Validity and reliability of the Korean version of the Humanism Scale Short Form: A cross‐sectional study

To examine the reliability and validity of a Korean version of the Humanism Scale Short Form.


| BACKG ROU N D
Humanism is a philosophical stance that emphasizes humanity. The word originates from the Latin term, humanitas, which means "making people more human" (Veyne, 1993, p. 342). Throughout history, humanist thoughts have found representation both in Eastern (e.g. the Donghak concept of Innaecheon) and Western cultures (e.g. Renaissance period). However, since ancient times, humanism has undergone several changes and has been influenced by other ideologies characteristic of each era. Thus, only the fundamental framework of respect for humanity has been maintained (Doosan Encyclopedia Editorial Staff, 2010).
In the field of human health, a humanistic, person-centred approach was first proposed in 1940 by American psychologist Carl Rogers (Morgan & Yoder, 2012). This approach was based on the idea that the relationship between counsellor and client should be one of equality. In a similar context, Abdellah (1960) proposed human-centred nursing because of the need for an individualized approach that incorporated human-centred values into the profession.
This new philosophy differed from the disease-oriented nursing that prevailed until then; there occurred a shift in focus towards patients' values and needs (Hong, 2018). In order to practice human-centred nursing, above all, nurses must approach the discipline with humanistic ideals.
The role of humanism in nursing has been reported in many studies on hospice or palliative care, in connection with the extended lifespans seen in recent years. Hopkinson (1999) applied humanistic nursing theory in a study on patients' perceptions of hospice day care. Coward (1991) studied humanism in nurses who cared for patients with breast cancer and found that nurses' humanism affected patients' psychological stability. Although humanism has been criticized for its limited role in promoting science and has been considered objectively indiscernible (Traynor, 2009), it has also been claimed to be a necessary quality among nurses with its emphasis on fostering ideals and practices for improved patient care (Wu & Volker, 2012). Nevertheless, there is no objective tool to measure this competency in nurses in Korea.

| Design
This was a cross-sectional descriptive study designed to translate the Humanism Scale Short Form, as revised in 2016, into Korean and test its validity and reliability. Study reporting followed STROBE guidelines (File S1).

| Participants
This study was conducted in a general hospital in D city, Korea.
The specific participant selection criteria for nurses were as follows: (a) having worked at the general hospital for at least 3 months, (b) patient care as the primary job function, (c) understanding the study's purpose and (d) voluntarily consenting to participate. The required sample size was estimated to be 150 participants, given that a sample size of 5-10 times the final number of items is considered adequate for exploratory factor analysis (EFA; Tab achnick & Fidell, 2001). However, considering possible issues with the survey return rate and response completeness, an online self-report questionnaire was distributed among 200 nurses. Seven questionnaires were not returned, and two had incomplete responses; thus, a total of 191 questionnaires were available for analysis.

Humanism scale short form
To assess levels of humanism, the Humanism Scale Short Form was used with permission from the developer. The scale was translated into Korean through a process of translation and back-translation, and revised based on the results of a test for content validity. The scale consists of 15 items on a seven-point Likert scale, where 1 = strongly disagree and 7 = strongly agree. A total score is obtained by summing the items. The higher the total score, the higher the level of humanism. At the time of scale development, the Cronbach's α, which is the index for reliability, was 0.93 (Nilsson, 2015).

Empathy quotient scale
The Korean version of the Empathy Quotient Scale was used to assess empathy. This scale was originally developed by Baron-Cohen and Wheelwright (2004), tested for reliability and validity by Lawrence et al., (2004), and translated into Korean by Heo and Lee (2010). This 17-item scale consists of 10 cognitive empathy items assessing cognitive aspects, three emotional reactivity items assessing emotional aspects, and four social skills items assessing behavioural aspects. All items are rated on a five-point Likert scale, with a higher total score indicating a higher empathy quotient. In this study, the average score, obtained by dividing the total score by the number of items, was analysed. The Cronbach's α was 0.83 in Heo and Lee's (2010) study.

Compassion competence scale
Compassion among nurses was assessed using the Compassion Competence Scale developed by Lee and Seomun (2016). This scale consists of a total of 17 items that are rated on a five-point Likert scale. Total scores range from 17-85 points, and the higher the score, the higher the level of compassion. The Cronbach's α value was 0.93 in Lee and Seomun's (2016) study. Back-translation. A nursing doctoral student, unaware of the contents of the original instrument, whose first language is English and who is also proficient in Korean, translated the Korean items back into English. Then, a nursing professional proficient in both languages compared the back-translated items against the original instrument to check for any differences. Subsequently, three other nursing professionals examined the Korean wording to test the content validity and finalize items.

Pilot test
Before the survey was administered, a pilot test of the Korean version of Humanism Scale Short Form was conducted with 10 university students who were not in the main study sample to ensure readability, comprehensibility and adequacy of the wording. During the process, the survey duration and the reactions of the students while completing the survey were observed. The students were encouraged to give their opinions freely if they found any items unclear or did not understand any aspect of the wording.

Test for construct validity
Construct validity was tested using item analysis and EFA. In the item analysis, item means, standard deviation (SD), skewness, and kurtosis indices were examined to check the level of bias in the distribution of each item. Additionally, item-total correlation coefficients were computed to examine whether each item reflected the concept of humanism, and a criterion was established to discard items with a low contribution, that is, a correlation coefficient less than 0.30 (Streiner & Norman, 2011). EFA was performed using principal component analysis, which allows the determination of a stable pattern of consistency coefficients regardless of sample size. A varimax method was used for factor rotation, and the criterion for explained variance was set at 50% or more. Factors with an initial eigenvalue of 1.0 or more were extracted, and items with a coefficient of 0.40 or less were selected (Song, 2011).

Reliability analysis
The reliability of the Korean version of the Humanism Scale Short Form was assessed by computing Cronbach's α.

Test for concurrent validity
Concurrent validity was evaluated by simultaneously administering the Korean version of the Empathy Quotient Scale and the Compassion Competence Scale as these instruments have been reported to be highly correlated with humanism in previous studies (Burks & Kobus, 2012;Gaufberg & Hodges, 2016). Specifically, Pearson's correlation analysis was performed on the total scores of the three instruments to test the extent to which they were correlated.

| Ethics
The study was approved by the institutional review board of the concerned university (JIRB-2019040801-01-190429) and was conducted in accordance with the principles outlined in the Declaration of Helsinki. The selected individuals were first informed of the rationale and purpose of the study, and then an overview of the survey was presented. The survey was administered only to those who provided written informed consent. Participants were assured of anonymity, that the survey data would not be used for purposes other than the research, and that the data would be discarded on completion of the study. The participants were given a small gift for filling out the survey.

| Item analysis
Item analysis and EFA were conducted to evaluate construct validity. The corrected item-total correlation coefficients of the 15 items computed for the item analysis were in the range of 0.42 to 0.64. No item had a coefficient below the criterion (0.30); therefore, all 15 items underwent EFA. The lowest item mean was 4.51, and the highest was 6.26, while the lowest and highest values for item SD were 0.84 and 1.52, respectively. The mean total score was 5.64, and the SD was 0.67 (Table 2).

| EFA
The Kaiser-Meyer-Olkin statistic was 0.88, and the test statistic of Bartlett's test for sphericity was χ 2 = 1,128.18 (p < .001), confirming the goodness of fit of the EFA solution (Lee & Rho, 2015).  items were representative of the equal rights of all human beings in society. Thus, the first factor was named "human equality." The next six items, in factor 2, consisted of questions that focused on human beings' dignity, respect for individual identity, and respect for others. Therefore, the second factor was named "respect for human beings." The presence of the two factors in the instrument was confirmed (Table 2), and it was named the "Korean version of the Humanism Scale Short Form."

| Reliability analysis
The test for the internal consistency reliability of the Korean version of the Humanism Scale Short Form showed that the Cronbach's α was 0.88 for the entire scale, and 0.85 and 0.80 for factors 1 and 2, respectively ( Table 2).

| Concurrent validity
The Korean version of the Humanism Scale Short Form was positively correlated with the empathy quotient of the Korean version of the Empathy Quotient Scale (r = .42, p < .001) and compassion according to the Compassion Competence Scale (r = .58, p < .001).
As the correlation with the reference instruments was between 0.4-0.8, our tool has sufficient concurrent validity as per Lee et al., (2009, p. 572-578).

| D ISCUSS I ON
In the present study, we examined the reliability and validity of a Korean version of the Humanism Scale Short Form, based on the scale developed by Nilsson (2015), further refined in 2016, to objectively assess levels of humanism in nurses in Korea. The test for the reliability of the Korean version showed that the Cronbach's α was 0.88, which was higher than the criterion (0.70) proposed by Nunnally and Bernstein (1994). This demonstrates that the instrument was highly reliable. The Cronbach's α of the Korean version was found to be similar to the coefficients identified while developing the original instrument-0.83 in a study conducted with US university students, and 0.76 and 0.83 in two studies conducted with Swedish adults (Nilsson, 2015). This finding indicates that the  the EFA. Conducting a between-study comparison was not possible as factor analysis results regarding construct validity had not been determined when the Humanism Scale Short Form was developed (Nilsson, 2015); however, in this study conducted with nurses, the

| Limitations
Despite these advantages of the instrument, this study had some limitations. First, although the sample size was over 10 times the number of items, the participants were nurses from a single region. Hence, caution is required while generalizing the results, and additional research should be conducted to replicate the findings with a greater number of participants recruited from multiple regions. Second, the instrument was not tested for stability using the test-retest method. Future research is needed to evaluate the stability of the instrument. Third, in this study, we did not conduct a confirmatory factor analysis (CFA). Therefore, CFA is required to further evaluate the tool in further studies.

| CON CLUS IONS
This Korean version of the Humanism Scale Short Form is of considerable significance in that objective data assessing the level of humanism in nurses can be generated. For the clinical application of patient-centred care, it is essential to study the relationship between nursing and humanism. Accordingly, the Korean version of the Humanism Scale Short Form is expected to be used in future studies to measure the level of humanism, that is, the extent and level of humanistic attitudes, among nurses.
The Korean version of the Humanism Scale Short Form developed in this study is likely to help in assessing humanism in Korean nurses more effectively. Humanism is fundamental for humancentred care and empathetic understanding in nursing practice. The use of this scale enables a nurse's approach to holistic and integrated care to be more clearly determined in terms of humanism, and can provide fundamental data for the development of programs to improve integrated care.

ACK N OWLED G M ENTS
We would like to thank Editage (www.edita ge.co.kr) for English language editing.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
No data are available in online. All supporting data can be provided upon request to the authors.