FOUR

Transcultural Self-Efficacy Tool (TSET)

with contributions from Fatma Başalan İz, Ayla Baylk Temel, Jing Chen, Mark Fridline, Faye J. Grund, Margaret M. Halter, Maria Malliarou, Carol Reece, Pavlos Sarafis, Sharon See, and Lisa Young

Choice of a valid and reliable instrument will help decrease the limitations associated with measurement error and increase the confidence with which researchers interpret data.

SNAPSHOT SCENARIO

Administrator: Accountability for validly assessing, measuring, evaluating, and documenting outcomes prior to and after cultural competence education is being emphasized worldwide in academic and practice settings across health care disciplines.

Doctoral student: In my doctoral courses, I’ve been taking a measurement course and realize the complexity involved in creating a valid and reliable instrument that measures what it purports to measure.

Professor Numbers: Yes, a tool or questionnaire must also be comprehensive to cover the specified content area and validated by qualified experts in content, methodology, and psychometrics. Instrument design and psychometric evaluation is a detailed, time-consuming process. For example, scaling and ordering of subscales, item format and sequencing, and response options must be based upon an underlying conceptual framework that is also substantiated by statistical standard practices to decrease measurement error.

Nurse Map: An underlying conceptual model should be used as a map to guide educational interventions whether it’s with employee educational practice settings, continuing education (CE) programs through professional organizations, or academic settings. By planning evaluation during the design of educational interventions, educators will have evidence to continue with or modify teaching–learning strategies. Selecting a valid, reliable tool that connects to an underlying conceptual framework will enhance interpretation of findings and contribute to the expanding literature concerning cultural competence education, the selected evaluation tool, and the model.

Researcher: During my literature review, I noted that the Transcultural Self-Efficacy Tool, commonly referred to as the TSET, regularly demonstrated excellent psychometric properties consistent with its underlying conceptual model called the Cultural Competence and Confidence model or CCC model. The TSET has various versions, such as a multidisciplinary healthcare provider version, and it’s been translated and psychometrically tested in several languages. It’s been applied with small and large samples of students and with practicing health professionals in various entry-level and advanced practice settings and demonstrated statistically significant changes in cognitive, practical, and affective learning dimensions following educational interventions. Some researchers also examined TSET scores in relation to patient satisfaction scores on national questionnaires like HCAPS (Hospital Consumer Assessment of Healthcare Providers and Systems) used for hospital reimbursement or for train-the-trainer programs. Some researchers examined TSET data via expanded types of data analyses, statistical techniques, and focus on areas of research demanding attention in cultural competence education such as end-of-life care and LGBTQ (lesbian, gay, bisexual, transgender, and questioning) workplace and patient-care issues.

Nurse Ponder: It sounds like the TSET has many desirable psychometric properties and attributes already with the added potential for application in numerous settings and for issues needing much attention. I’m wondering what led to the design of the TSET. It’s good to know a questionnaire has current and future applications, but I’m curious about where it all began.

Professor J: During my doctoral dissertation study concerning first semester nontraditional associate degree student achievement and retention, individual item review of perceived self-efficacy for 60 select nursing skills suggested that students were least confident about learning specific communication skills as compared with skills in other categories. Eight of the 10 communication items received responses ranking in the 30th percentile, or least confident. Less confidence was reported for interviewing a client about “financial concerns,” “religious practices and beliefs,” and “ethnic food preferences.” These were the only items that dealt with cultural issues in terms of socioeconomic status (class), religion, and ethnicity.

So, I thought, “Why are students least confident about communication? Why were students less confident about items related to cultural issues? How would students have responded to items that further delineated the various dimensions of culture?” According to the renowned psychologist Dr. Albert Bandura, individuals with low confidence are at greater risk for task avoidance and decreased commitment. If students are avoiding tasks or are less committed to tasks associated with culture, then how can cultural assessments, culture-specific nursing care, culturally congruent care, and cultural competence be achieved? Cultural assessments must begin with effective transcultural communication, which requires awareness, sensitivity, knowledge, and skills. If students are less confident about general communication skills, how will cultural assessments be performed? Or will they be performed? The obvious gaps and lowered confidence raised the important question, “What teaching interventions are needed to promote culturally congruent care?”

I was then interested in learning more about students’ self-efficacy perceptions concerning specific transcultural nursing skills necessary for developing cultural competence. Specifically, my area of interest was to develop a composite of students’ needs, values, attitudes, and skills related to transcultural nursing care and the assessment of their changes (outcomes) over time. Assessing students’ needs, strengths, weaknesses, and perceptions would be the necessary precursor to the design of any teaching interventions. I believed that the initial and ongoing assessment of students’ self-efficacy perceptions (confidence) concerning culture care of diverse individuals would be a valuable component in transcultural nursing education and cultural competence. I wanted answers to the following questions:

These questions prompted the design of the TSET.

The beginning part of the chapter discusses the exploration, measurement, and evaluation of confidence through the use of the TSET (Jeffreys Toolkit 2016 Item 1). Major components, features, and psychometric properties (reliability and validity) of the TSET are highlighted. Tables summarize vital instrument information. TSET Research Exhibits, application strategies, and the “Educator-in-Action” vignette provide a user-friendly and reader-friendly approach for immediate application in a variety of settings. Additional assessment measures, applied uses, and evaluation strategies conclude the chapter.

TSET DESCRIPTION

Major Components and Features

The TSET was originally designed to measure and evaluate students’ confidence for performing general transcultural nursing skills among diverse populations. A generalist approach, focusing on general transcultural nursing skills, was considered most appropriate for learners without previous formal education and background in transcultural nursing, especially those who would care for clients of many different cultural backgrounds. The generalist approach emphasizes broad transcultural nursing principles, concepts, theories, and research study findings to care for clients of many different cultures (1). In contrast, a specialist approach is most appropriate for learners who have mastered general transcultural nursing skills developed through formalized educational and interactive experiences. A specialist approach aims to prepare an individual as a “specialist” in one or more select cultural groups, requiring a series of specialized transcultural courses and concentrated fieldwork (1–3).

To address the question, “How can TSE perceptions be explored, measured, and evaluated validly among novice and advanced students?” a literature review was conducted, focusing on transcultural nursing, cultural issues, self-efficacy, learning theories, and instrumentation. The majority of self-efficacy tools described in the nursing literature pertained to client education. One self-efficacy scale, the Cultural Self-Efficacy Scale (4), was designed to measure cultural self-efficacy perceptions of community health nurses within three specific client populations (Black, Latino, and Southeast Asian). No self-efficacy tool had been specifically designed that would measure student’s perceived self-efficacy for performing general transcultural nursing skills among clients of different cultures. Furthermore, no questionnaire had been designed for culturally diverse nursing students who must learn to care for many different groups of culturally diverse clients.

Therefore, the design of a new instrument was necessary. The author’s decision to create a new instrument was further substantiated by self-efficacy theory that supports the design of detailed questionnaires (focused on specific tasks or skills) within the desired domain of inquiry (5–7). The process of designing a new instrument included item development, item sequence, subscale sequence, expert content review, expert psychometric review, revised draft, pre-test, minor revisions, and a second pre-test (8).

Based on the literature and the results of a two-phase evaluation study (8), the 83-item TSET (9) contains three subscales presented in the following sequence: Cognitive (25 items), Practical (28 items), and Affective (30 items). Separate subscales were created for two main reasons: (a) the most comprehensive learning includes coordinated learning in the cognitive, practical (psychomotor), and affective domains; and (b) self-efficacy theory purports that different dimensions within a specific domain of inquiry require separate subscales for accurate measurement and evaluation (5, 6). The Cognitive Subscale asks respondents to rate their confidence about their knowledge concerning the ways cultural factors may influence nursing care. The Practical Subscale asks respondents to rate their confidence for interviewing clients of different cultural backgrounds to learn about their values and beliefs; 28 culture-related interview topics are presented as items. Attitudes, values, and beliefs are addressed in the Affective Subscale. Empirical evidence concerning TSET subscale sequence presentation supported presentation of the Cognitive Subscale first and Affective Subscale last to prevent anchoring effects (8).

Consistent with self-efficacy measurement tool guidelines (5, 6, 10), items are close-ended and positively phrased; a 10-point rating scale from 1 (not confident) to 10 (totally confident) is used. In addition, TSET pilot testing of a 10-point scale versus a 6-point rating scale indicated that the 10-point rating scale was more valid, thus further substantiating Bandura’s recommendations and the underlying conceptual framework (8). Considering that most people can relate to a 10-point scale in daily life, such as the decimal and financial system and pain scales, this is not surprising. Approximate completion time is 20 minutes. Table 4.1 depicts the major components and features with the underlying rationale for each component and feature. (Details about the initial TSET design and testing may be found in Jeffreys and Smodlaka [8].)

PSYCHOMETRIC PROPERTIES: VALIDITY

A series of four studies were initially conducted to estimate the psychometric properties of the TSET. Psychometric properties broadly refer to the results obtained from specific statistical tests for estimating instrument validity and reliability. Results provide estimates for the validity and the reliability of the instrument. Details about the initial validity and reliability tests and their results are outlined in the following sections. Subsequent validity and reliability estimates from subsequent studies by the TSET author and other researchers are highlighted. In addition, TSET reliability, validity, and applications are integrated throughout the chapter’s narrative text, in the TSET Research Exhibits in various chapters, and in the Appendix.

Validity is concerned with the degree to which an instrument measures what it is supposed to measure. For the TSET, a general question was asked: Does the TSET accurately measure what it is supposed to measure? In order to answer this broad question, three more specific questions were asked, differentiating between content validity, construct validity, and criterion-related validity.

TABLE 4.1 Transcultural Self-Efficacy Tool (TSET): Major Components, Features, and Rationale

¹Bloom, B. S., Enlgehart, M. D., Furst, E. J., Hill, W. H., & Krathwohl, D. R. (1956). Taxonomy of educational objectives: Handbook I, cognitive domain. New York, NY: McKay.

²Harrow, A.J. (1972). A Taxonomy of the psychomotor domain. New York, NY: McKay.

³Krathwohl, D.R., Bloom, B.S., & Masia, B. (Eds.). (1964). Taxonomy of educational objectives: Handbook II, affective domain. New York, NY: McKay.

⁴Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215.

⁵Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122–145.

⁶Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

⁷Bandura, A. (1989). Regulation of cognitive processes through perceived self-efficacy. Developmental Psychology, 25(5), 729–735.

⁸Sudman, S., & Bradburn, N. M. (1991). Asking questions: The definitive guide to questionnaire design. San Francisco, CA: Jossey-Bass.

⁹Jeffreys, M. R., & Smodlaka, I. (1996). Steps of the instrument-design process: An illustrative approach for nurse educators. Nurse Educator, 21(6), 47–52. [Erratum, 1997, 22(1), 49].

¹⁰Cervone, D., & Peake, P. K. (1986). Anchoring, efficacy, and action: The influence of judgmental heuristics on self-efficacy judgments and behavior. Journal of Personality and Social Psychology, 50, 492–501.

Content Validity

Content validity is concerned with whether the instrument and its items are representative of the desired content area and is best assessed by content experts (11, 12). For the TSET, the question posed was: Is the TSET (and its items) representative of the desired content area? The identified content area targeted the transcultural nursing skills necessary for the transcultural nurse generalist who may be caring for clients of many different cultural backgrounds. Additionally, items need to represent this content domain and be readable and appropriate for novice undergraduate nursing students. Information about the intended purpose, desired content area, and self-efficacy theory was distributed to the content experts. Content validity was established by six doctoral-prepared nurses certified in transcultural nursing (8).

Construct Validity

Assessment of construct validity evaluates the degree to which a tool measures the construct being studied. Construct validation attempts to validate the tool’s underlying theoretical concepts and proposed relationships between the concepts (12–14). An instrument whose performance is consistent with the underlying conceptual expectations demonstrates adequate construct validity (15). To answer the question, “To what degree does the TSET measure TSE?,” a contrasted group approach and a factor analysis were conducted.

Contrasted Group Approach

Two initial studies used the contrasted group approach for estimating construct validity and addressed the question, “Are mean scores on the TSET significantly different between two contrasted groups (novice students and advanced students)?” Consistent findings from the initial longitudinal study and cross-sectional study supported that the TSET detected differences in TSE perceptions within groups and between groups on all of the subscales (16–18). Several underlying theoretical assumptions were supported in these studies, namely:

Subsequent studies using contrasting groups of learners in the academic setting (students) and clinical agency setting (nurses, other health professionals, and employees) also demonstrated significant differences prior to and after cultural competence educational intervention, thus lending continued support for the TSET’s construct validity (19–35). See other studies in TSET Research Exhibits in this chapter; Chapters 6, 7, 8, 9, 13, 16; and the Appendix.

Factor Analysis: Initial Approach

Next, factor analyses evaluated the degree to which individual items clustered around one or more conceptual dimensions. Items that cluster together (to become a “factor”) should make sense conceptually, thus supporting the underlying conceptual framework and attesting to the construct validity of the instrument. The factor analysis studies (36–37) broadly explored the question: “What is the factor composition of the TSET?”

Several different statistical procedures may be employed in factor analysis. Each procedure contains certain errors or assumptions; therefore, all factor analysis options were appraised. Review of statistical theory and collaboration with a psychometric expert guided the decision-making process. The process employed in the initial factor analysis study is presented first, followed by the subsequent study utilizing statistical procedures and techniques not yet developed in 1995. The first step of factor analysis begins with a matrix of correlation coefficients between items (13, 38). Items should contribute uniquely and satisfactorily to the instrument, avoiding any redundancy between items. Generally, items that correlate below 0.30 are not sufficiently related and therefore do not contribute to the construct’s measurement; items that correlate above 0.70 are considered redundant and unnecessary (39). The question “Do all items on the TSET contribute uniquely and sufficiently to the TSE construct?” was evaluated via an inter-item correlation matrix. All items correlated between 0.30 and 0.70; therefore, it was assumed that all TSET items contributed uniquely and sufficiently to the TSE construct (16, 37).

The next issue pertained to the assessment of intercorrelations between the three subscales. Within self-efficacy theory, separate subscales should be designed to measure the distinct dimensions within a domain of inquiry. In the assessment of self-efficacy, intercorrelations between subscales corroborate that each subscale measures different dimensions within the domain (6). To answer the question “Are the Cognitive, Practical, and Affective Subscales correlated with each other?” subscale scores were computed. Intercorrelations between subscales were statistically significant and ranged from 0.53 (Cognitive and Affective) to 0.62 (Cognitive and Practical) and 0.68 (Practical and Affective). These results helped validate the following assumptions in the underlying framework.

Factor analysis studies are sometimes used for determining parsimony through item deletion, meaning that items that do not load on a factor are deleted to make the instrument shorter. Using conventional procedures for factor analysis as reported in the literature prior to 1995 (40–42), a Principal Component Analysis with the varimax rotation yielded a nine-factor structure for the study sample (16, 37). With factor loadings set at 0.50, 70 items loaded on the nine factors; none of the 70 items loaded significantly on any other factor. “All nine factors had eigenvalues greater than 1.00, accounted for 62% of the total variance, and contained at least three items whose difference in loading on the other factors was greater than 0.30” (37, p. 223). The question, “Should any items be dropped?” was considered. Because dropping items after just one factor analysis study should be viewed cautiously (38) and because the inter-item correlation matrix indicated that each item contributed uniquely and significantly to the TSE construct, no items were dropped.

Factors derived from an analysis provide one interpretation of the data and are useful for understanding relationships within a specific domain; however, the interpretation should be confirmed with other types of evidence (38). Additionally, construct validation should focus on the degree to which an instrument is consistent with the related literature and underlying conceptual framework (15); therefore, this was also examined. For the TSET, supporting evidence included several key theoretical issues. First, the factors related to the literature in transcultural nursing and the underlying conceptual framework. Items that clustered together made sense conceptually and were accordingly labeled: recognition, kinship and social factors, professional nursing care, cultural background and identity, life cycle transitional phenomena, awareness of cultural gap, communication, self-awareness, and appreciation (16, 37). Second, items that clustered together were from the same subscale, clustering exclusively on the Cognitive, Practical, or Affective Subscale. Subscale exclusivity most likely implies that within each broad educational category, as captured in the three TSET subscales, there are several underlying theoretical dimensions that contribute to the TSE construct (37). Third, students were most confident about Affective Subscale factors and least confident about Cognitive Subscale factors. Fourth, changes in mean scores on each of the nine factors occurred in the expected direction for novice and advanced students (means were higher for advanced students on all of the factors).

Results from this phase of the initial factor analysis study helped validate the following assumptions.

Factor Analysis: Latest Techniques and Comparative Findings

Since the initial factor analysis study, new statistical techniques have been developed that permit more sophisticated analyses. Using a sample of 272 culturally diverse associate degree nursing students, Jeffreys and Dogan (36) explored the factorial composition of the TSET by conducting a Common Exploratory Factor Analysis (CEFA). The reader is encouraged to note that CEFA is different from Principal Component Analysis (PCA); the rationale for conducting another factor analysis study using new statistical techniques is described here briefly.

“Both Exploratory Factor Analysis (EFA) and Principal Component Analysis (PCA) are multivariate statistical techniques widely used in social and behavioral sciences (43). PCA is a more popular method for factor extraction, because it is the default method provided in many popular statistical software packages. As Costello and Osborne (44) point out, however, PCA is not a true method of factor analysis ‘and there is disagreement among statistical theorists about when it should be used, if at all’ (p. 2)” (36).

Jeffreys and Dogan (36) also implemented two other similarly underused (if not ignored) methods in analyzing survey data. First, the most recent data imputation techniques were used in dealing with missing data. Second, the standard errors of factor loadings were used in determining the factor structure of the TSET. Specifically, standard errors for factor loadings were computed and utilized in deciding if a given item loads significantly on a factor and whether the difference between the factor loadings of two or more items on the same factor are statistically significant. (Details about these procedures and the technical differences between the PCA and CEFA may be found in Jeffreys and Dogan [36].)

The CEFA comprising 69 of the 83 items yielded four factors: “Knowledge and Understanding,” “Interview,” “Awareness, Acceptance, and Appreciation,” and “Recognition,” with internal consistency ranging from 0.94 to 0.98. Notably, all of the items within a factor came from a single subscale of the TSET. Consistent with the item-exclusive factor loadings on subscales demonstrated in the previous factor analysis study (37), what this most probably implies is that within each broad educational learning domain, as captured by the three TSET subscales, there are several underlying theoretical dimensions that contribute to the construct of TSE. Furthermore, the internal consistency of each of the factors was quite high, ranging from 0.94 to 0.98, attesting to the coherence of the underlying conceptual structure; reliability of the total instrument was 0.99 (36). Unlike the previous study in which 70 items yielded nine factors consisting of items that loaded exclusively within the originally conceptualized subscale (37), the number of factors came closer to the initial conceptualization of subscales. However, validation of a construct must focus on the degree to which an instrument is consistent with the underlying theoretical framework and related literature. If the performance of an instrument is consistent with the theoretically proposed expectations, then it is concluded that the instrument is construct valid (15).

The factorial composition of the TSET continues to be consistent with the underlying theoretical framework and the related literature in nursing, education, and psychology (self-efficacy). Notably, Factor 1 incorporated all 25 items on the Cognitive Subscale with factor loadings ranging from 0.63 to 0.80 and all items correlating positively with the total score (0.68 to 0.83). The Cognitive Subscale targeted the cognitive learning domain conceptualized by Bloom et al. (45), focusing on the first two levels of “Knowledge and Understanding.” Using the more valid and newer CEFA approach yielded a coherent factor structure within the cognitive learning domain whereby all items loaded exclusively and significantly on this subscale, demonstrating high internal consistency (Cronbach’s alpha = 0.97). The previous PCA statistical approach (37) separated items in the Cognitive Subscale into two factors: “Professional Nursing Care” (nine items) and “Lifecycle Transitional Phenomena” (eight items). The content validity experts previously validated that the 25 items (developed from the transcultural literature) within the Cognitive Subscale were appropriate and relevant for this learning domain within the full context of TSE. The 2009 CEFA analyses (36) supported the ratings by the content validity experts, thereby lending greater validity to the TSET. Similarly, a factor analysis of the TSET–Greek version administered to 500 nurses and student nurses resulted in all 25 items on the Cognitive Subscale loading on Factor 1 (46).

Educational experts support that sufficient background knowledge and understanding is needed in order to apply information practically. As conceptualized in the TSET and the CCC model, interviewing clients about cultural values and beliefs is essential for the completion of a comprehensive health history, development of a culture-specific/culturally competent and congruent care plan, and provision of culturally congruent care. Using the CEFA approach, Factor 2 (Interview) contained 22 of the 28 items on the Practical Subscale with factor loadings ranging from 0.67 to 0.86, all items correlating positively with the total score (correlation coefficients ranged from 0.77 to 0.89), and demonstrated high reliability (Cronbach’s alpha = 0.98). The labeling of Factor 2 as “Interview” was considered appropriate for the underlying conceptual framework that purports that the teaching and learning of cultural competence must carefully weave together cognitive, practical, and affective learning. Although Factor 2 excluded six of the items dealing with asking clients about verbal and nonverbal communication, space and touch, time perception, language preference, and English comprehension, all of the other essential expert-rated interview topics loaded exclusively on this one factor. One possible explanation for the item exclusion may be that the respondents assumed that in order to conduct an interview, assessment of English comprehension and so on would be obviously essential. Previously (37), these six items loaded separately on Factor 7, labeled “Communication.”

Transcultural experts and educational leaders continually attest to the importance of the affective learning domain on integrating foundational and key professional values within a professional education (47–77). Although the 1998 study (37) yielded four separate factors consisting of items loading exclusively on the Affective Subscale, the most recent CEFA study analyses generated two factors labeled “Awareness, Acceptance, and Appreciation” (Factor 3) and “Recognition” (Factor 4) (36). Factor 3 addressed: (a) all three items dealing with awareness of own cultural group and biases, (b) awareness of insensitive and prejudicial treatment among different cultural backgrounds; (c) all three items concerned with accepting differences and similarities between cultural groups and client’s refusal of treatment based on beliefs; and (d) all five items associated with appreciation. The factor loadings for this factor ranged from 0.51 to 0.87, all items correlated positively with the total score (correlation coefficients ranged from 0.61 to 0.85), and the factor was highly reliable (Cronbach’s alpha = 0.94).

Factor 4 (Recognition) contained 10 of the 11 items that addressed recognition in the affective dimension. Respondents were asked to recognize, among clients of different cultural backgrounds, the impact of select elements (political factors, values, roles, socioeconomic factors) on health care practices, the need to foster cultural care through nursing strategies (preservation/maintenance, accommodation/negotiation, repatterning/restructuring), importance of home remedies and folk medicine, and the need to prevent cultural imposition and ethnocentric views. Excluded from the factor was the item about inadequacies in the U.S. health care system. The factor loadings for this factor ranged from 0.55 to 0.80. Cronbach’s alpha for the set of items loading on this factor was 0.94. All items correlated positively with the total score; the correlation coefficients ranged from 0.68 to 0.83.

Originally conceptualized as one dimension tapping on affective learning, the latest analyses (36) may suggest that within the respondents’ perspective, recognition (Factor 4) involved a different type or level of affective learning. Several transcultural experts have identified “awareness” as the essential, first, or key element required for cultural competence development (49, 50, 59, 62, 66, 70–77). This awareness includes self-awareness as well as awareness of various aspects of culturally different clients. Recognition may include a different level of assessment that blends together knowledge, practical skills, and awareness. But, notably, all of these components are considered essential in the development of cultural competence and in the provision of culturally congruent nursing care. Perhaps focus groups with similar and/or different learner/health care provider professional populations will lend greater insight into the perceived differences between “Awareness, Acceptance, and Appreciation” and “Recognition.”

Consistent with the 1998 study (37), virtually all Affective Subscale items under the heading “Recognition” clustered together. In the CEFA study (36), only the item about recognizing inadequacies in the U.S. health care system was excluded. This item was added to the TSET upon the suggestion of several of the content validity experts. Previously, all items were clustered together under “Recognition.” Since the development of the TSET in 1994 (9), changes within the U.S. health care system have been ongoing; some of these changes may be reflected in the respondents’ views of this item. Additionally, the 2009 study sample included students within one associate degree program (n = 272). Generally, associate degree nursing students do not receive as much background course work grounded in the complexities of the U.S. health care system. The study also occurred prior to large political debates, documents, and decision making leading up to the Affordable Care Act and major changes in the U.S. health care system. Researchers in other countries have also requested changing this item to mention their respective country and/or deleting this item, or changing to a generic version such as “the nation’s health care system.” Because one purpose of factorial analyses may be to refine instrument items and create more parsimonious instruments that capture a valid measure/assessment of the targeted construct, consideration of deleting this item following repeated CEFA studies using different health care professional populations may be considered; however conceptual and practical reasons discourage deletion of this item. The reality is that the structure and set-up of a health care system (or lack of) influences health care access, quality, and cost and awareness (or lack of) concerning the existing health care system within a governing body and/or geographic location, and is important to the delivery of culturally competent care. The generic term “the nation’s health care system” is an option that is open to comparison more globally and within changing political and economic times.

Factor Analysis: Conclusive Summary

Conclusively, the latest CEFA study (36) continues to support that the TSET assesses the multidimensional nature of TSE while also differentiating between three types of learning: cognitive, practical, and affective. Additionally, each TSET item continues to contribute to the reliability of the underlying constructs, as evidenced by the positive and high item to total score correlations. Furthermore, the TSET continues to demonstrate high levels of internal consistency and had a coherent factor structure supportive of the underlying conceptual framework (CCC model). Moreover, researchers exploring the factor analyses of the TSET–Chinese (78, 79), TSET–Greek (46), and TSET–Turkish (80) conclude that the underlying factor structure continues to be consistent with the underlying framework and related literature (see TSET Research Exhibits 4.1–4.3). The benefits of this support allow the researcher/educator to move beyond mere assessment to the design, implementation, and evaluation of diagnostic-prescriptive teaching strategies for cultural competence education. The findings also suggest that the TSET and underlying CCC model have worldwide application and validity and that cultural competence education (and the measurement and evaluation of changes following cultural competence educational strategies) is a global interprofessional priority. In addition, researchers conducting factor analyses on the TSET–Chinese, TSET–Greek, and TSET–Turkish report findings that are generally consistent with the underlying CCC model framework at high levels of internal consistency (Cronbach’s alpha) (29, 46, 79–80). Researchers for the TSET–Chinese and TSET–Turkish reported that the factor with the highest loading (Factor 1) was comprised of items in the Affective Subscale (79, 80); the original TSET (American English) and TSET–Greek (46) reported that the factor with the highest loading (Factor 1) was comprised of items in the Cognitive Subscale (see TSET Research Exhibits 4.1–4.3).

---

TSET RESEARCH EXHIBIT 4.1

The Reliability and Validity of the Transcultural Self-Efficacy Tool–Turkish (TSET–Turkish)

Ayla Baylk Temel, PhD, RN

Professor of Nursing

Ege University Nursing Faculty

Community Health Nursing Department

I˙zmir, Turkey

Fatma Başalan İz, PhD, RN

Assistant Professor

Süleyman Demirel University Faculty of Health Sciences

Community Health Nursing Department

Isparta, Turkey

Research Report

Purpose: To examine whether the TSET is a valid and reliable tool for nursing students in Turkey.

Research questions:

  1.  What is the factor composition of the Turkish version of TSET?

  2.  Is there a similarity in the composition of the TSET–Turkish version and the original TSET?

  3.  To what degree are items on the TSET–Turkish internally consistent?

  4.  Is the internal consistency of the TSET–Turkish adequate?

  5.  Does the TSET–Turkish accurately measure what it is supposed to measure?

Study design: Methodological

TSET–Turkish translation: The translation-back-translation method was used for determining language equivalence. Three nursing faculty members and the researchers translated the TSET from English to Turkish. Researchers reviewed the translated items and prepared a Turkish form. A linguistic specialist translated it back to English. Any uncertainties regarding terminology were resolved through discussions between the translators and the researchers. The instrument was pilot tested with 50 nursing students and comprehensibility was evaluated through student interviews. Next, a panel of five experts (nursing faculty members) reviewed the instrument’s items for its content validity, linguistic properties, clarity, and understanding. Experts rated each item of the TSET–Turkish using a content validity index (CVI) form containing a 4-point rating scale (1 = not relevant to 4 = very relevant), resulting in a CVI of 0.83 as determined by the proportion of experts who rated the item as relevant (a rating of 3 or 4). After making some minor changes in wording, the instrument was retested again on 10 nursing students, resulting in the final TSET–Turkish version.

Sample:

Size: 485

Type of learner: Fourth semester BSN students enrolled in 1 of 7 (out of 11 total) nursing programs in Turkey.

Demographics: Average age was 22.4 ±1.28; 97% born in Turkey; 63% spent his or her life in cities; 93% lived in nuclear families. Turkish was the language spoken at home generally; however, 76% spoke another language in addition to Turkish. Approximately 76% cared for patients from different cultures during their clinical practicum; 39% reported language and communication difficulties while providing care.

TSET data collection: At the beginning of the semester

Educational interventions/Teaching–learning strategies: Cultural competency was studied throughout the curriculum. Topics such as culture of health, values, beliefs, health behaviors, folk medicine, and nurses attitudes’ concerning culturally diverse groups are common in Turkish nursing curricula.

TSET Reliability (Cronbach’s alpha):

Data analysis:

Correlation coefficient matrix: Total item correlation of the scale ranged between 0.31–0.84. Five items had a total item correlation lower than 0.40. (All TSET–Turkish items contributed uniquely and sufficiently to the TSE construct.)

Factor analysis using principal component analysis with varimax rotation yielded three factors. Factor 1 accounted for 51.43% of the variance (Cronbach’s alpha = .96) and contained all items from the Affective Subscale (all but three achieved loadings above .4). Factor 2 contained all items from the Practical Subscale, accounting for 5.47% of the variance (Cronbach’s alpha = .97) (all but one item loaded above .4). All Cognitive Subscale items loaded at levels of .39 or above on Factor 3, accounting for 3.71% of the variance (Cronbach’s alpha = .96).

TSET scores*:

---

	Mean	Standard Deviation
Total	7.95	1.27
Cognitive	7.79	1.32
Practical	7.79	1.46
Affective	7.95	1.31

---

*80% of student responses were in the medium range (between 3 and 8 on the 10-point scale).

ANOVA of TSET scores and select demographics: Age (older), to be employed, ability to speak a second (another) language, and caring for an individual from a different culture were found to be significant predictive variables for internal criterion validity (p < 0.05).

Discussion:

  1.  The TSET–Turkish is a valid, reliable, and useful tool for assessing nursing students’ transcultural nursing skills and transcultural self-efficacy (TSE) in Turkish nursing schools.

  2.  The factor structure of the TSET–Turkish was found to be compatible with the original TSET and underlying conceptual framework, although the Affective Subscale accounted for the most variance and the Cognitive Subscale accounted for the least variance.

Implications:

  1.  Results from the TSET–Turkish can guide curricular revisions to enhance cultural competence education throughout the curriculum.

  2.  Conduct longitudinal studies prior to and after cultural competence teaching–learning intervention(s).

  3.  Continue to collect data using the TSET–Turkish and compare with other TSET research (original version and other language TSET translations).

---

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. A Guide for Interpreting Learners’ Transcultural Self-Efficacy (TSE) Perceptions, Identifying At-Risk Individuals, and Avoiding Pitfalls
2. Faculty Advisement and Helpfulness: A Culturally Congruent Approach
3. Cultural Competence Clinical Evaluation Tool (CCCET) and Clinical Setting Assessment Tool–Diversity and Disparity (CSAT–DD), with contributions from Enis Dogan and Lynn Schmidt
4. Case Exemplar: Linking Strategies—Spotlight on Employee Orientation Programs to Enhance Cultural Competence, with contributions from Cynthia Karczewski

Stay updated, free articles. Join our Telegram channel

Join