Psychometric properties of Nurses Professional Values Scale-Revised: An Iranian version

Professional values are the basis of development of professional identity and commitment. Evaluation of professional values provides critical information regarding effective strategies for their integration in professional performance and clinical learning. The present study aimed to determine the psychometric properties of Nurses Professional Values Scale-Revised (NPVS-R) in Iranian population. This methodological study was conducted on 439 students in the last semester of nursing selected from Iran’s universities using stratied single-stage cluster sampling. After back translation, its face, content, and construct validity were assessed. Its reliability was also evaluated by determining Cronbach’s alpha using internal consistency method. All analyses were performed using the SPSS statistical software, version 19.

1998) [5], which is used to direct accreditation of curricula of B.Sc. and higher education levels [5]. The professional values planned in AACN include human dignity, integrity, autonomy, altruism, and social justice [6]. American Nurses Association (ANA) has also presented directions for professional values in "Code of Ethics for Nurses with Interpretive Statements" [7]. In order to introduce issues related to spiritual and ethical health in the 21st century, nurses and their colleagues have to refer to professional values to direct their professional performance, behavior, and decision-making [7]. Although nursing curricula may follow reliable guides for bene tting from professional values, a limited number of them might be aware of the rate of success in acquiring learning outcomes related to professional values [8,9].
Understanding how to develop professional values is of particular importance in future of nursing profession [10,11]. In this context, evaluation of professional values can provide valuable information about effective strategies for integration and application of professional values in professional performance and clinical learning [12].
In order for an instrument to be usable in a community, it should bene t from appropriate reliability and validity [13]. Validity refers to accuracy and credibility of a study [14]. Additionally, reliability means obtaining similar results in repeated measurements using a single instrument [14]. Therefore, reliability and validity of NPVS-R are the requirements of its applicability in Iranian community.
Limited studies have been conducted on nurses' professional values in Iran [15][16][17] and Asia [18][19][20]. On the other hand, using valid and reliable Persian version of NPVS-R can enhance the importance of professional values and ethical codes for nurses in Iranian society [21]. Considering the importance of institutionalizing and applying professional values, the present study aims to assess the psychometric properties of NPVS-R in Iranian population.

Background And Conceptual Framework
Considering investigation of professional values, various instruments were used in researches between 1970 and 1977, which resulted in creation of different databases that were not generalizable. Although there are several instruments for evaluation of professional values, only one; i.e., Nurses Professional Values Scale (NPVS), is based on ANA's code of ethics for nurses. Modi cation of nurses' code of ethics in 2001 accelerated the process of reviewing NPVS [22], which resulted in creation of Nurses Professional Values Scale-Revised (NPVS-R). Due to general application of ANA's codes and their similarity to laws issued by the International Council of Nurses and other codes, such as those developed in Iran [21], this instrument can be used in different communities including Iran.
In the present study, permission was obtained from the developers of the instrument. The scale was rst translated to Persian by a bilingual translator who was familiar with nursing profession. Then, three experts in nursing evaluated the scale regarding appropriateness with Iranian culture. After slight modi cations, the scale was back translated to English. The back translated version was quite similar to the original version, indicating no need for modi cations in the Persian version.
Validity refers to the extent an instrument measures what it is supposed to measure [24,25]. In the current study, face validity, content validity, and construct validity were taken into account. Face validity refers to evaluation of an instrument's outward credibility, being logical, appropriateness, attractiveness, rationality, logical sequencing, and comprehensibility from target population's perspective [26]. In addition, content validity refers to the extent an instrument contains appropriate items to the intended construct [26]. Finally, construct validity refers to the extent an instrument measures what it is supposed to measure [26].
In order to assess the face and content validity of NPVS-R, it was given respectively to 10 nursing students and 12 experts in nursing and instrumentation in Shiraz and Tabriz universities of medical sciences. Considering qualitative face validity, the items were evaluated with respect to level of di culty, vagueness, failure in meaning, appropriateness (relationship between the items and the main goal of the scale), simplicity, uency, and understandability. In quantitative face validity, item impact method was employed using a 5-point Likert scale ranging from not important (1) to very important (5) to determine the impact score. The items with impact scores > 1.5 were maintained for further analyses.
In order to assess the content validity of NPVS-R, Content Validity Rate (CVR) and Content Validity Index (CVI) were computed. In doing so, assessors were required to evaluate the scale items based on Lawshe table. Lawshe invented a method to determine content validity. In this method, a questionnaire is given to a panel to mention their opinions about each item within a predetermined judgement scale. Then, the panel's responses are coded as follows: essential, useful but not essential, and not necessary [27,28]. After obtaining the experts' opinions in the present study, items with CVR > 0.6 were maintained and the rest were omitted. In this formula, Ne represents the number of experts who consider the item essential and N represents the total number of experts.
After that, the experts were required to determine the items' relevance, clarity, and simplicity using a score between 1 and 4 based on Waltz and Bausell's index. Accordingly, items with CVI > 0.79 were appropriate, those with 0.70 ≤ CVI ≤ 0.79 were questionable requiring modi cation, and those with CVI < 0.70 were unacceptable.
Finally, construct validity was evaluated using exploratory factor analysis with varimax rotation. In order to extract factors, Eigenvalues > 1 and scree plots were employed. It should be noted that factor load value was considered to be above 0.4. Besides, nearly 17 samples per item completed the questionnaires. In this way, after getting permission from the Ethics Committee of Tabriz University of Medical Sciences (No: IR.TBZMED.REC.1390.112), 439 senior nursing students were selected from Iran's universities using strati ed single-stage cluster sampling. Among the participants, 286 (65.4%) were female and 151 (34.6%) were male. In addition, their mean age was 22.56 ± 1.08 years.
Reliability means obtaining similar results in repeated measurements using a single instrument [26]. Reliability coe cients > 0.7 are acceptable, but those between 0.85 and 0.95 are preferred [29]. In the present study, the reliability of the instrument was assessed using internal consistency by determining Cronbach's alpha coe cient. In doing so, a descriptive study was conducted by the researcher. Considering the necessity to use Cronbach's alpha coe cient for investigating reliability in questionnaires with Likert scales [30], internal consistency method by determining Cronbach's alpha coe cient was used in the present study.

Results
For investigation of face validity in this study, quantitative and qualitative approaches were taken into account. Considering qualitative face validity, the questionnaire was given to 10 nursing students and their opinions were applied in form of slight changes. In quantitative face validity, item impact method was used to determine impact scores. Based on the results, all components' impact scores were above 1.5 (Table 1). Therefore, all questionnaire items were entered into the next evaluation stage, which was related to content validity. Considering investigation of qualitative content validity, the questionnaire was given to experts and they were required to provide their written opinions about content coverage, grammar, usage of appropriate phrases, proper scoring, time required for completion, appropriateness of domains, and location of items.
The obtained results were in line with those of qualitative face validity. Accordingly, the questionnaire components were quite simple and clear and, consequently, no modi cations were performed.
CVR and necessity of items were assessed using Lawshe table. Accordingly, the items with CVR > 0.6 were maintained. Based on the results, CVR of all questionnaire items was between 0.6 and 1 (Table 2). Thus, all components were maintained at this stage.
Content validity was evaluated using CVI proposed by Waltz and Bausell [28] and the results have been presented in Table 2. As the table depicts, CVI was above 0.79 for all items and above 0.9 for the whole scale. Thus, all items were accepted at this stage. In addition, all items showed appropriate simplicity and clarity and, consequently, no modi cations were needed. The results of factor analysis and factor loads have been presented in Table 3. Accordingly, factor load of "not participating in cares that are ethically opposed to my professional values" was 0.29 and, as a result, the item was omitted from the questionnaire (Table 3). Investigation of the reliability of the questionnaire revealed Cronbach's alpha coe cient of 0.91 for the whole instrument. Indeed, Cronbach's alpha coe cients of 0.83, 0.81, 0.72, 0.75, and 0.53 were computed for care, professionalism, activism, responsibility, and social commitment factors, respectively.

Discussion
The study results indicated that NPVS-R bene tted from appropriate validity (face and content validity) and reliability, which is consistent with the results of other studies conducted on the issue [21,31,32].
Studies have also proved the validity of English, Spanish, Chinese, Turkish, and Korean versions of NPVS-R. Similarly, Weis and Schank (2009) assessed qualitative face validity of NPVS-R among B.Sc. and M.Sc. nursing students and clinical nurses [21]. Considering quantitative face validity, all components showed impact scores > 1.5 in the current study and, consequently, all items were entered into the next stage that was related to content validity.
With respect to qualitative content validity in the research by Weis and Schank (2009), NPVS-R was investigated regarding su ciency, relatedness to ethical codes, clarity, and meaning. The results revealed 100% agreement and all items were maintained with slight changes in words. That study also con rmed the content validity of the original version of this instrument [21].
In the present study, CVR of all components of the scale ranged from 0.6 to 1. Therefore, all components of the scale were maintained at this stage. Review of the literature showed that no studies have assessed the CVR of NPVS-R.
Our study also demonstrated that CVIs of the items and the whole scale were equal to 1. Thus, all items were accepted. In the same line, Lin and Wang evaluated content validity using CVI. In that study, seven experts including two nurses, two head nurses, one supervisor, and two assistant professors scored the Chinese version of NPVS-R with regard to relevance, clarity, and simplicity. Based on the results, CVI ranged from 0.8 to 1 and was equal to 0.9 for the nal version [33]. With regard to construct validity in the current study, KMO was 0.93 and Bartlett's sphericity test was statistically signi cant. In addition, exploratory factor analysis determined ve main domains of NPVS-R by varimax rotation and exclusion of items with factor loads < 0.4. These factors included care (6 items), professionalism (7 items), activism (5 items), responsibility (4 items), and social commitment (3 items), which explained 52.26% of the total variance. It should be noted that item 17 (not participating in cares that are ethically opposed to my professional values) was excluded because its factor load was below the threshold value.
In line with the present study, Weis and Schank (2009), Geckil et al. (2012), and Moon et al. (2014) reported KMO criteria to be 0.93, 0.92, and 0.95, respectively. Bartlett's sphericity test was also statistically signi cant in the above-mentioned studies (p < 0.001), which revealed appropriateness of factor analysis for construct validity of the data [21,31,32]. In the research by Weis and Schank (2009) The Persian version of NPVS-R can be used for assessment and supervision prior to interventions, evaluation and development of nursing professional values after interventions, and investigation of nursing professional values over time. Moreover, using NPVS-R can enhance nurses' awareness about the importance of professional values and codes of ethics.

Conclusions
The

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and/or analysed during the current study are not publicly available due to patient con dentiality but are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was supported by Tabriz University of Medical Sciences.