g., appointment scheduling, transportation, language interpretation for therapy, housing, health care service, etc.). Since the practicum began in 2007, it has served refugee survivors of trauma and torture (hereafter referred to as ‘refugee survivors’) from Iraq, Syria, Republic of Congo, Ethiopia, Somali, Sudan, Eritrea, Afghanistan, Bhutan, Myanmar, and Colombia. In many cases, these refugees’ past sufferings are among the most severe forms of social injustice and atrocity perpetrated against humanity (Gorman 2001).

Didactic Seminars, Therapy Sessions, and Clinical Supervision

The practicum begins with 5–6-weeks of pre-therapy seminars, surveying critical content knowledge and logistical information pertaining to working with refugee survivors and with the community agency, including vicarious traumatization, refugee trauma, torture, and recovery, newcomers’ acculturation concerns/issues, skills in working with a language interpreter, etc. Following these didactic seminars, actual therapy with refugee clients occurs for a maximum of approximately 18 weekly sessions, from October to April (which occurs within a full academic semester).

The practicum adopts the treatment model for refugee survivors rooted in the principles of multiculturally-informed traumatology and trauma therapy as propositioned by Gorman (2001). As follows, the practicum aims to facilitate refugee clients’ healing through the processes/stages of safety, reconstruction, and reconnection (see Kuo and Arcuri 2014 for a more detailed discussion). Finally, clinical supervision for the therapist trainees takes the form of weekly 2-h group supervision. The supervision approach is framed and grounded within the foundation of the Tripartite MCC Model of Sue et al. (1992), traumatology, and social justice, advocacy and service. Following this Tripartite model, specific emphases during the supervision are placed on trainees’ learning and cultivation of cultural awareness, knowledge, skills, and relationships, and on their personal and professional self-reflection, examination, and development.

The Current Study

The evaluative component of this present research was to assess the effects of this refugee-serving MC practicum on therapist trainees’ reported MC development, with respect to their MC competency, MC self-efficacy, and MC therapy skills. As follows, in this study we attempted to address two central research questions:

  1. 1)

    Research Question #1 – “How does providing direct therapy to refugee clients through the multicultural therapy practicum course impact the therapist trainees’ development of multicultural counseling competencies (i.e., their level of cultural awareness, knowledge, skills, and relationships with clients), and their sense of multicultural self-efficacy (i.e., MC intervention, assessment, and session management)?

  2. 2)

    Research Question #2 (including sub-questions) – “How might we capture the way in which therapist trainees’ MC therapy skills evolve and develop within this multicultural practicum, as assessed through the analyses of trainees’ written weekly critical incident journal entries (qualitative)? Is this process of trainees’ growth in these core MC therapy skills linear or non-linear? Is there intraindividual variability within these developmental patterns of MC therapy skills among trainees?

For Research Question #1, we hypothesized that trainees would show significant improvement (increases) in each domain of MC competency and MC self-efficacy, respectively, between their baseline at pre-test and at post-test on completion of the practicum. For Research Question #2, based on an analysis of trainees’ qualitative CIJs, we hypothesized that participant trainees would show a gradual, incremental or ‘linear’ increase over the course of the practicum. Individual variability in their starting point and the slope (i.e., rate of change over time) among trainees would also be expected. Additionally, we anticipated that the trainees’ qualitative CIJs would reveal ‘thick,’ nuanced, and insightful gains in trainees’ cultural and clinical learning as they worked and interacted directly with refugee clients at the micro-process level of counseling and therapy.

Methods

Participants

The study received full ethics approval by the Research Ethics Board at the University of Windsor. The 14 participants (13 female and 1 male), who were doctoral-level clinical psychology trainees at the University of Windsor enrolled in the multicultural therapy practicum course across the 2010–2011 (N = 4), 2011–2012 (N = 5), and 2012–2013 (N = 5) cohort years, read and signed the informed consent form prior to the start of the study. Participants ranged from 25 to 31 years of age, with an average age of 26.79 (SD = 1.93). In terms of their racial/ethnic backgrounds, eleven self-identified as “White/Caucasian” and three as “East Asian”. All but two reported that they were born in Canada.

Quantitative Data Collection and Analysis Procedures

To address Research Question #1 and to quantitatively assess participants’ self-report learning outcomes in terms of their temporal change in MC development, the trainees completed the Multicultural Counseling Inventory (MCI) and Multicultural Counseling Self-Efficacy Scale – Racial Diversity Form (MCSE-RD) at the beginning of the practicum course in the Fall of the academic year (as the pre-test) and then at the end of the practicum course (as the post-test). These pre- and post-test scores were then analyzed with a series of t-tests to help answer this research question. The measures used are described below.

Multicultural Counseling Inventory (MCI)

The MCI (Sodowsky et al. 1994) is a 40-item measure of MCC scaled on a 4-point Likert-scale ranging from 1 (very inaccurate) to 4 (very accurate). The MCI consists of four factors/sub-scales that assess a respondent’s MCC with respect to MC awareness, knowledge, skills, and relationship. In the current study, the MCI showed an acceptable to good level of reliability with the following Cronbach’s alphas: MC Knowledge (pre-test: .70; post-test: .80); MC Awareness (pre-test: .21; post-test: .75); MC Skills (pre-test: .67; post-test: .70); MC Relationship (pre-test: .79; post-test: .77); and the overall MCI (pre-test = .74; post-test: .90).

Multicultural Counseling Self-Efficacy Scale-Racial Diversity Form (MCSE-RD)

The 37-item MCSE-RD (Sheu and Lent 2007) is a domain-specific measure of the MC skills dimension of the Tripartite MCC Model of Sue et al. (1992) scaled on a 10-point Likert scale ranging from 0 (no confidence at all) to 9 (complete confidence). It assesses participants on the dimensions of: Multicultural Intervention; Multicultural Assessment; and Multicultural Counseling Session Management. In the current study, the MCSE-RD demonstrated good reliability: Multicultural Intervention (pre-test: .90; post-test: .94); Multicultural Assessment (pre-test: .79; post-test: .91); Multicultural Counseling Session Management (pre-test: .83; post-test: .80); and the overall MCSE-RD (pre-test: .92; post-test: .95).

Qualitative Data Collection and Analysis Procedures

To address Research Question #2 concerning the process, the pattern of change, and the ‘lived experience’ of trainees’ learning and implementation of MC therapy skills over time, we adhered to the following methods and procedures.

Critical Incident Journal (CIJ)

To capture the therapist trainees’ patterns of change and personal narratives about the evolvement of their cultural competency over the course of the MC practicum, the CIJ method was employed. “Critical incidents” have been defined as interpersonal encounters that impact an individual’s personal and professional development (Coleman 2006). CIJ is comprised of a target incident followed by a series of guided personal and professional reflections and has been successfully used in numerous MCT studies (e.g., Arthur and Januszkowski 2001; Coleman 2006; Sammons and Speight 2008).

In the present study, therapist trainees were instructed to complete their CIJs by organizing their narratives around four question prompts after each therapy session. These CIJ prompts asked trainees to recall and identify the critical incident in the session, their reactions, the impact of the incident on their MC development, and the extent to which the incident led to their personal and/or therapy adjustment. Specifically, the questions included: “What has been the most critical or impactful event for you in this week’s session with your client? Please clearly describe the nature of the incident/experience”; b) “How would you describe your cognitive, affective, behavioral, and interpersonal reactions to this incident/experience?”; c) “In what ways might this incident or experience have affected you in terms of your cultural awareness, knowledge, skills, and relationship with your client?”; and d) “How has this incident/experience prompted you in making changes or adjustment in yourself and/or in your counseling work with the client in subsequent sessions?” In this study, we used trainees’ CIJs for their first 10 scheduled sessions with their individual refugee clients as the parameter for the analysis. Due to variabilities in the number of sessions attended by the clients (e.g., due to missed appointments) among trainees, a total of 121 journals were obtained from the trainees and used in the subsequent analyses.

Coding of the CIJs

To ensure the rigorousness and integrity of the data analysis process of trainees’ CIJ entries, a systematic procedure was implemented. First, a panel of 6 coders, which was comprised of 5 upper-level psychology major undergraduate students and one master’s-level clinical graduate student from the first author’s research lab was formed. The coders were blind to the research questions and the hypotheses of the project. Second, the first and the second author worked together to develop a standardized ‘coding rubric’ to help guide the process of analyzing the CIJs based on the MCT literature. The coding rubric was revised and refined several times with the input of the entire research team (8 individuals), including the first and the second authors as the faculty researchers and the 6 student coders, for depth, coverage, and clarity.

The coding rubric being employed to assess and code the trainees’ CIJs contains three ‘global domains’ of MCC grounded in the existing MCC frameworks (i.e., Sue et al. 1992; Arthur and Januszkowski 2001): MC Awareness; MC Therapy Skills; and MC Therapeutic Alliance. Within each of these three global domains, 3–4 refined thematic rating scale items were further specified to capture nuanced responses within each domain.

Thirdly, for the actual journal coding process, each coder independently and individually read, reviewed, and rated each journal entry assigned to them. To remove potential bias and expectation by the coders, the serial-temporal order of the trainees’ journal entries was ‘scrambled’ by yet another independent research assistant, and then the journal entries were randomly assigned to the coders for rating and analysis. Fourthly, the coders were instructed to first read through their assigned trainees’ journal entries. Then, they were instructed to rate each entry by adhering to the established coding rubric.

In the case of evaluating trainees’ ‘MC Therapy Skills’ particularly, the following three subdomains/rating items were specified: a) Theme A: “Therapist showing effort/attempt to adjust conventional therapy approach or her/his usual therapy approach to meet the need of the refugee client; b) Theme B: Therapist showing flexibility in her/his therapy and counseling approach in working with the refugee client; and c) Theme C: Therapist showing effort to manage misunderstandings, conflict, and negative feeling that arose from the client-therapist relationship during therapy. Using these rating thematic items on the coding rubric, the coders rated the extent to which each of these sub-themes were observed in trainees’ weekly CIJ entries by assigning ‘0’ for ‘No evidence,’ ‘1’ for ‘Some evidence,’ or ‘2’ for ‘Strong evidence’.

Lastly, coders met weekly in pairs to discuss and reconcile their ratings to arrive at a set of ‘resolved codings' – i.e., agreed upon rating scores for all journal entries across all domains and subthemes. The resolved codings for all the CIJs from all the coder pairs were then combined and collated for the data analyses process described below.

Analysis Procedure of the Coded CIJ Data

For the analysis, we selected to focus on analyzing the trainees’ journals for the ‘MC Therapy Skills’ domain only, due to publishing space constraints. Hence, to examine the ‘process’ or ‘trajectory’ of changes in trainees’ MC Therapy Skills based on coded CIJs, we implemented MLM as the analytic strategy (Tasca and Gallop 2009). This approach is a more sensitive and powerful test than traditional repeated measure methods because it allows researchers to simultaneously examine not only intraindividual change (i.e., the within-person patterns of variance) across multiple occasions, but also between-person differences in within-person change over time (Tasca and Gallop 2009). To the authors’ knowledge, the MLM analytic procedure has not previously been applied to examine the processes and the patterns of change in the development of therapist cultural competencies within a longitudinal research design. Yet, the methodological utility of multilevel modeling dovetails with the focus of the present research and the future direction of MC research as advocated by scholars and researchers (Sammons and Speight 2008; Smith et al. 2006).

Results

Analysis of Quantitative Measures: Pre- and Post-Tests

A series of paired-samples t-tests were first conducted with the pre- and the post-test scores on the overall scales and the subscales of the MCI and MCSE-RD. Means and standard deviations are reported in Table 1. The distributions of each pre-and post-test scores were examined, and all were deemed to be normal and no outliers were detected. Consistent with our hypothesis, a statistically significant increase in trainees’ pre- to post-test scores in the following variables were found: the overall MCI, t(12) = −4.136, p = .001, d = 1.147; the Multicultural Knowledge subscale, t(12) = −2.467, p = .030, d = .684; the Multicultural Awareness subscale, t(12) = −3.473, p = .005, d = .963; the Multicultural Counseling Relationship subscale, t(12) = −2.300, p = .040, d = .638; and the Multicultural Counseling Skill subscale, t(12) = −4.947, p < .001, d = 1.372. Furthermore, statistically significant increases were also found in trainees’ pre- to post-test scores in multicultural self-efficacy: the overall MCSE-RD, t(12) = −8.130, p < .001, d = 2.255; the Multicultural Intervention subscale, t(12) = −8.259, p < .001, d = 2.291; the Multicultural Assessment subscale, t(12) = −3.079, p = .010, d = .854; and the Multicultural Counseling Session Management subscale, t(12) = −6.672, p < .001, d = 1.850. Hence, across all the MC domains measured, the practicum trainees reported improvement, with effect size ranging from medium to large (i.e., d > .8), with the largest effects observed in the components of the MCSE-RD. Unequivocally, the pre- and the post-test results point to significant growth and improvement on all aspects of trainees’ MC competency and efficacy, based on trainees’ self-report quantitative MC measures (see Table 1).

Table 1 Means and standard deviations for each subscale
Full size table

Analysis of Trainees’ Change Pattern in MC Therapy Skills

Next, we examined the development of trainees’ MC Therapy Skills over the course of the practicum. We analyzed the inter-individual variability for each trainee. To reduce the complexity of trainees’ change trajectories over time, the rating scores by the coders for the journal entries for each trainee, described in the previous section, were aggregated across three phases of therapy (Hill 2005): a) The early phase (aggregate of sessions 1–3); b) The middle, working, phase (aggregate of sessions 4–7); and c) The late phase (aggregate of sessions 8–10). An inspection of the distribution of the slopes in Fig. 1 suggested that the patterns of changes across the aggregated session scores were quadratic. Three non-linear unconditional growth models were specified for each MC Therapy Skill sub-domain as shown below using Hierarchical Linear Modeling (HLM) v. 7.01. HLM is used to examine longitudinal data where observations are nested within participants or individuals, such as multiple number of data points collected from trainees over time in the current study. The longitudinal HLM model used in the current research is also known as growth curve model.

Fig. 1
figure 1

Spaghetti plot for trainees’ change patterns on the Multicultural Therapy Skills Domain – Theme A based on their coded journal ratings aggregated by three phases of therapy over 10 sessions. Individual trainees are denoted by their ‘id’ numbers and their change trajectories are represented by different pattern of dash lines

Full size image

Main HLM Findings

With these results, the fixed effects are denoted by scripts γ00,γ10, and γ20 and represent unbiased estimates of population parameters for the intercept, which is set at baseline, the linear slope, and the quadratic slope terms respectively for each skill theme. At baseline, Theme A and B (see the corresponding thematic descriptors presented under the “Coding of the CIJs” section above) differed significantly from 0 (.90 and 1.12 respectively), while Theme C did not differ significantly from 0 (.77). The linear and quadratic slope terms did not differ significantly from 0 for each of the three themes (.-.15, −.06, .45, for linear slope respectively, and .02,,001, and − .17, for the quadratic slope respectively) reflecting a lack of mean-level change in these MC Therapy Skills themes over the three aggregated session blocks.

The random effects denoted by the terms υ0J, υ1J, υ2J are estimates of between-therapist variability on the intercept, linear slope, and quadratic slope for each skill theme, respectively. As predicted, there was significant between-therapist variability on the intercept υ01, linear slope υ11, and quadratic slope υ21 terms for the targeted MC Therapy Skills theme (see Table 2). Trainees’ growth curve for therapy skills reflected a non-linear pattern with a peak in these domains during the working phase of therapy. Furthermore, as exhibited in Fig. 1, among the 14 trainees, their reported MC Therapy Skills at the early phase of their therapy work with refugee clients was more divergent (i.e., more scattered points; see the left oval circle highlighted in Fig. 1) than towards the late phase of the 10 therapy sessions (i.e., more convergent and less scattered points; see the right circle in Fig. 1). Moreover, a close visual inspection of Fig. 1 across the 14 trainees suggests an approximate ‘spike’ pattern to trainees’ progress in their MC Therapy Skills over the therapy sessions, going from low-to-high in the early phase of therapy, peaking at the middle or working phase, and then tapering off near the late phase of the therapy process.

Table 2 Summary of unstandardized RML estimates for HLM growth model of the Multicultural (MC) Therapy Skills domain for themes A, B, and C
Full size table

Illustrative Narratives from Trainees’ CIJs

To help better contextualize and illustrate the observed pre- and post-test changes and the emerged trainee growth curves reported above, in this section we attempt to supplement ‘rich and thick description’ of trainees’ internal, experiential processes in exploring and reflecting on their MC development, based on selective journal narratives of trainees. The following entries were selected as poignant examples and illustrations of the trainees’ process of change, as they struggled with and progressed through their development of MC therapy skills over the course of the therapy work with their refugee clients. Note that the bracketed entries in the subsequent clients’ quotes are supplemented by the researchers to ensure clarity.

Following her first session working with an Iraqi woman in her 50s who was a survivor of the bombing of a Christian church in Iraq, a female therapist trainee of European Canadian descent identified her first-hand, heightened awareness of cross-cultural communication barriers with her client, even with the aid of an Arabic-speaking language interpreter:

This is the first time I feel that I have a real concept of the barrier that language and culture presents, beyond the intellectual understanding of it. Trying to explain psychological questions in plain straightforward language [through interpretation] was much more challenging than I expected. I also felt that this must seem a strange experience for her [the client] as well.

The same therapy session had elicited the therapist’s cultural empathy towards the client as a Christian minority living in a predominately Muslim country to the forefront, as she wrote:

Understanding her [the client’s] experience from her point of view was difficult coming from a different culture, but also a society in which I have never really lived in fear. I have never experienced even a fraction of what happened in that Church, and seeing such things on the news is so much more distant. Furthermore, looking at the grieving community left behind after this incident [the Church bombing] was interesting for me in terms of deepening my understanding of being a Christian in a fairly extreme Islamic country.

     In the 6th session, the therapist identified a critical incident that pertained to her Iraqi client’s refusal to talk about her trauma memories in the session due to the client’s cultural beliefs. In this working phase of the therapy, however, the trainee’s journal clearly revealed a more nuanced and culturally-informed approach in framing and approaching her therapy interview with the client. This development coincides with the ‘peaking’ or the elevated level of MC Therapy Skills in the trainees’ growth curves/change trajectories evident in Fig. 1. The therapist recounted:

It was [an] interesting experience because it brought to mind the training we received in the previous multicultural class….I did make sure not to communicate to my client that this belief was unacceptable or [needed to] change, which I felt was important. I wanted the relationship with my client to remain strong despite differences in beliefs; however, I also recognized the need to seek supervision and information in order to not let this belief prevent me from being helpful to the client. This experience indicated a strong need to balance the beliefs of the client with doing what will be helpful for the client. I am willing to find a compromise that will be both beneficial to the client, but that will also make her feel comfortable.

     In another example, a female trainee of Jewish Canadian descent was assigned to work with a 31-year-old refugee woman from Bhutan, who had lived in a refugee camp for many years prior to being resettled in Canada. The client reported multiple physical pains and emotional distresses since her arrival in Canada that were without apparent, identifiable causes. The quote below from this trainee’s first post-session journal entry epitomizes the common, initial challenge that many therapists and clinicians faced in cross-cultural counseling interaction, due to verbal and non-verbal barriers. Despite the aid of a Nepali-speaking interpreter, the therapist’s inability to motivate her Bhutanese refugee client to speak in the session had inadvertently led up to the trainee’s self-doubt about her clinical and cultural skills and competence:

Regarding myself, I felt a bit incompetent and frustrated that I could not get her to open up more about her experiences. Because I was surprised that I was not getting the information I was expected from the client, I am worried that my reactions were insensitive and that the client and I are not on the same page regarding the purpose of therapy.

However, this impasse, as the critical incident, in turn forced the trainee to seek adaptation in her therapeutic conceptualization and approaches with the client as she anticipated the next session. The trainee reflected:

[In the next session] I will view the client more from the perspective of her strengths rather than assuming that she has experienced a traumatic event that needed to be talked about in therapy. Focusing on the client’s empowerment to cope adaptively with her everyday stressors may be one way that I can be more in line with the client’s current needs and concerns.

As follows, in session 3 the therapist identified a crucial turning point in her therapy process and relationship with her Bhutanese client. With the encouragement of the practicum supervisor, the trainee shifted her perspective by adapting the ‘acculturation framework’ and a strength-based schema in reconceptualizing her refugee client’s presenting concerns, as opposed to her original trauma- and victimization- oriented perspective. This ‘ah ha’ insight culminated in her growing confidence as a therapist, which was evident in her post-session journal:

Through the session it became quite clear to me that these issues (client’s isolation and limited mobility in a new city) can all be combined to some degree under the heading of acculturative stress….. [With this framework] I found that I was able to ask better questions that the client was happy to respond to. I was finally getting somewhere in understanding where the client is coming from and it appeared she was becoming more comfortable speaking to me about the important things in her life, including her vulnerabilities…It has increased my knowledge about acculturative stress in general and how it may manifest in different people’s lives.

     Here, we see that with the support of culturally-informed supervision the trainee’s direct therapy experience with her refugee client within this practicum provided her with an exceptional opportunity to integrate her cultural knowledge (i.e., acculturation and migration) into actual multicultural therapy skills and intervention. A similar ‘shift in perspective’ was observed among several trainees in the study, whereby a re-examination of Western psychotherapy assumptions was invoked as trainees interacted with their refugee clients. This was evident in the case of a female therapist trainee of Polish immigrant background who treated a Karen-speaking refugee woman in her late 50s, originating from Burma. In their third session, the trainee described feeling deeply moved by seeing the client’s family photographs, which the client spontaneously brought to the session – they included photos of her sister, her sister’s 3-year-old son, and husband, all of whom tragically died a year prior. The trainee reflected on this critical incident:

She [the client] became tearful as she wondered aloud how they died, focusing especially on the innocent child. There was a long stretch of very little interpretation, where eye contact and nonverbal communication served to express sadness from the client and compassion from me. … I felt very honored that she wanted me to “meet” her family. It made me feel as though our relationship was growing and that she trusted me. I noticed that our body language changed - from sitting back to leaning close to each other.

The evidence of deep empathy and non-verbal communication was perceptible in this critical client-therapist exchange. In response, the trainee poignantly described the way in which she incorporated the client’s cultural worldview in adapting and broadened her ‘typical’ repertoire of clinical skills and therapy stance, such as by her own self disclosure:

…., given her collectivist cultural roots, in order for me to know the client, it is necessary for me to know her in her community context. For individuals from other [non-Western] cultures, to tell about themselves implies having to share where they come from – who they are is who their family is I suppose. ….I felt as though the client was reassured that I was interested in her and where she came from, and that I had responded to her sharing of her family stories in an acceptable manner for her to trust me. She even asked me about my own life - if I had children, etc. - and I felt comfortable sharing with her some things – things I would not have shared with a Western client, I don’t think. …. I think it was a necessary part of the “give and take” dynamic that had developed in our relationship. To have been vague or avoided the questions would have been rude and equivalent to rejection of the relationship that had formed.

     Part of this expanding MC therapeutic skills and repertoire for the trainee appeared to stem from a critical evaluation of Western assumptions in therapy, in which the trainee was deeply entrenched in through previous training. In concluding her third session CIJ, the trainee revealed a deep sense of cultural humility as she reflected:

I think that this has opened my eyes a bit more to the idea that the boundaries defined by clinical psychology as I learned it (in a Western context) may not be as absolute when working with those from other cultures… Perhaps Western clients feel a certain element of comfort with a psychologist who is the consummate professional – caring, soft, compassionate, but definitely observant of professional conduct. However, such an “antiseptic” stance may alienate individuals from other cultures. For them (client with a collectivistic orientation), the idea of a helper is someone who knows them and who is somewhat known to them. As such, it is important to not be dismissive of an individual’s attempts to forge their own connection with you. If that means disclosing a bit about yourself, so be it. If that means taking an extra 10 minutes at the start of the session for social “chit chat” to respect a client’s idea of social etiquette, so be it. There is certainly a volume of literature on mechanisms and process of successful therapy. However, these mechanisms, approaches and processes only can work if the client is engaged, and to get a client from a different culture to that point may require one have an open mind with which to forge a good working alliance.

     These above excerpts provide narrative examples from trainees’ CIJs that are commensurate to Level 1 and Level 2 of the coded journal rating procedure by the coders, described in previous sections. These qualitative, narrative responses highlight the breadth, depth, and richness of trainees’ learning impacts and developmental processes through their direct and intimate therapeutic encounters with their refugee clients.

Discussion

The current research represents one of the first empirical studies within the MCT literature to comprehensively discern and evaluate trainees’ outcomes and change processes associated with their MC development within a culturally-focused clinical practicum, at a nuanced, micro-process counseling level. The key findings and the implications of the study are expounded below.

Quantitative Pre- and Post-Test Outcomes

First of all, the quantitative, pre- and post-test results of the current research lend support to the positive learning gains trainees acquired from this multicultural practicum – accordingly, it addresses the study’s Research Question #1. As hypothesized, trainees who completed the practicum reported statistically significant increases in all domains of their MC competence (as measured by the MCI) and in their MC self-efficacy (as measured by the MCSD-RD), with medium to large effect sizes (i.e., d > .8). In a recent meta-analytical study of 24 MC education programs, Smith and Trimble (2016) found the average weighted effect size across these studies between pre- and post-test scores to be d = .95. In comparison, in the current study the score improvement for trainees in this MC practicum was significantly larger, with d = 1.15 for overall MCI and d = 2.3 for overall MCSE-RD, respectively. Consequently, in contrast with the findings of previous MC education studies, the magnitude of change and learning gain for therapist trainees in the present study is substantial and striking.

These results lend support for previous cross-sectional MCT research that highlighted the critical value of hands-on MC practica and direct counseling/therapy/clinical experiences with culturally diverse populations (e.g., Arthur and Januszkowski 2001; Dickson and Jepsen 2007). Additionally, the study’s findings speak to the pedagogical advantage of MC practicum particularly for facilitating trainees’ actual practice and acquisition of culturally-informed assessment, intervention, and management of session, as attested by the trainees’ significant increase in the MCSD-RD scores.

Qualitative Pattern and Process of Change

Secondly, the findings of the study also helps to shed light and further our understanding on how learning changes occur for trainees undergoing MC training and education across time - that is, the ‘process’ and the ‘pattern’ of acquiring MCC – Research Question #2. Accordingly, the MLM growth curve analysis of trainees’ coded CIJs over 10 therapy sessions with their refugee clients points to a non-linear oscillation pattern in their learning and demonstration of MC Therapy Skills. Two notable characteristics stood out.

First, as shown in Fig. 1 there was significant divergence in trainees’ reported change patterns on MC Therapy Skills at the initial, early phase, but followed by an increasing convergence at the later phase of the 10 therapy sessions with refugee clients. This initial divergence could be explained by the inter-individual differences among the 14 trainees coupled with random variabilities among the refugee clients (e.g., clients’ diversities in terms of therapy needs, presenting concerns, familiarity with and motivation for therapy, cultural backgrounds and demographics, etc.). Based on the ratings of trainees’ CIJs, it seems that with increased therapy practice and continuous clinical supervision trainees reported a more flexible use of their therapy approaches with refugee clients. In turn, they perceived and reported gradual improvement on and assimilation of new clinical skills into their therapy repertoire over time.

Second, across the 14 trainees a rough ‘spike’ or ‘peaking pattern was also observed in trainees’ growth curve trajectories of MC Therapy Skills, at approximately between the 4th and the 7th sessions over the 10-session period, which coincided with the working phase of therapy (see Fig. 1). The relatively low-level MC Therapy Skills demonstrated by the trainees during the early sessions might be explained by trainees’ initial anxiety, uncertainty, and preoccupation over the perceived cultural distance and difference between themselves and their refugee clients. In fact, these initial tentativeness and misgivings on the part of the trainees were well reflected in trainees’ journal narrative excerpts described in the previous section.

This aforementioned phase was then followed by trajectories of growth in trainees’ reported MC Therapy Skills, which showed a peak of MC competence during the middle or working phase of the therapy with refugee clients (see Fig. 1 for the ‘peaking’ pattern in the trajectories). Beyond this middle phase, however, the trainees’ growth curves slid to lower levels of salience over the late phase of therapy. It is possible that, during the later phase of the MCT therapy process, the initial ‘newness’ or ‘saliency’ of cultural issues and distance between refugee clients and their therapists had receded into the ‘background’ of the therapy process as the client-therapist therapeutic alliance continued to grow and strengthen. Therefore, the ‘culture aspect’ of the therapy no longer occupied the therapist’s attention. It is equally possible that what was being assessed and reported through the CIJs, especially during the earlier phases of the therapy process, reflected trainees’ initial, new and immediate awareness and understanding about working with refugee clients. These cultural awareness and knowledge elements that emerged early on might have naturally plateaued or leveled off at the late phase of the therapy process.

As follows, it can be reasoned that in an effective culture-integrated and informed MC counseling and therapy practice, clinicians would be able to conduct counseling and therapy and to establish positive therapeutic relationships with their culturally-diverse clients in much the same way they would with any client in conventional counseling/therapy, but remain vigilant to cultural concerns and issues as they arise. Indeed, two decades ago Fischer et al. (1998) conceptualized and argued that culturally competent counseling and therapy should embody therapeutic common factors that underlie all universal healing practices (i.e., conventional therapy), but be supplemented with counselors’ acquired cultural knowledge of the client and the therapy process. This current finding and interpretation align with Fisher et al.’s (1998) observation and assertion.

Furthermore, the observed curvilinear property of trainees’ trajectories in their development and demonstration of MC Therapy Skills can be conceptualized and understood from the perspective of the Dynamic Systems Theory for psychotherapy process (Hayes and Strauss 1998). According to the Dynamic Systems Theory, “psychological growth is a lifelong process that is characterized by periods of stability and instability” (Hayes and Strauss 1998, p. 940). Similarly, as seen in Fig. 1 and Table 2, trainees’ rating scores on the 3 themes of the MC Therapy Skills Domain fluctuated across sessions in a non-incremental and non-linear fashion, before they settled into more stable levels towards the end of the 10 sessions. Once again, trainees’ initial unsettling experiences were corroborated by their early CIJ entries presented in the trainees' narrative section. For example, these initial tensions were observed in the first trainee’s profound struggle with responding to her refugee client and the second trainee’s feelings of incompetence and self-doubt for not being able to help her refugee client to open up in the session. However, as predicted by the Dynamic Systems Theory, this process was then followed by a period of reorganization or change in trainees’ schemas, as they attempted to incorporate newly gained clinical and cultural learning and insights into their therapy work with refugees, either via direct therapy experience and/or through input received from clinical supervision. Such a cultural and therapy learning process is likely to undergo reiteration several times before it is eventually consolidated and assimilated into counselor and therapist trainees’ clinical repertoire and competence.

Qualitative Narratives of Trainees

Finally, the trainees’ post-session CIJ narratives further afforded the current study a unique lens through which to discern the nuanced quality of trainees’ experiences and processes in developing cultural and clinical competence at a micro-process counseling level. As illustrated by excerpts from the journal narratives of the trainees in the previous section, profound and invaluable challenges and impacts at the cognitive, affective, behavioral, and interpersonal levels were reported by the trainees as the result of participating in this multicultural practicum and interacting directly with refugee clients. Trainees’ reports of cultural humility and critical reflections was observable through their journal accounts. In this sense, the CIJ has once again proven to be an empirically useful research method to help reveal and document trainees’ lived experiences as they worked with their refugee clients, with ‘thick and rich’ descriptions. By extension, the current study lends further support for the use of mixed-methods research and qualitative methods in MCT research (e.g., Coleman 2006; Sammons and Speight 2008).

Limitations

The results of the current research need to be viewed with caution given several limitations. First, despite having collected extensive quantitative and qualitative data from the 14 participants in the present study, the sample size was small. Therefore, while the findings of the study shed new and unique light on the learning impacts and processes of practicum-based, experiential MCT on counselor and therapist trainees, the results should be regarded as preliminary. The generalizability of the study’s findings beyond the present sample cannot be assumed without further replication with a larger sample of trainees. Second, it is important to recognize that the assessments of trainees’ MCC in this study, both quantitatively and qualitatively, were based on the trainees’ self reports - that is, their personal perceptions and reporting of the changes occurring during their therapy work with refugee clients. Therefore, trainees’ cultural competencies as investigated in the present study were established on their ‘perceived’ competencies; the evaluation of trainees’ ‘actual,’ demonstrable MCC is beyond the scope of this current research. Third, the gender imbalance among the 14 participating trainees in the study is conspicuous, as there was only one male trainee in the sample – simply a default of the gender characteristics of these three cohort years of clinical students. Clearly, the study’s results need to be cautiously considered in view of this female-skewed sample. Finally, given the naturalistic setting of this MC psychotherapy practicum, the refugee client characteristics and backgrounds, such as the types and the severity of their concerns, could not be controlled. It is unclear to what extent these between-client variabilities might or might not have affected trainees’ overall therapy experiences with their refugee clients, and, in turn, their development of perceived cultural competence and skills in working with their respective clients. This answer awaits further research.

Implications for Future Multicultural Counseling and Therapy Training and Research

MC Training and Practice

The results of the study hold a number of important practical and research implications for MC training in counselor education, counseling psychology, and clinical psychology programs in Canada, the U.S. and beyond. Firstly, the findings of the present study provide initial, empirical evidence in support of the efficacy and the usefulness of multicultural therapy practica for clinical and counseling trainees, as has been advocated by previous MCT scholars (e.g., Abreu et al. 2000; Dickson and Jepsen 2007). As follows, the current findings should provide graduate training programs in counseling, psychology, and other mental health fields with added, evidence-based, confidence in designing and incorporating supervised MC practicum into their training and education. In particular, a well-implemented MC practicum can help to bridge the critical gap between cognitive- and didactic - based learning of multicultural knowledge and implementation of MCC in actual counseling and clinical practice with diverse client populations (Kuo 2012; Smith and Trimble 2016). As demonstrated in this study, this practicum training experience has significantly helped improve trainees’ scores on measures of MC intervention, assessment, and session management between pre- and post-practicum.

Secondly, professional counseling and psychology graduate programs might consider tailoring their MC therapy practica to provide counseling, therapy, and psychological services to specific underserviced culturally-diverse populations in their local communities, such as racially and ethnically diverse youth, adults, and families, immigrant individuals and families, migrant workers, international students, sexual minorities, religiously diverse populations, etc. Through such a service-based MC practicum and practice, graduate students and trainees will be afforded the sorely-needed opportunity to practice skills in addressing issues of social inequality and injustice and to engage in social and resource advocacy for disadvantaged and underserviced client populations (Pieterse et al. 2009), such as refugee survivors as in the current study. Thirdly, the present refugee-serving MC therapy practicum provides one tangible example/model of university-community collaboration in advancing MCT through community-focused, service-based learning for counseling and clinical graduate students, as advocated by both Canadian (Arthur and Collins 2010; Kuo and Arcuri 2014) and U.S. MC scholars (APA 2010; Dickson and Jepsen 2007). Therefore, counseling and clinical psychology graduate programs might wish to consider and explore similar model of university-community partnership, to augment their cultural and diversity training.

MC Research

In terms of research implications, future MCT research would greatly benefit from continuous efforts to systematically assess, evaluate, and analyze the dynamic processes associated with students’ and trainees’ acquisition of MC competence, self-efficacy and therapy skills in actual cross-cultural counseling/clinical/therapy contexts involving real multicultural clients. This line of research would effectively contribute to the shaping and the formulation of ‘best practices’ for MCT and empirically-supported MC education and intervention. Methodologically speaking, the current study’s application of a longitudinal, mixed-method design and the use of MLM growth curve analysis are both novel and unique in examining trainees’ multicultural development. Assessing trainees’ development of MCC with the combined use of quantitative and qualitative data, not only helped shed light on the effects of MCT intervention on the trainees in a complementary way, but also elucidated the dynamic process through which trainees struggle with and develop their cultural competence across time. Therefore, the present study might serve as an initial framework, a basis for which future mix-methods MCT research can replicate and/or build on.

Finally, it would also be profitable for future MCT studies to examine the perceptions of culturally-diverse clients who are the recipients of culture-based counseling and therapy interventions, such as the refugee clients in the current study. For obvious reasons, this type of client-based MCT research would be invaluable for verifying the consistency (or lack thereof) between the views of the counselor/therapist and that of the client on what constitutes culturally responsive, useful, and competent counseling and therapy interventions. The information will help verify the relationship between trainees’ reported or ‘perceived’ MCC and ‘actual’ MCC as indicated by counseling outcomes and/or clients’ evaluations. Correspondingly, this information can help guide and inform more effective MCT and provision of counseling, therapy and psychological services to the receiving client population.

Conclusion

In concluding their recent comprehensive review and evaluation of accumulative MC literature, Smith and Trimble (2016) asserted: “In short, an obvious gap exists between multicultural psychology as practiced and as frequently defined. It is time to bridge that gap, starting with an evaluation of reality. Taking inventory of what multicultural psychology ‘is’ in the real world can help determine where we are relative to what it ‘ought’ to be, with the aim of achieving the envisioned ‘ought’”. Considering this call, this present study provides an initial attempt in support of this stated vision. The importance of practically-relevant, clinically-situated, and evidence-based MCT for clinical and counseling trainees for highly multicultural countries and societies, such as Canada and the U.S., cannot be overstated. We hope that the present research and its preliminary findings may encourage other MC scholars, practitioners, and educators, as well as counseling, psychology, and other mental health training programs to consider incorporating and implementing similar hands-on, service-based MC practica into their cultural-clinical training. Additionally, we hope to see continuous MCT research with sophisticated and comprehensive designs to pursue a deeper and more nuanced understanding of the processes of learning, change, and development for clinical and counseling trainees undergoing MCT and education.