- Email: [email protected]
Measuring peer caring behaviors of nursing students: Scale development
ARTICLE IN PRESS
International Journal of Nursing Studies 44 (2007) 105–114 www.elsevier.com/locate/ijnurstu
Measuring peer caring behaviors of nursing students: Scale development Chien-Lin Kuoa, Michael A. Turtonb, Jane Lee-Hsieha,, Hung-Fu Tsengc, Chin-Lung Hsua a Department of Nursing, Junior College, Fooyin University, Kaohsiung, Taiwan Department of Applied Foreign Languages, Chaoyang University, Taichung, Taiwan c Institute of Medical Research, Chang-Jung University, Tainan, Taiwan
b
Received 28 August 2004; received in revised form 20 July 2006; accepted 20 July 2006
Abstract Background: Caring is one of the most important domains of nursing research, peer caring among student nurses, and its potential effects on nurse caring behaviors remains largely unexplored. Few tools in the literature target peer caring interactions, and which were either irrelevant to our research purpose or culturally inappropriate for nursing student population in Taiwan. Objectives: The purpose of this study was to develop a culturally sensitive instrument to measure peer caring behavior from the student perspective and to offer a descriptive answer to ‘‘what is peer caring in Taiwan.’’ Design: The study uses both qualitative and quantitative methods in developing the ‘‘Peer Caring Measurement (PCM)’’ questionnaire. Settings: Students from a 5-year associate degree nursing program in a university of technology in southern Taiwan were recruited into this study. Participants: Thirteen first- through fourth-year student volunteers between the ages of 16 and 20 were interviewed to explore caring behavior in student peer-to-peer interactions. Two classes from each of the first, second, third, and fourth year students, a total of 360 students were randomly selected to assess the internal consistency of ‘‘PCM’’, and 47 first-year students were conveniently selected to examine the stability of the tool. Methods: Interviews were conducted in an unstructured manner. Qualitative data were analyzed by a constant comparative method. The questionnaire survey was used to assess the validity and reliability of ‘‘PCM’’. Results: A 17-item ‘‘PCM’’ was developed; the content validity, construct validity, and reliability of the tool were ensured by expert review, factor analysis, and internal consistency. Three factors, labeled assistant caring, academic caring, and affective caring, accounted for 63.197% of the variance. Conclusions: The ‘‘PCM’’ addressed the multidimensional construct of peer caring. It was validated in a Chinese language version and can be used in college settings to evaluate student interactions and their peer caring behaviors. r 2006 Published by Elsevier Ltd. Keywords: Peer caring; Instrument development; Nursing education; Taiwan
Corresponding author. Tel./fax: +886 7 7828258.
E-mail address: [email protected] (J. Lee-Hsieh). 0020-7489/$ - see front matter r 2006 Published by Elsevier Ltd. doi:10.1016/j.ijnurstu.2006.07.025
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
106
What is already known about the topic?
Caring
is one of the most important domains of nursing research; peer caring among student nurses and its potential effects on nurse caring behaviors remain largely unexplored. There are few tools in the literature which target peer caring interactions, and most of them are from western perspective.
What this paper adds
The
‘‘Peer Caring Measurement’’ describes the meaning of peer caring from the perspective of Taiwanese students by assessing three factors: assistant caring, academic caring and affective caring. The ‘‘Peer Caring Measurement’’ can be applied in nursing school settings to evaluate student interactions and their peer caring behaviors.
1. Introduction Researchers generally agree that caring is a complex and crucial component of nursing knowledge (Schoenhofer, 2001). Though caring is one of the most important domains of nursing research, peer caring among student nurses, and its potential effects on nurse caring behavior, remain largely unexplored. Caring theorists such as Noddings (1992) have argued that caring begins with the self and those closest to oneself. Peer caring behavior is thus an indicator of the status of a student nurse’s growth in caring attitudes and abilities. In 2002, the researchers established a concept-based caring curriculum framework for the 5-year associate degree nursing program at our university in southern Taiwan. The caring curriculum was based on two models, the meaning of Caring model and the dynamic caring model, developed out of the researcher’s prior phenomenological study of caring in a Taiwan context (Hsieh et al., 1998). The meaning of caring contains two major domains: a psychosocial domain and a caring action domain. The psychosocial domain includes the categories of love, unconditional acceptance, empathy, genuineness, respect, and treating others as family members. The Caring action domain includes the categories of providing professional behaviors, fulfilling needs, sensational approaching, self-offering, supporting, and providing a suitable environment. The new curriculum consists of six credits of required coursework, including a first-year course (two credits) entitled ‘‘Introduction to Caring,’’ a third-year course (two credits) entitled ‘‘Application of Caring Concepts’’ and a fifth-year course (2 credits) entitled ‘‘Professional
Caring.’’ Introduction to Caring covers personal and interpersonal caring, including peer caring (Lee-Hsieh et al., 2004). The purpose of the Introduction to Caring course is to create a caring atmosphere throughout the entire campus, and increase student peer-to-peer, family, and teacher–student interpersonal caring. The ultimate goal of the course is to foster student ability to self-offer in caring action. Currently, the first three stages of the curriculum development process, planning, design, and implementation, have been completed for the Introduction to Caring course. However, without the final stage, evaluation, there is no way to know whether the course has been effective in fostering student learning of caring attitudes and behavior. The ‘‘Peer Caring Measurement (PCM)’’ instrument is the final stage in the curriculum development process for this course. The purpose of the PCM to explore the peer caring behavior of students who have taken the Introduction to Caring course in order to evaluate their ability to apply the caring skills and concepts they studied to their real-life peer interactions. These concepts include such ideas as listening and expressing, encouraging and praising, accepting and trusting, empathy and respect, and love and sincerity. Two factors impelled us to create our own evaluation instrument. First, few tools in the literature target peer caring interactions. For example, Watson (2002) collected and compared 16 survey instruments. Of these, 15 examined caring behavior or abilities from the patient or nurse point of view, while only one, that of Hughes (1998), explored peer caring behavior among student nurses. Second, the instrument developed by Hughes, the ‘‘Peer Group Caring Interaction Scale (PGCIS),’’ was inappropriate for our purposes, for three reasons. First, while four of the 16 items on the PGCIS addressed clinical practice issues, the student nurses we wanted to evaluate had not yet performed their clinical practice. Second, the students in Hughes’ (1998) study had a mean age of 25.6, ranging from 19 to 51, years older than the 16–20-year-olds who comprise our junior college nursing student population. Third, based on our previous experience attempting to use evaluation instruments developed in western cultural settings (LeeHsieh et al., 2005), we expected to encounter items that were culturally inappropriate or irrelevant.
2. Literature review The educational philosopher Noddings (1984, 1988, 1992), who has written extensively on caring in education, has argued that caring begins with the care of self and then expands outward to care of intimates, strangers, distant others, and the natural and man-made world. Noddings (1984) views caring education as
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
having four components: modeling, dialogue, practice, and confirmation. Peer groups provide all four of these components. Adolescents use their peer groups as resources for developing moral judgments and values (Hartup, 1996). Relationships with peers enable adolescents to explore important caring skills such as reciprocity, self-disclosure, trust, and conflict resolution skills (Collins and Repinski, 1994). Recognizing the importance of peers in shaping adolescent views of caring, Noddings has argued that the peer group may be particularly useful in providing an environment where student nurses can learn caring. Although findings from several studies suggest that peer interactions may help student nurses learn caring, few studies have attempted to describe such effects. Hughes (1998) noted that ‘‘little attention has been focused on peer relationships as an important dimension of educational experience in schools of nursing’’(p. 206). Beck (2001) was able to locate only two studies focusing on caring among nursing students, her own of 1992 and the study of Hughes (1993). Beck (1992) argued that when nursing students experience caring from fellow nursing students, they learn caring. In her phenomenological study of 190 descriptive expressions collected from student nurses, Beck analyzed the experience of caring into four necessary constituents, authentic presencing, self-less sharing, fortifying support, and enriching effects. She further suggested (1993) that student nurses share their experiences of clinical caring with fellow students to increase familiarity with practical ways to demonstrate caring to patients, and allow students to serve as caring role models for each other. Thus, Beck’s research provides support for the idea that student nurses who care for each other will be more caring for parents as well. Caring, in her view, is contagious. Hughes (1993, 1998) developed an instrument to measure caring peer group interactions, the PGCIS. The PGCIS consisted of 16 items distributed across two factors, caring behavior (nine items), and giving assistance (seven items). Hughes recommended that a survey instrument should be evaluated by both quantitative and qualitative approaches.
3. Methodology and results This study used both qualitative and quantitative methods in constructing the ‘‘PCM’’ tool. The study was carried out from the spring of 2003 to the summer of 2004. 3.1. Phase I The initial phase of the development of the instrument consisted of three steps. First, the researchers conducted
107
interviews and analyzed the contents of the interviews. Thirteen first- through fourth-year student volunteers were interviewed to explore caring behavior in student peer-to-peer interactions. Students signed up to be interviewed after the researchers requested volunteers for the research project through announcements in their own classrooms. Open-ended interviews were used to gather data on the participant perceptions, experiences, and feelings about student peer caring behavior. The researchers used at least three students from each year as participants. Data saturation was reached with the 13th participant. 3.1.1. Ethical considerations Before interviewing the students, the researchers informed the students of the purpose of the research, and obtained written consent from each participant. The researchers also obtained participant permission to audiotape each interview for the purpose of content analysis and audit trail. 3.1.2. Step I: interview process Interviews were conducted in an unstructured manner. Each interview began by asking the participant to express their experience of peer caring behaviors. Interviews began with the question ‘‘What do you think caring is?’’ This might be followed by questions designed to elicit specific caring experiences, such as ‘‘what was your most unforgettable peer caring experience?’’ Participants were also asked to describe what kinds of caring behaviors they would like to receive from their peers. The following are examples of prompts used to obtain further details from participants: ‘‘What you mean isy’’; ‘‘Can you tell me more about that?’’; ‘‘Is there anything more you would like to tell me?’’ Interviews lasted between 45 and 90 min. 3.1.3. Step II: content analysis and instrument drafting Interview results were immediately transcribed verbatim. Data were analyzed by a modified version of the constant comparative method (Glaser and Strauss, 1967; Lincoln and Guba, 1985). The constant comparative method contains four steps: (1) comparing incidents applicable to each category, (2) integrating categories and their properties, (3) delimiting the construction, and (4) writing the construction (Glaser and Strauss, 1967). In the first step, meaningful descriptions of peer caring were identified, coded, compared, and then a preliminary classification into categories was performed. Analyzed data from each interview were constantly compared to each other in order to fit them into the most appropriate category. In the second stage, descriptions of caring were compared to the preliminary version of the definitions of the various categories. Descriptions were resorted to increase the coherence of each category. Stage three was completed when redundant categories
ARTICLE IN PRESS 108
C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
had either been deleted or integrated into existing categories. Data collection and analysis were conducted until data saturation was reached. With the concepts from the Introduction to Caring course as a foundation, the researchers used the analysis of the student interview data to construct a 25-item PCM instrument. The instrument used a six-point Likert scale (1 indicating ‘‘strongly disagree,’’ 2 indicating ‘‘moderately disagree,’’ 3 indicating ‘‘slightly disagree,’’ 4 indicating ‘‘slightly agree,’’ 5 indicating ‘‘moderately agree,’’ and 6 indicating ‘‘strongly agree’’). 3.2. Phase II 3.2.1. Step I: assessing content validity In step 1, the 25-item PCM instrument was submitted to a panel of ten experts, including five experts from the fields of education and nursing, and five instructors who had taught the Introduction to Caring course at our university, in order to evaluate its clarity and relevance. A four-point scale was used. The panel determined whether each item was appropriate, accurate, and representative (1 ¼ irrelevant and should be deleted; 2 ¼ seemingly relevant but large-scale revision was required; 3 ¼ relevant but in need of small adjustments; and 4 ¼ relevant, clear and precise). The instrument was then revised accordingly. 3.2.1.1. Item wording revisions. For the sake of readability and clarity, revisions were made to eliminate awkward wording in items 1, 4, 5, 11, and 17. For example, the item ‘‘when a student is in need, other students will keep him or her company, and not leave her alone with her problems’’ was changed to ‘‘When a student is lonely and needs company, other students will keep him or her company.’’ 3.2.1.2. Content revisions. The item ‘‘For the good of the class, or for the good of a particular individual, students will offer suggestions or attempt to persuade the individual to change their behavior’’ included two different levels, the group and the individual. To clarify this, it was changed to ‘‘for the good of a specific individual, classmates will offer suggestions’’. The experts felt that the language of the item ‘‘providing needed extra help is the responsibility of the teacher’’ was unclear, for students would take the word ‘‘extra’’ to mean that the help was beyond their ability to provide and should be given by the teacher, not peers. If no help was given, it was the teacher’s responsibility, not because they as peers were uncaring. Hence, the item failed to measure student willingness to help, as was intended. Consequently, it was revised to ‘‘helping students who need help is solely the teacher’s responsibility.’’ One expert felt that the item ‘‘interpersonal trust between students’’ was not relevant to caring, while another
expert felt that ‘‘trust’’ was too broad an idea, and the question was not clear enough. Trust is generally seen as an important component of caring. A foundation of interpersonal trust can raise the willingness to offer caring (Hsieh et al., 1998). Due to this understanding, the researchers decided not to delete this item, instead revising it to ‘‘the students trust each other and do not lie to one another.’’ The item ‘‘students chat and share their feelings with each other,’’ contained two different levels of intimacy, chatting and sharing of feelings, and was therefore split into two items by the researchers, ‘‘students have classmates to chat with’’ and ‘‘students have classmates to share their feelings with.’’ Most of the experts felt that the item ‘‘there is intense competition between the students’’ simply described a common phenomenon in Taiwan schools, given the ruthless competition for places in good schools under the local testing system. Generally, competition may be positive or negative. This item was thus revised to ‘‘there is negative competition between the students.’’ 3.2.1.3. Item order revisions. Most of the experts argued that the content of items originally designated 1, 2, 8, and 20 was quite similar, and suggested that some be deleted. Based on their reasoning in designing these items, the researchers believed that item 1, ‘‘when a student encounters difficulties, other students will notice,’’ and item 2, ‘‘when a student asks for help, other students are happy to help,’’ addressed different aspects of the interpersonal action, one passive (being noticed), the other active (asking for help). Though item 8, ‘‘when a student encounters difficulties, other students will offer guidance and solutions’’ and item 20, ‘‘when a student encounters difficulties, other students will offer suggestions’’ seem similar, the researchers felt that they showed different levels of caring, as item 8 addressed the offering of concrete solutions for difficulties, unlike item 20. All four items were retained. However, to clarify the different levels of caring represented by these questions, the researchers placed them in series, one after another, so their similarities and differences would be immediate and obvious (items 1, 2, 3, and 4 in the resulting 27-item version of the instrument). After revising the wording, content, and item order, the original 25 items of the PCM instrument became 27 (Table 1). 3.2.2. Step II: assessing face validity The language of the instrument was then reviewed for clarity by the 13 students who were originally interviewed for the study. During the process of completing the survey, the students did not express any problems in
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
109
Table 1 27-item version of PCM from phase II 1.a 2.a 3. 4.a 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.b 15. 16. 17.b 18. 19.a 20.a 21. 22. 23.b 24.a 25. 26. 27.b a
When a classmate encounters difficulties, other classmates notice. When a classmate encounters difficulties, other classmates offer solutions. When a classmate encounters difficulties, other classmates take the initiative to help When a classmate asks for help, other classmates are happy to help Classmates are concerned with each other’s health. When a classmate is sick, other classmates accompany them to get medical help. When a classmate is lonely and needs company, other classmates are willing to keep her company. If there is a conflict between classmates, other classmates help smooth things out. Classmates respect the different opinions of other classmates. For the good of a specific individual, classmates will offer suggestions Classmates discuss assignments with each other. Classmates who perform well academically are willing to instruct students who need help. When a classmate encounters difficulties with assignments, other classmates offer suggestions. Helping students who need help is solely the teacher’s responsibility (reverse score item). Classmates listen to each other’s troubles and problems. Classmates trust each other and do not lie. Certain classmates are excluded by other classmates (reverse score item). Classmates are concerned with each other’s needs. Classmates help other classmates only when it is in their best interest to do so (reverse score item). Classmates have someone to chat with. Classmates have someone to share feelings with. There is deep friendship between classmates. Classmates are too busy to help each other (reverse score item). Classmates praise each other. Classmates encourage each other. Classmates share notes and references with each other. There is negative competition between classmates (reverse score item).
Deleted due to average inter-item correlations less than .3 or greater than .7 (6 items). Deleted due to item-to-total correlations less than .3 (4 items).
b
understanding the item wording and meaning, demonstrating its face validity. 3.2.3. Step III: assessing internal consistency Two classes from each of the first-, second-, third-, and fourth-year students at our school, a total of 360 students in eight classes (approximate 45 per class), were selected and given the tool. Classes were selected randomly. Class instructors informed the students of the purpose of the study and distributed and collected the questionnaires during last 10 min of class. Students were also informed that they did not have to fill out the questionnaire, and that there would be no punishment for those who decided not to participate. Questionnaires were anonymous and participant consent was obtained on a tear-off form. It took roughly 10 min to complete the questionnaire. Because the questionnaires were collected on the spot, the return rate was 100%. Data from the 360 valid returned questionnaires were analyzed using SPSS for Windows 10.0 (SPSS, 2000 Statistical Package for the Social Sciences) Internal consistency was measured by Cronbach’s a coefficient. Individual item characteristics were obtained by
corrected average inter-item correlation coefficients, inter-item correlations, and item-to-total correlations (Ferketich, 1991; Reeve, 1994; Younger, 1993). Items with an item-to-total correlation of less than .3 were deleted, leaving 23 items (Table 2). Next, inter-item correlation coefficients were obtained. Items with an inter-item correlation coefficient of less than .3 or greater than .7 were deleted (Ferketich, 1991), leaving 17 items (Table 1). The 17-item PCM Instrument had a Cronbach’s a of .9347. 3.3. Phase III Using SPSS for Windows 10.0, the researchers then examined the 17-item PCM using exploratory factor analysis with principal component and varimax rotation. Three factors emerged, accounting for a total of 63.197% of the variance (Table 3). Each factor was labeled based on a theme appropriate for the content of the factor. The first factor, comprising nine items, was labeled Assistance Caring. The second factor, composed of four items, was labeled Academic Caring. The third factor, also composed of four items, was labeled Affective Caring.
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
110
Table 2 Descriptive statistics and reliability analysis of the 27-item version of PCM
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22 Q23 Q24 Q25 Q26 Q27
Mean
Std. Dev.
Item-total correlation
Alpha if item deleted
4.24 4.39 4.37 4.57 4.73 5.03 4.32 4.18 4.12 4.26 4.83 4.74 4.60 4.97 4.71 3.74 4.13 4.33 3.38 5.05 4.92 4.52 4.01 4.40 4.64 4.66 3.92
1.020 1.076 1.081 1.079 1.030 .993 1.140 1.173 1.217 1.148 1.042 1.144 1.140 1.312 1.002 1.263 1.366 1.215 1.370 1.096 1.024 1.229 1.326 1.056 1.078 1.173 1.400
.5576 .6352 .6786 .6318 .6519 .5583 .6452 .6149 .6014 .6264 .6151 .6007 .6411 .0908 .6372 .5054 .0748 .5669 .4011 .4470 .6179 .6545 .2403 .6730 .7120 .6386 .2722
.9140 .9127 .9120 .9127 .9126 .9140 .9124 .9128 .9130 .9127 .9131 .9131 .9125 .9223 .9128 .9147 .9256 .9136 .9169 .9156 .9131 .9121 .9197 .9122 .9115 .9124 .9194
Table 3 17-item version of PCM, exploratory factor analysis with principal component and varimax rotation Factor 1 Q8 Q7 Q9 Q3 Q10 Q5 Q16 Q18 Q6 Q11 Q12 Q13 Q26 Q21 Q22 Q15 Q25 Eigenvalue Percent of total variance explained Cumulative percent
Factor 2
Factor 3
.754 .743 .719 .695 .684 .646 .567 .520 .503 .797 .797 .727 .643
8.415 26.295
1.315 19.645
.831 .760 .650 .529 1.012 17.257
26.295
45.940
63.197
had the highest factor loading (.831), followed by Item 22, ‘‘There is deep friendship between classmates’’ (.760). 3.4. Phase IV
3.3.1. Factor 1: assistance caring This factor involves offering assistance and suggestions when needed by classmates, and helping classmates handle problems. Item 8, ‘‘When there is a conflict between classmates, other classmates will help smooth things out,’’ had the highest factor loading, (.754), followed by item 7, ‘‘When a classmate is lonely and needs company, other classmates are willing to keep him or her company’’ (.743). 3.3.2. Factor 2: academic caring This factor involves academic information sharing and peer instruction. Item 11, ‘‘Classmates discuss assignments with each other,’’ and item 12 ‘‘Classmates who perform well academically are willing to instruct students who need help’’ had the highest factor loading (both .797). 3.3.3. Factor 3: affective caring This factor involves peer ability to listen, share, encourage, and support each other. Item 14, ‘‘Classmates have someone to share their feelings with,’’
To examine the stability of the PCM, the researchers gave it to 47 first-year students in the first and again in the third week of the semester. The difference between the first and second test was either 1 or 0 for at least 80% of participants on all 17 items (six-point Likert scale). The Pearson’s correlation coefficient (g) between the first and the second test was .72 (po:001), and the regression slope (b) was .724 (po:001). These results indicated the stability of the instrument.
4. Discussion The conceptual-theoretical basis of our research was provided by the work of Noddings (1984, 1992), who argued that caring begins with care of the self and those closest to oneself. The PCM was constructed through a step-by-step development process (Table 4). It is capable of meaningfully evaluating student attitudes toward their own caring behavior and that of others.
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
111
Table 4 Phases and steps of the peer caring measurement (PCM) development process Development process
Purpose
Phase I
Develop a new instrument for the measurement of peer caring behavior from the nursing student perspective Develop a set of peer caring behaviors (1) Develop items based on the interview data (2) Develop a preliminary version of the instrument Assess content validity and face validity Assess clarity and relevance of the items on a 4-point rating scale Assess clarity of wording and face validity of the 2nd version instrument Assess internal consistency of the 2nd version instrument
Step I Step II
Phase II Step I Step II
Step III
Phase III
Phase IV
Methods/data
Items remained
Student interviews Literature review, content analysis, and item construction
25 items
Expert analysis N ¼ 10
27 items (item wording, content and item order revisions)
Student assessment N ¼ 13
27 items
Test (N ¼ 360) Cronbach’s a, inter- 17 items (deleted 10 items) item correlations, item-to-total correlations (1) Investigate the construct validity Factor analysis (N ¼ 360) 17 items of the final version 17-item instrument (2) Define conceptual categories for the factors Assess instrument stability Test and retest (N ¼ 47) 17 items
Comparison of this instrument with that of Hughes (1998) yields a number of important divergences that reflect differing cultural practices in education, caring, and behavior. A key difference of the Taiwanese educational system lies in the definition of ‘‘classmate.’’ In Taiwanese colleges, students are typically enrolled in a ‘‘Class’’ when they enter college. For example, if 100 students enter the English program that year, they might be divided into two ‘‘Classes’’ of 50 each, typically designated 1A and 1B, or 1-1 and 1-2. The maximum size of both the intake and the Class is determined by the Taiwan Ministry of Education. These Classes then become surrogate families for the students throughout their college career. They are supposed to engage in collective activities such as trips and outings together, take care of each other, be closer to each other than to other students at the same university, and present a united front toward the world. Students in a Class take all of their required courses together, year after year, with the university dictating when they will take them, which students will be in class together, and which teachers they will take. For example, all 900 first-year students (18 Classes) take the required Introduction to
Caring course, and each Class will take it as a Class, all 50 together, 1-1 with instructor Lee, 1-2 with instructor Wang, and so forth. Thus, both informally (through expectations about behavior), and formally (through the university administration and course framework), students in a Class spend most of their weekdays together. In effect, a Class is more or less a potential caring group. Thus, when our questionnaire asked students to consider peer caring behavior from the standpoint of ‘‘classmates,’’ the term designates a specific and formally defined relationship. ‘‘Classmates’’ are those in the same group of 50. It is not, as it might be in the US, a loose descriptor of someone who happens to be in school or in a course with oneself. Hughes’ PGCIS used the terms ‘‘students at this school’’ or ‘‘classmates’’ when it asked about peer caring behavior, as if they were interchangeable. We chose the term ‘‘classmate’’ to reflect the fact that in a Taiwan college students have more limited contact with students outside their Class. Hughes’ (1998) PGCIS did not ask about conflicts between students. Because each Class involves the same 50 students year after year, intrapersonal conflicts represent serious problems, both for the poisoning effect
ARTICLE IN PRESS 112
C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
they have on interpersonal relations within the Class, and on the image the Class presents to the outside world. The importance of this issue can be gauged from the fact that Item 8, ‘‘If there is a conflict between classmates, other classmates help smooth things out,’’ had the highest factor loading, .754, in Assistance Caring. Even translating this item threw into sharp relief the differing cultural assumptions of the researchers. The American automatically selected the word ‘‘resolve,’’ reflecting his assumption that conflicts are things that demand solving. However, the Chinese word ‘‘tiao jie’’ does not call for a solution, but merely a restoration of the previous placid surface of things, a smoothing over rather than an ending (the primary authors are comfortable in both languages). Item 7, ‘‘When a classmate is lonely and needs company, other classmates are willing to keep him or her company,’’ had the second-highest factor loading in Assistance Caring. Loneliness was not mentioned in Hughes’ PGCIS. In all probability, this is due to the higher collectivism of Chinese culture (Hofstede, 1980, 1983; Duncan, 1994), in which people who are by themselves are more likely to be viewed as needing, and expecting, company. Cultural preferences for collectivism also explain the wording of item 10, ‘‘for the good of a specific individual, classmates will offer suggestions.’’ The wording does not quite capture the implied meaning, which is that the individual’s good can be defined either by their own needs, or the needs of the Class. The Chinese wording thus contains an element of social reinforcement of group norms that is much less obvious to the English reader. Cultural preferences in communication appeared in the idea of listening versus talking. Hughes’ questionnaire contained 3 items that referred to ‘‘talking’’ with others and none that addressed ‘‘listening.’’ Item three of Hughes’ instrument ‘‘Students at this school talk with each other about their problems and concerns’’ contrasts with our item 15, ‘‘Classmates listen to each other’s troubles and problems.’’ In our interviews, almost every student gave an example of someone listening to them as an instance of authentic caring. The word typically used to describe such listening, qing ting, expresses the idea of a listening from the heart in which the self is offered in caring for others. Another cultural difference not captured by Hughes’ PGCIS was the use of intermediaries, especially when a student is sick. Invariably, when a student requires medical help, regardless of the seriousness of the problem, a second student will take her to the clinic. Thus, item 6, ‘‘When a classmate is sick, other classmates accompany them to get medical help,’’ had no parallel on Hughes’ questionnaire. Chinese culture is legendary for its emphasis on academic performance. Thus, one important aspect of
peer caring is offering help to classmates having academic difficulties. Collective face is also at stake, since average scores for each Class are often announced at year-end meetings and publicly posted, to encourage performance. In our study, academic caring was a factor in its own right, and the four items in it all had relatively high factor loadings (.797, .797, .727, and .643). By contrast, Hughes’ PGCIS findings revealed relatively lower factor loading for ‘‘share class notes, books, and articles,’’ and ‘‘giving suggestions about school assignments.’’ Item 1 of Hughes’ PGCIS, ‘‘Students at this school anticipate the needs of their classmates,’’ had no counterpart on our PCM. No student mentioned this peer caring behavior in the interview phase.
5. Recommendations The process of constructing and implementing this instrument has yielded several important recommendations for future research into peer caring. The development process for this instrument underscores the findings of our previous study (Lee-Hsieh et al., 2005) that while caring may be universal among humans, its expression is entirely local. Psychologists working in Chinese cultural settings have long expressed dissatisfaction with the limited applicability of western measurement instruments (Yang, 1987; Yang and Bond, 1990), and we can only echo that. It is apparent that behavioral measurement instruments must be developed within particular cultural settings, a conclusion that implies a development process that partakes of both qualitative and quantitative approaches, each informing the other. Though this measurement instrument is aimed at nursing students, the items therein examine psychological, social, and behavioral aspects of caring. These include empathy, sincerity, respect, praise, and communication (Lee and Fang, 2000; Hsieh et al., 1998). Because it looks at aspects of caring with broad application, the instrument may be used in almost any type of Chinese school setting with many different kinds of students. It may be used as a before/after test to explore the effects of coursework in caring, as well as a tool administrators can use to take a campus-wide snapshot of the status of caring at their university.
6. Limitations This instrument was developed to assess peer caring among student nurses in a junior college in Taiwan between the ages of 16 and 20. It may not be appropriate for other age groups, institutional, and cultural settings. The original PCM is in Chinese.
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
7. Conclusion This study used qualitative methods to develop a PCM questionnaire. The content validity, construct validity, and reliability of the 17-item PCM were ensured by expert review, factor analysis and internal consistency, respectively. The PCM was validated in a Chinese language version. A translated version in any other language would also require appropriate validation. The 17-item PCM instrument developed in this study addressed the multidimensional construct of peer caring. It can be used in college settings to evaluate peer caring behavior. It is easy to administer and takes less than 10 min to complete. This study explores peer caring from the perspective of students and offers a descriptive answer to ‘‘what is peer caring in Taiwan.’’ For nursing programs hoping to develop specific teaching–learning objectives for their nursing curricula, this study offers directions to explore in understanding the meaning of caring in nursing as well as teaching strategies. Instructors could motivate and encourage nursing students to provide peer caring behaviors by using the PCM to evaluate the peer caring in their institutions. If school administrators seek to promote and support a caring atmosphere on campus, they may use the PCM to evaluate student interactions and the effectiveness of peer caring initiatives. The cross-cultural issues raised by this study suggest several directions for future research. For example, peer caring is only one component of a caring campus; research is needed into student–family, faculty–student, faculty–faculty, and other caring relationships on campus. This study has made contributions to caring education in Taiwan by providing a description of the meaning of peer caring and offering a way to evaluate it. We hope that further exploration of this vital and interesting topic will enhance our understanding of the issues. Acknowledgements The researchers would like to thank the National Science Council for the grant that enabled this study, and our students for their unfailing patience and openness. A special debt of thanks is owed to our dedicated and overworked assistant, Ms. Mei-Ling Chen. Special thanks also to Dr. Clyde Warden of National Changhua University for his suggestions. References Beck, C.T., 1992. Caring among nursing students. Nurse Educator 17 (6), 22–27.
113
Beck, C.T., 1993. Caring relationships between nursing students and their patient. Nurse Educator 18 (5), 28–32. Beck, C.T., 2001. Caring within nursing education: a metasynthesis. Journal of Nursing Education 40 (3), 101–109. Collins, W.A., Repinski, D.J., 1994. Relationships during adolescence: continuity and change in interpersonal perspective. In: Raymond, M., Gerald, R.A., Thomas, P.G. (Eds.), Personal Relationships During Adolescence. Sage Publications, Thousand Oaks, pp. 7–36. Duncan, W.J., 1994. Why some people loaf in groups while others loaf alone. The Academy of Management Executive 1 (8), 79–80. Ferketich, S., 1991. Focus on psychometrics aspects of item analysis. Research in Nursing and Health 14, 165–168. Glaser, B.G., Strauss, A.L., 1967. The Discovery of Grounded Theory. Aldine, Chicago. Hartup, W., 1996. The company they keep: friendships and their developmental significance. Child Development 67, 1–13. Hofstede, G., 1980. Culture’s Consequences: International Differences in Work-Related Values. Sage, London and Beverly Hills. Hofstede, G., 1983. Dimensions of national cultures in fifty countries and three regions. In: Deregowski, J.B., Dziurawiec, S., Annis, R.C. (Eds.), Expiscations in CrossCultural Psychology. Swets and Zeitlinger, Lisse, Netherlands, pp. 335–355. Hsieh, J., Fang, Y.Y., Chiang, L.Y., 1998. An exploration of the meaning of caring in Taiwan. International Journal for Human Caring 2 (2), 12–22. Hughes, L., 1993. Peer group interactions and the students perceived climates for caring. Journal of Nursing Education 32, 78–83. Hughes, L., 1998. Development of an instrument to measure caring peer group interaction. Journal of Nursing Education 37, 202–207. Lee, J.C., Fang, Y.Y., 2000. Caring: the core of nursing education. The Journal of Nursing 47 (3), 22–27. Lee-Hsieh, J., Kuo, C.K., Tsai, Y.H., 2004. An action research on the development of a caring curriculum in Taiwan. Journal of Nursing Education 43 (9), 391–400. Lee-Hsieh, J., Kuo, C.K., Tseng, H.F., Turton, M.A., 2005. Development of an instrument to measure caring behaviors in nursing student in Taiwan. International Journal of Nursing Studies. Lincoln, Y.S., Guba, E.G., 1985. Naturalistic Inquiry. Sage Publications, Beverly Hills, CA. Noddings, N., 1984. Caring: A Feminine Approach To Ethics and Moral Education. University of California Press, Berkeley, CA. Noddings, N., 1988. An ethic of caring and its implications for instructional arrangements. American Journal of Education 96 (2), 215–230. Noddings, N., 1992. The Challenge to Care in School: An Alternative Approach to Education. Teachers College Press, New York. Reeve, M.M., 1994. Development of an instrument to measure effectiveness of clinical instructors. Journal of Nursing Education 3, 15–20.
ARTICLE IN PRESS 114
C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
Schoenhofer, S.O., 2001. Infusing the nursing curriculum with literature on caring: an idea whose time has come. International Journal for Human Caring 5 (2), 7–14. Watson, J., 2002. Assessing and Measuring Caring in Nursing and Health Science. Springer, New York. Yang, K.S., 1987. Chinese personality and its change. In: Bond, M.H. (Ed.), The Psychology of the Chinese People. Oxford University Press, New York, pp. 106–170.
Yang, K.S., Bond, M.H., 1990. Exploring implicit personality theories with indigenous or imported constructs: the Chinese case. Journal of Personality and Social Psychology 58 (6), 1087–1095. Younger, J.B., 1993. Development and testing of the mastery of stress instrument. Nursing Research 42 (2), 68–73.
International Journal of Nursing Studies 44 (2007) 105–114 www.elsevier.com/locate/ijnurstu
Measuring peer caring behaviors of nursing students: Scale development Chien-Lin Kuoa, Michael A. Turtonb, Jane Lee-Hsieha,, Hung-Fu Tsengc, Chin-Lung Hsua a Department of Nursing, Junior College, Fooyin University, Kaohsiung, Taiwan Department of Applied Foreign Languages, Chaoyang University, Taichung, Taiwan c Institute of Medical Research, Chang-Jung University, Tainan, Taiwan
b
Received 28 August 2004; received in revised form 20 July 2006; accepted 20 July 2006
Abstract Background: Caring is one of the most important domains of nursing research, peer caring among student nurses, and its potential effects on nurse caring behaviors remains largely unexplored. Few tools in the literature target peer caring interactions, and which were either irrelevant to our research purpose or culturally inappropriate for nursing student population in Taiwan. Objectives: The purpose of this study was to develop a culturally sensitive instrument to measure peer caring behavior from the student perspective and to offer a descriptive answer to ‘‘what is peer caring in Taiwan.’’ Design: The study uses both qualitative and quantitative methods in developing the ‘‘Peer Caring Measurement (PCM)’’ questionnaire. Settings: Students from a 5-year associate degree nursing program in a university of technology in southern Taiwan were recruited into this study. Participants: Thirteen first- through fourth-year student volunteers between the ages of 16 and 20 were interviewed to explore caring behavior in student peer-to-peer interactions. Two classes from each of the first, second, third, and fourth year students, a total of 360 students were randomly selected to assess the internal consistency of ‘‘PCM’’, and 47 first-year students were conveniently selected to examine the stability of the tool. Methods: Interviews were conducted in an unstructured manner. Qualitative data were analyzed by a constant comparative method. The questionnaire survey was used to assess the validity and reliability of ‘‘PCM’’. Results: A 17-item ‘‘PCM’’ was developed; the content validity, construct validity, and reliability of the tool were ensured by expert review, factor analysis, and internal consistency. Three factors, labeled assistant caring, academic caring, and affective caring, accounted for 63.197% of the variance. Conclusions: The ‘‘PCM’’ addressed the multidimensional construct of peer caring. It was validated in a Chinese language version and can be used in college settings to evaluate student interactions and their peer caring behaviors. r 2006 Published by Elsevier Ltd. Keywords: Peer caring; Instrument development; Nursing education; Taiwan
Corresponding author. Tel./fax: +886 7 7828258.
E-mail address: [email protected] (J. Lee-Hsieh). 0020-7489/$ - see front matter r 2006 Published by Elsevier Ltd. doi:10.1016/j.ijnurstu.2006.07.025
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
106
What is already known about the topic?
Caring
is one of the most important domains of nursing research; peer caring among student nurses and its potential effects on nurse caring behaviors remain largely unexplored. There are few tools in the literature which target peer caring interactions, and most of them are from western perspective.
What this paper adds
The
‘‘Peer Caring Measurement’’ describes the meaning of peer caring from the perspective of Taiwanese students by assessing three factors: assistant caring, academic caring and affective caring. The ‘‘Peer Caring Measurement’’ can be applied in nursing school settings to evaluate student interactions and their peer caring behaviors.
1. Introduction Researchers generally agree that caring is a complex and crucial component of nursing knowledge (Schoenhofer, 2001). Though caring is one of the most important domains of nursing research, peer caring among student nurses, and its potential effects on nurse caring behavior, remain largely unexplored. Caring theorists such as Noddings (1992) have argued that caring begins with the self and those closest to oneself. Peer caring behavior is thus an indicator of the status of a student nurse’s growth in caring attitudes and abilities. In 2002, the researchers established a concept-based caring curriculum framework for the 5-year associate degree nursing program at our university in southern Taiwan. The caring curriculum was based on two models, the meaning of Caring model and the dynamic caring model, developed out of the researcher’s prior phenomenological study of caring in a Taiwan context (Hsieh et al., 1998). The meaning of caring contains two major domains: a psychosocial domain and a caring action domain. The psychosocial domain includes the categories of love, unconditional acceptance, empathy, genuineness, respect, and treating others as family members. The Caring action domain includes the categories of providing professional behaviors, fulfilling needs, sensational approaching, self-offering, supporting, and providing a suitable environment. The new curriculum consists of six credits of required coursework, including a first-year course (two credits) entitled ‘‘Introduction to Caring,’’ a third-year course (two credits) entitled ‘‘Application of Caring Concepts’’ and a fifth-year course (2 credits) entitled ‘‘Professional
Caring.’’ Introduction to Caring covers personal and interpersonal caring, including peer caring (Lee-Hsieh et al., 2004). The purpose of the Introduction to Caring course is to create a caring atmosphere throughout the entire campus, and increase student peer-to-peer, family, and teacher–student interpersonal caring. The ultimate goal of the course is to foster student ability to self-offer in caring action. Currently, the first three stages of the curriculum development process, planning, design, and implementation, have been completed for the Introduction to Caring course. However, without the final stage, evaluation, there is no way to know whether the course has been effective in fostering student learning of caring attitudes and behavior. The ‘‘Peer Caring Measurement (PCM)’’ instrument is the final stage in the curriculum development process for this course. The purpose of the PCM to explore the peer caring behavior of students who have taken the Introduction to Caring course in order to evaluate their ability to apply the caring skills and concepts they studied to their real-life peer interactions. These concepts include such ideas as listening and expressing, encouraging and praising, accepting and trusting, empathy and respect, and love and sincerity. Two factors impelled us to create our own evaluation instrument. First, few tools in the literature target peer caring interactions. For example, Watson (2002) collected and compared 16 survey instruments. Of these, 15 examined caring behavior or abilities from the patient or nurse point of view, while only one, that of Hughes (1998), explored peer caring behavior among student nurses. Second, the instrument developed by Hughes, the ‘‘Peer Group Caring Interaction Scale (PGCIS),’’ was inappropriate for our purposes, for three reasons. First, while four of the 16 items on the PGCIS addressed clinical practice issues, the student nurses we wanted to evaluate had not yet performed their clinical practice. Second, the students in Hughes’ (1998) study had a mean age of 25.6, ranging from 19 to 51, years older than the 16–20-year-olds who comprise our junior college nursing student population. Third, based on our previous experience attempting to use evaluation instruments developed in western cultural settings (LeeHsieh et al., 2005), we expected to encounter items that were culturally inappropriate or irrelevant.
2. Literature review The educational philosopher Noddings (1984, 1988, 1992), who has written extensively on caring in education, has argued that caring begins with the care of self and then expands outward to care of intimates, strangers, distant others, and the natural and man-made world. Noddings (1984) views caring education as
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
having four components: modeling, dialogue, practice, and confirmation. Peer groups provide all four of these components. Adolescents use their peer groups as resources for developing moral judgments and values (Hartup, 1996). Relationships with peers enable adolescents to explore important caring skills such as reciprocity, self-disclosure, trust, and conflict resolution skills (Collins and Repinski, 1994). Recognizing the importance of peers in shaping adolescent views of caring, Noddings has argued that the peer group may be particularly useful in providing an environment where student nurses can learn caring. Although findings from several studies suggest that peer interactions may help student nurses learn caring, few studies have attempted to describe such effects. Hughes (1998) noted that ‘‘little attention has been focused on peer relationships as an important dimension of educational experience in schools of nursing’’(p. 206). Beck (2001) was able to locate only two studies focusing on caring among nursing students, her own of 1992 and the study of Hughes (1993). Beck (1992) argued that when nursing students experience caring from fellow nursing students, they learn caring. In her phenomenological study of 190 descriptive expressions collected from student nurses, Beck analyzed the experience of caring into four necessary constituents, authentic presencing, self-less sharing, fortifying support, and enriching effects. She further suggested (1993) that student nurses share their experiences of clinical caring with fellow students to increase familiarity with practical ways to demonstrate caring to patients, and allow students to serve as caring role models for each other. Thus, Beck’s research provides support for the idea that student nurses who care for each other will be more caring for parents as well. Caring, in her view, is contagious. Hughes (1993, 1998) developed an instrument to measure caring peer group interactions, the PGCIS. The PGCIS consisted of 16 items distributed across two factors, caring behavior (nine items), and giving assistance (seven items). Hughes recommended that a survey instrument should be evaluated by both quantitative and qualitative approaches.
3. Methodology and results This study used both qualitative and quantitative methods in constructing the ‘‘PCM’’ tool. The study was carried out from the spring of 2003 to the summer of 2004. 3.1. Phase I The initial phase of the development of the instrument consisted of three steps. First, the researchers conducted
107
interviews and analyzed the contents of the interviews. Thirteen first- through fourth-year student volunteers were interviewed to explore caring behavior in student peer-to-peer interactions. Students signed up to be interviewed after the researchers requested volunteers for the research project through announcements in their own classrooms. Open-ended interviews were used to gather data on the participant perceptions, experiences, and feelings about student peer caring behavior. The researchers used at least three students from each year as participants. Data saturation was reached with the 13th participant. 3.1.1. Ethical considerations Before interviewing the students, the researchers informed the students of the purpose of the research, and obtained written consent from each participant. The researchers also obtained participant permission to audiotape each interview for the purpose of content analysis and audit trail. 3.1.2. Step I: interview process Interviews were conducted in an unstructured manner. Each interview began by asking the participant to express their experience of peer caring behaviors. Interviews began with the question ‘‘What do you think caring is?’’ This might be followed by questions designed to elicit specific caring experiences, such as ‘‘what was your most unforgettable peer caring experience?’’ Participants were also asked to describe what kinds of caring behaviors they would like to receive from their peers. The following are examples of prompts used to obtain further details from participants: ‘‘What you mean isy’’; ‘‘Can you tell me more about that?’’; ‘‘Is there anything more you would like to tell me?’’ Interviews lasted between 45 and 90 min. 3.1.3. Step II: content analysis and instrument drafting Interview results were immediately transcribed verbatim. Data were analyzed by a modified version of the constant comparative method (Glaser and Strauss, 1967; Lincoln and Guba, 1985). The constant comparative method contains four steps: (1) comparing incidents applicable to each category, (2) integrating categories and their properties, (3) delimiting the construction, and (4) writing the construction (Glaser and Strauss, 1967). In the first step, meaningful descriptions of peer caring were identified, coded, compared, and then a preliminary classification into categories was performed. Analyzed data from each interview were constantly compared to each other in order to fit them into the most appropriate category. In the second stage, descriptions of caring were compared to the preliminary version of the definitions of the various categories. Descriptions were resorted to increase the coherence of each category. Stage three was completed when redundant categories
ARTICLE IN PRESS 108
C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
had either been deleted or integrated into existing categories. Data collection and analysis were conducted until data saturation was reached. With the concepts from the Introduction to Caring course as a foundation, the researchers used the analysis of the student interview data to construct a 25-item PCM instrument. The instrument used a six-point Likert scale (1 indicating ‘‘strongly disagree,’’ 2 indicating ‘‘moderately disagree,’’ 3 indicating ‘‘slightly disagree,’’ 4 indicating ‘‘slightly agree,’’ 5 indicating ‘‘moderately agree,’’ and 6 indicating ‘‘strongly agree’’). 3.2. Phase II 3.2.1. Step I: assessing content validity In step 1, the 25-item PCM instrument was submitted to a panel of ten experts, including five experts from the fields of education and nursing, and five instructors who had taught the Introduction to Caring course at our university, in order to evaluate its clarity and relevance. A four-point scale was used. The panel determined whether each item was appropriate, accurate, and representative (1 ¼ irrelevant and should be deleted; 2 ¼ seemingly relevant but large-scale revision was required; 3 ¼ relevant but in need of small adjustments; and 4 ¼ relevant, clear and precise). The instrument was then revised accordingly. 3.2.1.1. Item wording revisions. For the sake of readability and clarity, revisions were made to eliminate awkward wording in items 1, 4, 5, 11, and 17. For example, the item ‘‘when a student is in need, other students will keep him or her company, and not leave her alone with her problems’’ was changed to ‘‘When a student is lonely and needs company, other students will keep him or her company.’’ 3.2.1.2. Content revisions. The item ‘‘For the good of the class, or for the good of a particular individual, students will offer suggestions or attempt to persuade the individual to change their behavior’’ included two different levels, the group and the individual. To clarify this, it was changed to ‘‘for the good of a specific individual, classmates will offer suggestions’’. The experts felt that the language of the item ‘‘providing needed extra help is the responsibility of the teacher’’ was unclear, for students would take the word ‘‘extra’’ to mean that the help was beyond their ability to provide and should be given by the teacher, not peers. If no help was given, it was the teacher’s responsibility, not because they as peers were uncaring. Hence, the item failed to measure student willingness to help, as was intended. Consequently, it was revised to ‘‘helping students who need help is solely the teacher’s responsibility.’’ One expert felt that the item ‘‘interpersonal trust between students’’ was not relevant to caring, while another
expert felt that ‘‘trust’’ was too broad an idea, and the question was not clear enough. Trust is generally seen as an important component of caring. A foundation of interpersonal trust can raise the willingness to offer caring (Hsieh et al., 1998). Due to this understanding, the researchers decided not to delete this item, instead revising it to ‘‘the students trust each other and do not lie to one another.’’ The item ‘‘students chat and share their feelings with each other,’’ contained two different levels of intimacy, chatting and sharing of feelings, and was therefore split into two items by the researchers, ‘‘students have classmates to chat with’’ and ‘‘students have classmates to share their feelings with.’’ Most of the experts felt that the item ‘‘there is intense competition between the students’’ simply described a common phenomenon in Taiwan schools, given the ruthless competition for places in good schools under the local testing system. Generally, competition may be positive or negative. This item was thus revised to ‘‘there is negative competition between the students.’’ 3.2.1.3. Item order revisions. Most of the experts argued that the content of items originally designated 1, 2, 8, and 20 was quite similar, and suggested that some be deleted. Based on their reasoning in designing these items, the researchers believed that item 1, ‘‘when a student encounters difficulties, other students will notice,’’ and item 2, ‘‘when a student asks for help, other students are happy to help,’’ addressed different aspects of the interpersonal action, one passive (being noticed), the other active (asking for help). Though item 8, ‘‘when a student encounters difficulties, other students will offer guidance and solutions’’ and item 20, ‘‘when a student encounters difficulties, other students will offer suggestions’’ seem similar, the researchers felt that they showed different levels of caring, as item 8 addressed the offering of concrete solutions for difficulties, unlike item 20. All four items were retained. However, to clarify the different levels of caring represented by these questions, the researchers placed them in series, one after another, so their similarities and differences would be immediate and obvious (items 1, 2, 3, and 4 in the resulting 27-item version of the instrument). After revising the wording, content, and item order, the original 25 items of the PCM instrument became 27 (Table 1). 3.2.2. Step II: assessing face validity The language of the instrument was then reviewed for clarity by the 13 students who were originally interviewed for the study. During the process of completing the survey, the students did not express any problems in
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
109
Table 1 27-item version of PCM from phase II 1.a 2.a 3. 4.a 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.b 15. 16. 17.b 18. 19.a 20.a 21. 22. 23.b 24.a 25. 26. 27.b a
When a classmate encounters difficulties, other classmates notice. When a classmate encounters difficulties, other classmates offer solutions. When a classmate encounters difficulties, other classmates take the initiative to help When a classmate asks for help, other classmates are happy to help Classmates are concerned with each other’s health. When a classmate is sick, other classmates accompany them to get medical help. When a classmate is lonely and needs company, other classmates are willing to keep her company. If there is a conflict between classmates, other classmates help smooth things out. Classmates respect the different opinions of other classmates. For the good of a specific individual, classmates will offer suggestions Classmates discuss assignments with each other. Classmates who perform well academically are willing to instruct students who need help. When a classmate encounters difficulties with assignments, other classmates offer suggestions. Helping students who need help is solely the teacher’s responsibility (reverse score item). Classmates listen to each other’s troubles and problems. Classmates trust each other and do not lie. Certain classmates are excluded by other classmates (reverse score item). Classmates are concerned with each other’s needs. Classmates help other classmates only when it is in their best interest to do so (reverse score item). Classmates have someone to chat with. Classmates have someone to share feelings with. There is deep friendship between classmates. Classmates are too busy to help each other (reverse score item). Classmates praise each other. Classmates encourage each other. Classmates share notes and references with each other. There is negative competition between classmates (reverse score item).
Deleted due to average inter-item correlations less than .3 or greater than .7 (6 items). Deleted due to item-to-total correlations less than .3 (4 items).
b
understanding the item wording and meaning, demonstrating its face validity. 3.2.3. Step III: assessing internal consistency Two classes from each of the first-, second-, third-, and fourth-year students at our school, a total of 360 students in eight classes (approximate 45 per class), were selected and given the tool. Classes were selected randomly. Class instructors informed the students of the purpose of the study and distributed and collected the questionnaires during last 10 min of class. Students were also informed that they did not have to fill out the questionnaire, and that there would be no punishment for those who decided not to participate. Questionnaires were anonymous and participant consent was obtained on a tear-off form. It took roughly 10 min to complete the questionnaire. Because the questionnaires were collected on the spot, the return rate was 100%. Data from the 360 valid returned questionnaires were analyzed using SPSS for Windows 10.0 (SPSS, 2000 Statistical Package for the Social Sciences) Internal consistency was measured by Cronbach’s a coefficient. Individual item characteristics were obtained by
corrected average inter-item correlation coefficients, inter-item correlations, and item-to-total correlations (Ferketich, 1991; Reeve, 1994; Younger, 1993). Items with an item-to-total correlation of less than .3 were deleted, leaving 23 items (Table 2). Next, inter-item correlation coefficients were obtained. Items with an inter-item correlation coefficient of less than .3 or greater than .7 were deleted (Ferketich, 1991), leaving 17 items (Table 1). The 17-item PCM Instrument had a Cronbach’s a of .9347. 3.3. Phase III Using SPSS for Windows 10.0, the researchers then examined the 17-item PCM using exploratory factor analysis with principal component and varimax rotation. Three factors emerged, accounting for a total of 63.197% of the variance (Table 3). Each factor was labeled based on a theme appropriate for the content of the factor. The first factor, comprising nine items, was labeled Assistance Caring. The second factor, composed of four items, was labeled Academic Caring. The third factor, also composed of four items, was labeled Affective Caring.
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
110
Table 2 Descriptive statistics and reliability analysis of the 27-item version of PCM
Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22 Q23 Q24 Q25 Q26 Q27
Mean
Std. Dev.
Item-total correlation
Alpha if item deleted
4.24 4.39 4.37 4.57 4.73 5.03 4.32 4.18 4.12 4.26 4.83 4.74 4.60 4.97 4.71 3.74 4.13 4.33 3.38 5.05 4.92 4.52 4.01 4.40 4.64 4.66 3.92
1.020 1.076 1.081 1.079 1.030 .993 1.140 1.173 1.217 1.148 1.042 1.144 1.140 1.312 1.002 1.263 1.366 1.215 1.370 1.096 1.024 1.229 1.326 1.056 1.078 1.173 1.400
.5576 .6352 .6786 .6318 .6519 .5583 .6452 .6149 .6014 .6264 .6151 .6007 .6411 .0908 .6372 .5054 .0748 .5669 .4011 .4470 .6179 .6545 .2403 .6730 .7120 .6386 .2722
.9140 .9127 .9120 .9127 .9126 .9140 .9124 .9128 .9130 .9127 .9131 .9131 .9125 .9223 .9128 .9147 .9256 .9136 .9169 .9156 .9131 .9121 .9197 .9122 .9115 .9124 .9194
Table 3 17-item version of PCM, exploratory factor analysis with principal component and varimax rotation Factor 1 Q8 Q7 Q9 Q3 Q10 Q5 Q16 Q18 Q6 Q11 Q12 Q13 Q26 Q21 Q22 Q15 Q25 Eigenvalue Percent of total variance explained Cumulative percent
Factor 2
Factor 3
.754 .743 .719 .695 .684 .646 .567 .520 .503 .797 .797 .727 .643
8.415 26.295
1.315 19.645
.831 .760 .650 .529 1.012 17.257
26.295
45.940
63.197
had the highest factor loading (.831), followed by Item 22, ‘‘There is deep friendship between classmates’’ (.760). 3.4. Phase IV
3.3.1. Factor 1: assistance caring This factor involves offering assistance and suggestions when needed by classmates, and helping classmates handle problems. Item 8, ‘‘When there is a conflict between classmates, other classmates will help smooth things out,’’ had the highest factor loading, (.754), followed by item 7, ‘‘When a classmate is lonely and needs company, other classmates are willing to keep him or her company’’ (.743). 3.3.2. Factor 2: academic caring This factor involves academic information sharing and peer instruction. Item 11, ‘‘Classmates discuss assignments with each other,’’ and item 12 ‘‘Classmates who perform well academically are willing to instruct students who need help’’ had the highest factor loading (both .797). 3.3.3. Factor 3: affective caring This factor involves peer ability to listen, share, encourage, and support each other. Item 14, ‘‘Classmates have someone to share their feelings with,’’
To examine the stability of the PCM, the researchers gave it to 47 first-year students in the first and again in the third week of the semester. The difference between the first and second test was either 1 or 0 for at least 80% of participants on all 17 items (six-point Likert scale). The Pearson’s correlation coefficient (g) between the first and the second test was .72 (po:001), and the regression slope (b) was .724 (po:001). These results indicated the stability of the instrument.
4. Discussion The conceptual-theoretical basis of our research was provided by the work of Noddings (1984, 1992), who argued that caring begins with care of the self and those closest to oneself. The PCM was constructed through a step-by-step development process (Table 4). It is capable of meaningfully evaluating student attitudes toward their own caring behavior and that of others.
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
111
Table 4 Phases and steps of the peer caring measurement (PCM) development process Development process
Purpose
Phase I
Develop a new instrument for the measurement of peer caring behavior from the nursing student perspective Develop a set of peer caring behaviors (1) Develop items based on the interview data (2) Develop a preliminary version of the instrument Assess content validity and face validity Assess clarity and relevance of the items on a 4-point rating scale Assess clarity of wording and face validity of the 2nd version instrument Assess internal consistency of the 2nd version instrument
Step I Step II
Phase II Step I Step II
Step III
Phase III
Phase IV
Methods/data
Items remained
Student interviews Literature review, content analysis, and item construction
25 items
Expert analysis N ¼ 10
27 items (item wording, content and item order revisions)
Student assessment N ¼ 13
27 items
Test (N ¼ 360) Cronbach’s a, inter- 17 items (deleted 10 items) item correlations, item-to-total correlations (1) Investigate the construct validity Factor analysis (N ¼ 360) 17 items of the final version 17-item instrument (2) Define conceptual categories for the factors Assess instrument stability Test and retest (N ¼ 47) 17 items
Comparison of this instrument with that of Hughes (1998) yields a number of important divergences that reflect differing cultural practices in education, caring, and behavior. A key difference of the Taiwanese educational system lies in the definition of ‘‘classmate.’’ In Taiwanese colleges, students are typically enrolled in a ‘‘Class’’ when they enter college. For example, if 100 students enter the English program that year, they might be divided into two ‘‘Classes’’ of 50 each, typically designated 1A and 1B, or 1-1 and 1-2. The maximum size of both the intake and the Class is determined by the Taiwan Ministry of Education. These Classes then become surrogate families for the students throughout their college career. They are supposed to engage in collective activities such as trips and outings together, take care of each other, be closer to each other than to other students at the same university, and present a united front toward the world. Students in a Class take all of their required courses together, year after year, with the university dictating when they will take them, which students will be in class together, and which teachers they will take. For example, all 900 first-year students (18 Classes) take the required Introduction to
Caring course, and each Class will take it as a Class, all 50 together, 1-1 with instructor Lee, 1-2 with instructor Wang, and so forth. Thus, both informally (through expectations about behavior), and formally (through the university administration and course framework), students in a Class spend most of their weekdays together. In effect, a Class is more or less a potential caring group. Thus, when our questionnaire asked students to consider peer caring behavior from the standpoint of ‘‘classmates,’’ the term designates a specific and formally defined relationship. ‘‘Classmates’’ are those in the same group of 50. It is not, as it might be in the US, a loose descriptor of someone who happens to be in school or in a course with oneself. Hughes’ PGCIS used the terms ‘‘students at this school’’ or ‘‘classmates’’ when it asked about peer caring behavior, as if they were interchangeable. We chose the term ‘‘classmate’’ to reflect the fact that in a Taiwan college students have more limited contact with students outside their Class. Hughes’ (1998) PGCIS did not ask about conflicts between students. Because each Class involves the same 50 students year after year, intrapersonal conflicts represent serious problems, both for the poisoning effect
ARTICLE IN PRESS 112
C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
they have on interpersonal relations within the Class, and on the image the Class presents to the outside world. The importance of this issue can be gauged from the fact that Item 8, ‘‘If there is a conflict between classmates, other classmates help smooth things out,’’ had the highest factor loading, .754, in Assistance Caring. Even translating this item threw into sharp relief the differing cultural assumptions of the researchers. The American automatically selected the word ‘‘resolve,’’ reflecting his assumption that conflicts are things that demand solving. However, the Chinese word ‘‘tiao jie’’ does not call for a solution, but merely a restoration of the previous placid surface of things, a smoothing over rather than an ending (the primary authors are comfortable in both languages). Item 7, ‘‘When a classmate is lonely and needs company, other classmates are willing to keep him or her company,’’ had the second-highest factor loading in Assistance Caring. Loneliness was not mentioned in Hughes’ PGCIS. In all probability, this is due to the higher collectivism of Chinese culture (Hofstede, 1980, 1983; Duncan, 1994), in which people who are by themselves are more likely to be viewed as needing, and expecting, company. Cultural preferences for collectivism also explain the wording of item 10, ‘‘for the good of a specific individual, classmates will offer suggestions.’’ The wording does not quite capture the implied meaning, which is that the individual’s good can be defined either by their own needs, or the needs of the Class. The Chinese wording thus contains an element of social reinforcement of group norms that is much less obvious to the English reader. Cultural preferences in communication appeared in the idea of listening versus talking. Hughes’ questionnaire contained 3 items that referred to ‘‘talking’’ with others and none that addressed ‘‘listening.’’ Item three of Hughes’ instrument ‘‘Students at this school talk with each other about their problems and concerns’’ contrasts with our item 15, ‘‘Classmates listen to each other’s troubles and problems.’’ In our interviews, almost every student gave an example of someone listening to them as an instance of authentic caring. The word typically used to describe such listening, qing ting, expresses the idea of a listening from the heart in which the self is offered in caring for others. Another cultural difference not captured by Hughes’ PGCIS was the use of intermediaries, especially when a student is sick. Invariably, when a student requires medical help, regardless of the seriousness of the problem, a second student will take her to the clinic. Thus, item 6, ‘‘When a classmate is sick, other classmates accompany them to get medical help,’’ had no parallel on Hughes’ questionnaire. Chinese culture is legendary for its emphasis on academic performance. Thus, one important aspect of
peer caring is offering help to classmates having academic difficulties. Collective face is also at stake, since average scores for each Class are often announced at year-end meetings and publicly posted, to encourage performance. In our study, academic caring was a factor in its own right, and the four items in it all had relatively high factor loadings (.797, .797, .727, and .643). By contrast, Hughes’ PGCIS findings revealed relatively lower factor loading for ‘‘share class notes, books, and articles,’’ and ‘‘giving suggestions about school assignments.’’ Item 1 of Hughes’ PGCIS, ‘‘Students at this school anticipate the needs of their classmates,’’ had no counterpart on our PCM. No student mentioned this peer caring behavior in the interview phase.
5. Recommendations The process of constructing and implementing this instrument has yielded several important recommendations for future research into peer caring. The development process for this instrument underscores the findings of our previous study (Lee-Hsieh et al., 2005) that while caring may be universal among humans, its expression is entirely local. Psychologists working in Chinese cultural settings have long expressed dissatisfaction with the limited applicability of western measurement instruments (Yang, 1987; Yang and Bond, 1990), and we can only echo that. It is apparent that behavioral measurement instruments must be developed within particular cultural settings, a conclusion that implies a development process that partakes of both qualitative and quantitative approaches, each informing the other. Though this measurement instrument is aimed at nursing students, the items therein examine psychological, social, and behavioral aspects of caring. These include empathy, sincerity, respect, praise, and communication (Lee and Fang, 2000; Hsieh et al., 1998). Because it looks at aspects of caring with broad application, the instrument may be used in almost any type of Chinese school setting with many different kinds of students. It may be used as a before/after test to explore the effects of coursework in caring, as well as a tool administrators can use to take a campus-wide snapshot of the status of caring at their university.
6. Limitations This instrument was developed to assess peer caring among student nurses in a junior college in Taiwan between the ages of 16 and 20. It may not be appropriate for other age groups, institutional, and cultural settings. The original PCM is in Chinese.
ARTICLE IN PRESS C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
7. Conclusion This study used qualitative methods to develop a PCM questionnaire. The content validity, construct validity, and reliability of the 17-item PCM were ensured by expert review, factor analysis and internal consistency, respectively. The PCM was validated in a Chinese language version. A translated version in any other language would also require appropriate validation. The 17-item PCM instrument developed in this study addressed the multidimensional construct of peer caring. It can be used in college settings to evaluate peer caring behavior. It is easy to administer and takes less than 10 min to complete. This study explores peer caring from the perspective of students and offers a descriptive answer to ‘‘what is peer caring in Taiwan.’’ For nursing programs hoping to develop specific teaching–learning objectives for their nursing curricula, this study offers directions to explore in understanding the meaning of caring in nursing as well as teaching strategies. Instructors could motivate and encourage nursing students to provide peer caring behaviors by using the PCM to evaluate the peer caring in their institutions. If school administrators seek to promote and support a caring atmosphere on campus, they may use the PCM to evaluate student interactions and the effectiveness of peer caring initiatives. The cross-cultural issues raised by this study suggest several directions for future research. For example, peer caring is only one component of a caring campus; research is needed into student–family, faculty–student, faculty–faculty, and other caring relationships on campus. This study has made contributions to caring education in Taiwan by providing a description of the meaning of peer caring and offering a way to evaluate it. We hope that further exploration of this vital and interesting topic will enhance our understanding of the issues. Acknowledgements The researchers would like to thank the National Science Council for the grant that enabled this study, and our students for their unfailing patience and openness. A special debt of thanks is owed to our dedicated and overworked assistant, Ms. Mei-Ling Chen. Special thanks also to Dr. Clyde Warden of National Changhua University for his suggestions. References Beck, C.T., 1992. Caring among nursing students. Nurse Educator 17 (6), 22–27.
113
Beck, C.T., 1993. Caring relationships between nursing students and their patient. Nurse Educator 18 (5), 28–32. Beck, C.T., 2001. Caring within nursing education: a metasynthesis. Journal of Nursing Education 40 (3), 101–109. Collins, W.A., Repinski, D.J., 1994. Relationships during adolescence: continuity and change in interpersonal perspective. In: Raymond, M., Gerald, R.A., Thomas, P.G. (Eds.), Personal Relationships During Adolescence. Sage Publications, Thousand Oaks, pp. 7–36. Duncan, W.J., 1994. Why some people loaf in groups while others loaf alone. The Academy of Management Executive 1 (8), 79–80. Ferketich, S., 1991. Focus on psychometrics aspects of item analysis. Research in Nursing and Health 14, 165–168. Glaser, B.G., Strauss, A.L., 1967. The Discovery of Grounded Theory. Aldine, Chicago. Hartup, W., 1996. The company they keep: friendships and their developmental significance. Child Development 67, 1–13. Hofstede, G., 1980. Culture’s Consequences: International Differences in Work-Related Values. Sage, London and Beverly Hills. Hofstede, G., 1983. Dimensions of national cultures in fifty countries and three regions. In: Deregowski, J.B., Dziurawiec, S., Annis, R.C. (Eds.), Expiscations in CrossCultural Psychology. Swets and Zeitlinger, Lisse, Netherlands, pp. 335–355. Hsieh, J., Fang, Y.Y., Chiang, L.Y., 1998. An exploration of the meaning of caring in Taiwan. International Journal for Human Caring 2 (2), 12–22. Hughes, L., 1993. Peer group interactions and the students perceived climates for caring. Journal of Nursing Education 32, 78–83. Hughes, L., 1998. Development of an instrument to measure caring peer group interaction. Journal of Nursing Education 37, 202–207. Lee, J.C., Fang, Y.Y., 2000. Caring: the core of nursing education. The Journal of Nursing 47 (3), 22–27. Lee-Hsieh, J., Kuo, C.K., Tsai, Y.H., 2004. An action research on the development of a caring curriculum in Taiwan. Journal of Nursing Education 43 (9), 391–400. Lee-Hsieh, J., Kuo, C.K., Tseng, H.F., Turton, M.A., 2005. Development of an instrument to measure caring behaviors in nursing student in Taiwan. International Journal of Nursing Studies. Lincoln, Y.S., Guba, E.G., 1985. Naturalistic Inquiry. Sage Publications, Beverly Hills, CA. Noddings, N., 1984. Caring: A Feminine Approach To Ethics and Moral Education. University of California Press, Berkeley, CA. Noddings, N., 1988. An ethic of caring and its implications for instructional arrangements. American Journal of Education 96 (2), 215–230. Noddings, N., 1992. The Challenge to Care in School: An Alternative Approach to Education. Teachers College Press, New York. Reeve, M.M., 1994. Development of an instrument to measure effectiveness of clinical instructors. Journal of Nursing Education 3, 15–20.
ARTICLE IN PRESS 114
C.-L. Kuo et al. / International Journal of Nursing Studies 44 (2007) 105–114
Schoenhofer, S.O., 2001. Infusing the nursing curriculum with literature on caring: an idea whose time has come. International Journal for Human Caring 5 (2), 7–14. Watson, J., 2002. Assessing and Measuring Caring in Nursing and Health Science. Springer, New York. Yang, K.S., 1987. Chinese personality and its change. In: Bond, M.H. (Ed.), The Psychology of the Chinese People. Oxford University Press, New York, pp. 106–170.
Yang, K.S., Bond, M.H., 1990. Exploring implicit personality theories with indigenous or imported constructs: the Chinese case. Journal of Personality and Social Psychology 58 (6), 1087–1095. Younger, J.B., 1993. Development and testing of the mastery of stress instrument. Nursing Research 42 (2), 68–73.