The Cultural Dependence of the Ethical Sensitivity Scale Questionnaire: The Case of Iranian Kurdish Teachers

A good theory-based tool for measuring ethical sensitivity, which is usable in different contexts, is scarce. In this study, we examined the Ethical Sensitivity Scale Questionnaire (ESSQ) in line with its seven-dimension structure. The scale was presented to a sample of 556 Iranian Kurdish teachers in primary, middle, and high schools. A confirmatory factor analysis was conducted to scrutinize the original factor structure of the ESSQ. The results confirmed that the ESSQ supports a reasonable model fit to study the seven dimensions of ethical sensitivity as it was developed in the original study. However, some modifications were conducted to free high error covariance between four pairs of items in the scale. This modification increased the fit indices and thus resulted in a good model fit. In addition to examining the satiability of the ESSQ, a further analysis showed that the level of ethical sensitivity in the targeted sample was high.


Introduction
Ethical sensitivity is a fundamental element of human moral conduct.Ethical sensitivity was coined for the first time by Rest [1] and it is the first important component of his 4component moral action theory.In a general sense, ethical sensitivity is the identification of salient aspects of a situation that involves the "good" and the "bad" of others.Weaver et al. [2] define ethical sensitivity as "the capacity to decide with intelligence and compassion, given uncertainty in a care situation. ..with additional ability to anticipate consequences and courage to act."According to Tirri and Nokelainen [3] "to respond to a situation in a moral way, a person must be able to perceive and interpret events in a way that leads to ethical action".The central feature of ethical sensitivity is the ability to read a caring situation in order to respond with an ethical action, that is, a human conduct whereby the others' benefits and loses are taken into consideration.As such, the primary assumption in research on ethical sensitivity is that something one might do or is doing can affect the welfare of someone else.Discerning that a situation requires a moral response is the first step in the process of moral action [4].
Recently there has been a significant conceptual insight into the topic in different studies [2,5].As an empirical concept, Weaver et al. [2] found that ethical sensitivity has five basic aspects in various professions and domains."Attributes" which refers to basic characteristics of the concept includes moral perception, affectivity, and dividing loyalties."Moral perception involves awakening and particularizing, which allows professional to perceive client situational needs. ... Affectivity is a relational component based on professional putting oneself in the place of clients and identifying comparable reactions. ..and dividing loyalties refers to strategies of interpretation, justification, and flexibility to deal with authenticate and contradict moral perception".Definition is the second aspect of ethical sensitivity, where the research shows there has been a multiplediscipline description and definition of ethical sensitivity but "decision-making within the uncertainty of professional practice" has been considered as a central feature to define ethical sensitivity.Preconditions is another aspect of ethical sensitivity, which refers to client suffering and vulnerability and professionals' perspectives of uncertainty, receptivity, responsiveness, and courage to deal with this suffering.As another aspect of ethical sensitivity, "Boundaries" passes on what is and is not, in this concept, based on previous aspects of ethical sensitivity.Moreover, ethical sensitivity has "outcomes" such as "integrity, comfort, well-being, personal growth, professional self-transcendence, and practical wisdom" for clients [2].
These conceptual insights into the topic encouraged researchers in various domains to conduct empirical studies on ethical sensitivity.This trend has shifted researchers' attention from Kohlberg's moral judgment to ethical sensitivity.Jordan states that one reason for the recent increase in attention to moral sensitivity may be that there are now several methods for measuring this construct, for example, the Dental Ethical Sensitivity Test, ethical sensitivity in television viewing, the Racial Ethical Sensitivity Test, moral sensitivity in counseling supervision, moral sensitivity in social work, ethical sensitivity in accounting, and several other measures [6].Using these measures, a significant body of research has been conducted to examine ethical sensitivity in various professions such as nursing and medicine [7][8][9], news [10], and business and accounting [11].These studies have also addressed different issues such as gender, cultural, nationality, and racial differences in regarding to ethical sensitivity.
As can be seen, most of these measures are domain specific and may not be applied in a different context.Drawing on Narvaez and Endicott [12], Tirri and Nokelainen have developed a measure, the Ethical Sensitivity Scale Questionnaire (ESSQ), which is more general and thus can be applied in different professional contexts.Using this measure, this research examined teachers' ethical sensitivity [13].
Numerous studies have showed that teaching is a moral activity by nature.Narvaez [14] believes that "because humans are wired for emotion signaling and social motivation, a caring, supportive teacher more easily foster students' empathy and prosocial behavior as well as motivation to learn. ... Children with poor early care will have brains that are less flexible, integrated, and attentive, represented by poor attachment but with patience and supportiveness, these children can be reached."Higgins stated that "talk about teaching is talk about the present and the past and the future of persons. ... Teaching is an act of caringcaring for the world and another human being.Teaching requires responding to the student as a worthy human being."[15] According to Hansen, every moment of the classroom life emits moral messages.Thus "If teachers sought to be constantly alert to the moral impact of their work, they might find themselves suddenly walking on eggshells in the classroom, fearful that at any moment a single word or deed might morally injure a student" [16].In a philosophical term, the moral foundation of teaching can be met with what has been called "teachers' manner".In line with this reflection, empirical evidence shows that along with method, which is used to convey knowledge to the learners, teachers should have good "manner" to convey virtue [17][18][19][20].
Narvaez has proposed a popular framework that can help educators gain insight into the moral foundation of teaching and teachers' manner.According to this framework, a moral agent and particularly a teacher should have four basic ethical skills namely ethical sensitivity, ethical judgment, ethical focus, and ethical action in order to develop moral character of students [14,21].Teachers with these ethical skills provide a sustainable pedagogical climate, in which children "foster secure attachment, develop more agreeable personalities, earlier conscience, and more prosociality".In this framework, ethical sensitivity is the "opening door" for conducting moral action because it deals with "concern for others".Most of pedagogical decisions in classroom life involve taking care of students in different ways.Different dimensions of ethical sensitivity are very critical skills with which teachers can regulate the emotional, sociological, and pedagogical aspects of their practice."Reading and expressions feeling" and "caring by connecting to others" help teachers provide an effective emotional relationship with students.With these two skills, teachers can read the existing emotional conditions of students and then care about them with appropriate actions.In addition, "Taking the perspective others," "working with interpersonal and group differences", and "Preventing social bias" are necessary skills to take social aspects of the students into consideration in different ways.These skills can be effective characters for teachers in a multicultural teaching context where students have different gender, ethnic, religious, and cultural backgrounds."Generating interpretation and options" and "Identifying the consequences of action" are other two dimensions of ethical sensitivity that can improve the pedagogical atmosphere of the teaching context in two ways."Generating interpretation and options" helps teachers be flexible in dealing with students when they try to have various pedagogical actions to deal with students with different interests and capability."Identifying the consequences of actions" can be an effective basis for reflective teaching when teachers try to be sensitive to the outcomes of their actions in very critical situations that involve ethical considerations.Figure 1 illustrates how dimensions of ethical sensitivity are connected to teaching as a moral activity.
According to Figure 1, teaching as a moral practice has three important bases, including pedagogical, emotional, and social aspects.Different dimensions of the ethical sensitivity are linked to three bases of teaching as moral practice.If teachers fail to perceive and recognize the moral issues of their work regarding the students, how then can they conduct a moral action?Higgins argued that teaching will turn to be a "care-less activity" if teachers fail to perceive the ethical loading of their pedagogical decisions [15].Empirical research has shown that ethical sensitivity, especially the skill to read and express emotions, is an important skill for teachers in establishing caring relationships with their students and their families, particularly in schools that have an increasing cultural diversity amongst students [22].Even though teaching is considered to be a moral action, there seems to be a paucity of empirical research on its first component, that is, ethical sensitivity, particularly in a crosscultural and international context.

Teaching and Teacher Education in Iran
Basic education is compulsory for 12 years in Iran and it is highly centralized and thus governed by the Ministry of Education.It has three different levels of schooling, primary, middle, and high school.Primary school (Dabestan) starts at the age of 7 and lasts for 5 years.Middle school, also known as the orientation cycle (Rahnamayi), goes from the sixth to the eighth grade.High school (Dabirestan) is divided between theoretical (science, mathematics, and humanities) and vocational/technical sections, each program having its own specialties; in the theoretical section, the students have one year studying as a preuniversity course [23].In line with its educational system, teacher education in Iran is also centralized in terms of its structure and curriculum.In general, particular teacher training centers and universities are responsible for preparing teachers for K-12 education.For primary and middle schools, teachers must generally have a postdiploma (associate degree, i.e., two years further education after graduating from high school); high school teachers must have a bachelor's degree in various subject matters [24].Primary teachers mainly receive pedagogical education to deal with basic subjects such as science, mathematics, reading, writing, religion, and Persian literature.In order to teach in middle and high schools, teachers, however, must complete both pedagogical and subject-specific education in the different subject matters they intend to teach.Moral competencies are also highlighted in Iranian teacher education.For all three levels of schooling, teachers must complete a few courses regarding the morality, specifically the moral principles of Islam [25].In addition, one of the main necessary requirements needed to become a teacher is to have moral and ideological competency [24].In other words, teachers must show their faith in Islam and Islamic values.This means that teachers' moral competencies should be based on Islamic values and principles.This institutional tendency to integrate Islamic ideology into education and teacher education was the consequence of the so-called Islamic revolution in Iran, which occurred in 1979.According to Cheng and Beigi, "this revolution brought a new era in Iran in which the new Islamic republic began to Islamize many sectors which had secular foundations prior to revolution" [23].The most striking feature of postrevolution schools is the moral education teacher (Morabbiye tarbitati).This teacher does not teach particular subjects but is responsible for teaching Islamic moral values at various religious events in schools [26].The primary qualification of this teacher is his/her knowledge of, and faith in, Islam.Other teachers are not allowed to discuss their own values and principles with students, except those regarding to Islamic moral values.

Specific Research Questions
Considering the context of Iran's teacher education, this paper examines the ethical sensitivity of Kurdish teachers in Iranian's context.Previous scholarship shows that ethics in teaching includes many culture-specific issues, for example the religious worldviews of teachers and students.In Iran's context teaching is influenced by Islamic worldview, which might have implications also for teachers' professional morality including their ethical sensitivity.In the Americanbased theory behind our instrument used [12], the Ethical Sensitivity Scale Questionnaire (ESSQ), ethical sensitivity is operationalized to include seven dimensions with concrete competencies in each dimension.The concrete skills for ethical sensitivity in the context of classroom teaching are discussed in Narvaez [27].The earlier empirical testing with the ESSQ includes Finnish students and teachers [13,28].Finnish context is very American-like European culture with an influence of Christian worldview and Protestant ethics among teachers [29].However, the instrument has been developed to be used in multicultural contexts and our goal is to test the culture-dependent aspects of the instrument with this specific population of Kurdish teachers whose worldviews differ very much from the Western teachers we  [12] the ESSQ measures seven dimensions of ethical sensitivity: reading and expressing emotions, taking the perspectives of others, caring by connecting with others, working with interpersonal and group differences, preventing social bias; generating interpretations and options, and identifying the consequences of actions and options.The measure consists of 28 items; each dimension is measured with 4 items.In the initial investigation on the psychometric properties of the measure, Tirri and Nokelainen found that, due to small sample size, there was a "quite low" level of common variance between items when inter-item correlation was examined [3].However, the direction of all 378 possible correlation combinations but one (i.e., item es1 4 with es7 28 had negative correlation; r = 0.01) was positive.Such positive correlations between items supports Naravaez' operationalization of ethical sensitivity [3,12].Further, reliability analysis of the dimensions showed good results: "Caring by connecting to others" and "Working with interpersonal and group differences" had the highest reliabilities (α = 0.78; α = 0.75); "Reading and expressing emotions" and "Preventing social bias" showed the lowest rates (α = 0.54; α = 0.50) [3].

Results
In line with the main research task, a confirmatory factor analysis (CFA) was conducted to examine the stability of the factor structure of the ESSQ, using LISREL 8.72.In other words, CFA was carried out to confirm the underlying seven structure dimensions of the ESSQ.In examining the model fit, we supported the model χ 2 statistic with both absolute and incremental fit indices [30,31].Absolute fit indices examine how well an original model reproduces the sample data under study.For this case, we have reported the root mean square error of approximation (RMSEA) and the standardized root mean square residual (SRMR).
Incremental fit indices assess model fit by comparing the target model with a baseline model.According to Hulpia et al., "typically, the null model in which all observed variables uncorrelated is used as a baseline model" [30].Regarding the incremental fit indices, the four most reported ones, including the comparative fit index (CFI), the goodness offit-index (GFI), the incremental fit index (IFI), and the normed fit index (NFI), are reported in our study.Conducting a CFA with a seven-factor structure of the ESSQ resulted in reasonable model fit (with χ 2 = 1248.57[df = 329; P < 0.00], SRMR = 0.064, RMSEA = 0.071 with a 90% confidence interval of 0.067 and 0.075, CFI = 0.86, GFI = 0.93, IFI = 0.93, NFI = 0.90).Different studies have suggested that for SRMR a value of 0.08 or lower indicates a good fit [31,32].For RMSEA a value less than 0.08 [33] explains a reasonable model fit, and more strictly values less than 0.06 shows a good model fit [32].
Considering incremental fit indices, it is generally suggested that a value close to 0.90 or above indicates a good model fit [30].As the CFA's results showed, most of the indices had satisfactory values and they confirmed that the seven-factor structure of the ESSQ had a reasonable model fit with the sample data.
In order to improve the goodness fit of the model, a further investigation was conducted to modify the model."The modifications are justified on the basis of assumed covariations between the errors of variances produced by the influences of social desirability and semantic similarity" [33].The results of examining the modification indices suggested high error covariance between the following four pairs of items: es7 27 (I am aware of the ethical issues I face at school) and es7 28 (I am better than other people in recognizing new and current ethical problems); es5 17 (I recognize my own bias when I take a stand on ethical issues) and es5 18 (I realize that I am tied to certain prejudices when I assess ethical issues); es3 9 (I am able to cooperate with people who do not share my opinions on what is right and what is wrong) and es3 10 (I tolerate different ethical views in my surroundings); es6 23 (I am able to create many alternative ways to act when I face ethical problems in my life) and es7 28.Accordingly, these four cases which had the highest error covariance between items were left out from the model.When these errors were freed, the results showed a significant improvement in the values of all fit indices (χ 2 = 1055.52[df = 325; P < 0.00], SRMR = 0.06, RMSEA = 0.06 with a 90% confidence interval of 0.059 and 0.068, CFI = 0.88, GFI = 0.94, IFI = 0.94, NFI = 0.92).This modification resulted in a good model fit of the sample data with regard to the seven dimensions of the ESSQ suggested in the original study [3].

Reliabilities of the ESSQ Scores
. Using Cronbach's Alpha coefficient the reliability of the ESSQ was examined.The results showed that the overall reliability of the questionnaire was sufficiently high (α =0.84).Researchers suggest that the overall reliability of the test score should be 0.80 or higher [30].From this, we can conclude that the test ESSQ had a high internal consistency with this sample.In addition, the reliability of each of the seven dimensions is presented in Table 2; the dimensions 5 (Preventing social bias) and 1 (Reading and expressing emotions) resulted in the lowest reliabilities (with α = 0.45; α = 0.46, resp.)Further inspection, with an inter-item covariance of both dimensions was made.The results suggested that, the item 5 18 (I realize that I am tied to certain prejudices when I assess ethical issues) and the item 1 3 (I notice if someone working with me is offended by me) had the highest covariance with other items in the mentioned dimensions.If we remove them, the reliabilities will increase to: α = 0.49 and α =0.50 for dimensions 5 and 1, respectively.
In general, the reliability of the ESSQ is good.The low reliability scores in two dimensions might be due to cultural differences in the sample of our study (Iranian culture) with original sample (Finnish culture).Highly abstract concepts involving such as ethical and moral issues could be differently interpreted in various cultures.

The Descriptions of Ethical Sensitivity in the Sample.
In this part of presentation of our results, we will present the nature and level of ethical sensitivity in the target sample.The results showed that the level of ethical sensitivity in the sample was high (M = 3.85, SD = 0.36).
Considering the differences in the dimensions of the scale with overall mean score of the ethical sensitivity, the results showed that except for "Identifying the consequences of actions and options", there were statistically significant differences between other dimensions with overall mean of the scale (as test value).Table 3 shows the results of other dimensions.As it is shown, "Taking the perspective of others" received the highest score (M = 4.38, SD = 0.53) and showed a large mean difference with the overall mean of the scale (Mean.d = 0.51).On the other side, "Reading and expressing emotions" had the lowest mean score (M = 3.26, SD = 0.59) and resulted in the largest mean difference with the overall mean of the scale (Mean.d = −0.59).
Considering the three subgroups of the sample, the results suggested that there were no statistically significant differences among primary, middle, and high school teachers in the six dimensions of the scale.As shown in Table 4, in all dimensions, the perceived ethical sensitivity of primary and middle teachers was higher than that of high school teachers.In the dimension "Identifying the consequences of actions and options", the difference was statistically significant; the primary teachers (M = 3.93, SD = 0.45) perceived their ethical sensitivity more highly than did the middle (M = 3.91, SD = 0.51) and high school teachers (M = 3.79, SD =   0.43).The Hochberg's Post Hoc inspection, however, showed that the differences between high school with primary and middle schools were statistically significant, while the difference between middle and primary teachers was not statistically significant.Looking at gender, the results showed that there was a statistically significant differences between male and female in three dimensions of ethical sensitivity: reading and expressing emotions (M: female = 3.07, male = 3.39; SD: female = 0.57, male = 0.56), caring by connecting to others (M: female = 3.60, male = 3.7; SD: female = 0.63, male = 0.59), and taking the perspective of others (M: female = 4.33, male = 4.3; SD: female = 0.51, male = 0.54).The other four dimensions resulted in no statistically significant differences.The results for gender differences are presented in Table 5.

Conclusion and Discussion
The present research was conducted on the premise that there was a need for measuring ethical sensitivity in different cultures and professional contexts.In line with this need, the Ethical Sensitivity Scale Questionnaire (ESSQ) was administered to a sample of 556 Iranian Kurdish teachers in order to (1) examine its validity in a different culture and (2) describe the level of ethical sensitivity in the targeted population.The ESSQ was originally developed by Tirri and Nokelainen and is based on Narvaez's theory [12] of seven dimensions of ethical sensitivity [3,13].
The results of an initial Confirmatory Factor Analysis (CFA) showed that the ESSQ reflected a reasonable model fit of these seven dimensions of ethical sensitivity.However, in order to improve the model's goodness of fit, we needed to do some changes based on suggested modification indices in the CFA.The modification indices were mainly related to high error covariance between four pairs of items (7 27 and 7 28; 5 17 and 5 18; 3 9 and 3 10; 6 23 and 6 28).These modification indices can be explained by their face and content similarities.In other words, these pairs of items are somehow similar and thus examine the same content.In this way, our study confirms the results of previous studies [3] that the ESSQ supports Narvaez's theory [12] of seven dimensions of ethical sensitivity.Like these studies, our results suggested that a systematic item reduction was needed in order to remove some items from the model, particularly items with a high error covariance.Therefore, when we freed the error covariance of the mentioned pairs of items, the model's goodness of fit significantly increased according to various indices.In addition to validity, the reliability analysis suggested that the overall reliability of the scale was good.Regarding the seven dimensions, "Generating interpretations and options" and "taking the perspectives of others" resulted in the highest scores.On the other side, "Preventing social bias" and "Reading and expressing emotions" had lowest reliability scores.Further inspection suggested that removing two items (es5 18 and es1 3) with a high covariance resulted in better reliability scores for these two dimensions.As a result of validity and reliability analyses, we believe that the ESSQ supports a good model fit for studying the seven dimensions of ethical sensitivity, provided that the items with a high error covariance are removed from the model.
Further descriptive analysis with the ESSQ suggested that the level of perceived ethical sensitivity in the sample was high (M = 3.88; SD = 0.36).Such a result supports the idea of "teaching as a moral activity", which has been reflected in several important studies [34][35][36][37][38].In these studies, "caring" is considered to be a fundamental competency for teachers in their day-to-day activities.The main idea of the so-called caring teaching is to conduct ethical and pedagogical actions so as to enhance the welfare of the students.In this way, ethical sensitivity is an initial and very important step in conducting caring pedagogy.This is why the teachers selfreported ethical sensitivity was high in this study and the other study in Finish context [28].
More in-depth analysis showed that the highest and the lowest self-reported dimensions of ethical sensitivity among Iranian Kurdish teachers were "Taking the perspective of others" and "Reading and expressing emotions," respectively.These results can be explained by two important phenomena.First, Iranians and particularly Kurds are known to have a collective culture and mentality in which "taking care of the perspective of others" has a significant place.Second, to be concerned about "the other" is the main idea of ethical sensitivity.As such, on the one hand, "taking the perspective of others" may reflect a part of Kurdish collective culture in which commitment to people and others is highlighted and considered as an essential value.Our results supported the previous findings where collective values predicted higher "empathic concern perspective taking" among Iranians [39].According to Ghorbani et al., empathic concern and perspective taking presumably are central to maintaining an adaptive allocentric sensitivity to others [39].On the other hand, "reading (and particularly) expressing emotions" somehow reflects "taking care" of "oneself " and thus may have caused less attention from Kurdish teachers.Other research shows that "assertiveness, autonomy, freedom, self-fulfillment, and a sense of personal uniqueness" are associated with individualist values in which commitment to reason is emphasized [39].
The last part of the descriptive results showed that the lower grade teachers generally perceived a high level of ethical sensitivity.However, it was only the dimension of "Identifying the consequences of actions and options" that resulted in statistically significant differences among primary, middle, and high school teachers.In general, primary and middle teachers reported higher ethical sensitivity than high school teachers, when it comes to identifying the consequences of their actions.This is due to the fact that the younger students are prone to more negative consequences as the result of careless activities on the part of their teachers.This part of our results opens up the idea that ethical sensitivity may be associated with different demographic and mediating variables.In various contexts, ethical sensitivity was found to be associated with age and the professional positions of the respondents [40], gender [41], culture [42] and giftedness [3].As such, a further line of research will be to model the path structure of ethical sensitivity in different contexts, including teaching and teacher education.
Considering the main task of this paper, we suggest that the ESSQ is a promising tool for measuring the seven components of ethical sensitivity.However, a systematic item reduction and modification will be needed to increase the model fit of the scale.In this study, some modifications based on covariations between four pairs of items were suggested.It would be worth to reexamining the stability of the scale in line with this modification in other studies.In addition, we should point out that moral competency including ethical sensitivity deals with a lot of complex meaning and dimensions in different cultures.Thus, quantitative study and the instrument we used in this study may not fully cope with this complexity.In order to deal with this limitation, we suggest further qualitative study to gain more insights into teachers' moral competencies including ethical sensitivity in an Islamic context.

-Figure 1 :
Figure 1: Dimensions of ethical sensitivity and link to teaching.

Table 1 :
Sample description.Data was collected from teachers who taught in three levels of K-12 Education in Kurdistan.Kurdistan is one of the 31 provinces in Iran and has about one and half million inhabitants, the majority of which has a Kurdish ethnic background.It has 8 major cities each with a population of over 100,000 inhabitants.The center of the province is Sanandaj with about 400,000 inhabitants.The total number of students and teachers in Kurdistan was 58,979 and 3,274 respectively in 2011.Using nonprobability sampling, data was collected from 556 teachers in the province of Kurdistan.The teachers were asked to complete a paper-and-pencil version of the Ethical Sensitivity Scale Questionnaire (ESSQ).Questionnaires were distributed among schools and delivered to the teachers by school principals.Table1illustrates the descriptive statistics of the sample.
(1)To what extent can the original constructs of the (ESSQ) be identified among Iranian teachers?(2)What is the nature and level of ethical sensitivity of Iranian Kurdish teachers?4.Methods4.1.Sampling Frame.

Table 2 :
Factor structure and alpha loadings of the ethical sensitivity scale.

Table 3 :
Mean differences of dimensions of ethical sensitivity with overall mean score of ethical sensitivity among Iranian Kurdish teachers (N = 556).

Table 4 :
Mean differences of dimensions of ethical sensitivity based on level of school (N = 556).