*2.6. Statistical Analysis*

The Kolmogorov–Smirnov test was utilized to confirm the normal distribution of the continuous data, with the result being <0.05, indicating that the data did not follow a normal distribution.

To analyze the reliability of the scale, a test–retest method was applied, with the calculation of the intraclass correlation coefficient (ICC) to evaluate the degree of consistency between the quantitative measurements obtained in the questionnaire. To examine the internal consistency, Cronbach's α was utilized, with a minimum value of 0.700 desired.

An exploratory factorial analysis (EFA) was performed. Before this, the Kaiser–Meyer– Olkin (KMO) and Bartlett's sphericity tests were performed to consider the adjustment of the values for the EFA. So that the factorial loads were consistent, the value had to be ≥0.40 for an item to be part of the factor selected [30].

Spearman's correlation coefficient, Welch's *t*-test and Welch's ANOVA were utilized. Values of *p* < 0.05 were considered significant. For the statistical analysis, the SPSS v21 software for Windows was utilized (SPSS, Inc., Chicago, IL, USA).

#### *2.7. Ethical Considerations*

Permission to use the English version of the 27-item standard research version was obtained via e-mail from Dr. Arthur Aron. The study was approved by the Ethics Board from the UCAM in June, 2019 (code CE 061902), considering the guidelines from the 1964 Declaration of Helsinki.

#### **3. Results**

Of the 284 students, 75% were women; 28.9% were enrolled in their first year, 25.4% in their second, 25% in their third, and lastly, 20.8% in their fourth year. The mean age was 21.6 ± 4.4 years. As for their training, 25.4% had some type of healthcare training. With respect to their family relationships, 8.8% described them as unsatisfactory. Additionally, 51.4% did not have a partner, and of those who did, 3.9% qualified their relationship as unsatisfactory. Lastly, 3.2% qualified their social relationships as very unsatisfactory.

#### *Initial Validation of the Reduced Versions of the Highly Sensitive Person Scale (R-HSP Scale)*

To verify the reliability of the scale, the consistency of the items was analyzed after repeating their measurement, through the application of the intraclass correlation coefficient. Table 1 shows the ICC value of all the items in the scale, with all of them being statistically significant, with a *p* < 0.005 value.

The reliability was also verified with the correlation analysis of the different measurements obtained after applying the scale multiple times, a procedure known as the split-halves method. Thus, in Table 1 we can verify a Spearman–Brown coefficient of r = 0.886, which indicates the high reliability of the questionnaire. Additionally, the Cronbach's α value was 0.705 on the initial test, and 0.760 on the retest, both of which were above 0.700, which verifies the reliability of the questionnaire (Table 1).

Table 2 shows the results from the correlation analysis of all the items on the questionnaire. Table 3 shows the results obtained in the homogeneity analysis of the items in the questionnaire. The Cronbach's α value obtained was 0.702. No items were eliminated, as the Cronbach's α value barely increased (Table 3).

As previously mentioned, the scale is composed of 16 items, with range in values between 0 and 16, with a higher score indicating a higher sensitivity. The mean was 9 ± 3.1 points, with 34.5% of the participants being HSP.

To analyze the validity of the construct, a factorial analysis was performed (Table 4). The Kaiser–Meyer–Olkin test provided a value of 0.729, with the Bartlett sphericity test being statistically significant, *p* < 0.001. The factorial analysis showed a structure composed of six factors, which as a set, explained up to 54.9% of the total variance of the results. Factor 1 with a value of 19.1%, factor 2 with 8.1%, factor 3 with 7.7%, factor 4 with 7%, factor 5 with 6.7%, and factor 6 with a value of 6.3%. Factor 1 included items 11, 12, 13, 14, and 16, which were considered related with "Instability". Factor 2 consisted of items 3, 9, 12, and 15, related with "Surroundings". Factor 3 was composed of items 1, 8, 10, and 13, "Interaction with others". Factor 4 was composed of items 5 and 6, "Sensoperception". Factor 5 included items 2 and 4, "Sensitivity". Lastly, factor 6 was composed of items 7 and 14, "Insecurity" (Table 4). Figure 1 provides a scree plot as a graphical representation of the extracted factors.

**Figure 1.** Scree Plot.


F: Statistic's value. *p*: Significance level. ICC: Intraclass correlation coefficient. CI: Confidence interval.

With respect to the associations between the total HSP scale and the personal and academic factors of the university students, differences were only found in women (9.61 ± 2.99 vs. 7.1 ± 2.71; *p* < 0.001), and those who indicated having unsatisfactory family relations (10.16 ± 2.92 vs. 8.89 ± 3.10; *p* = 0.049) (Table 5).




Spearman's correlation coefficient. *p*: Significance level.

rho: **Table 3.** Homogeneity analysis of the questionnaire.


**Table 4.** Factor loading of the questionnaire items. Rotated components matrix.


**Table 4.** *Cont*.


**Table 5.** Association between each factor of the R-HSP scale and personal attribute of Spanish university students.



**Table 5.** *Cont*.

rho: Spearman's correlation coefficient. *p*: statistical significance.

#### **4. Discussion**

To measure environmental sensitivity, the most utilized scale with university students or adults is the High Sensitivity Persons scale (HSP scale) developed by Aron and Aron [1]. However, for field studies in which time is highly prized, this original version of 27 items is inconvenient due to its length, and a need was detected to validate a Reduced High Sensitivity Persons scale (r-HSP) for nursing university students. This scale included the items that were habitually used in Spain by the Association of Persons with High Sensitivity (Aspase) for the diagnosis of Environmental Sensitivity, but it is necessary to show that the r-HSP is a simple tool that can be used to identify students who are highly sensitive.

The results showed that the tool had good psychometric characteristics. More specifically, the test and the retest showed a good reliability, with Cronbach's α values >0.700. Many studies related with the HSP scale showed one to three factors [1,28,31]. In the present study, six factors were identified. This structural model of the reduced HSP scale suggests that the general sensitivity score, as well as the scores of the six factors, are adequate for measuring the environmental sensitivity of Nursing students.

Aron and Aron [1] estimated that a high sensitivity was present in 20% of the general population. However, in the present study, 34.5% were identified, a value that is much higher than the one mentioned previously. In our study, we found significant differences, with women being much more sensitive than men. This finding is similar to the results from other authors. However, in these studies, the differences between gender groups were not statistically significant [4,32,33].

On the other hand, age was correlated with three of the six factors. The older the student, the fewer interactions with the rest, and more sensoperception and sensitivity. Costa-López et al. also found positive correlations between age and environmental sensitivity [27].

It has been described that individuals with a high level of environmental sensitivity can show over-stimulation, sensorial sensitivity, deep cognitive processing, and emotional reactivity [34]. In their social and personal relations, they are characterized as being empathetic and intuitive [3]. This means that these individuals relate better with others. These characteristics were not observed in our study, as a higher score in the scale was observed in those with unsatisfactory family relations.

Our study provides new evidence on the association between HSP and important aspects of the students, which could be considered as current life stressors, such as their relationships with their families and/or partner. The results show that those who had unsatisfactory family relations were HSP with higher scores on the scale. These results are similar to other studies, which verified that family problems of students increased their level of stress [21,22]. In other studies on HSP, research was not performed on current personal life aspects, and which directly influence their well-being. In general terms, the differential susceptibility of the adult subjects was not analyzed, including items in the questionnaires that were focused on the analysis of their childhood.

In summary, although the study used a small sample, the test–retest reliability showed ICC values (Table 1) ranging from poor (item 14, 15) to moderate and good. Furthermore, although the correlations between items were generally low, the reliability of the scale (Cronbach's alpha in Table 3) was acceptable.

## *Limitations*

Just as the original scale, most of the items put emphasis on the negative traits ("I become overwhelmed when I have a lot of things to do and little time"). In future studies, it would be interesting to focus on the advantageous aspects of being an HSP, and to conduct a more in-depth analysis of other aspects such as processing ability, empathy, the emotional response ability, and the sensitivity to subtle aspects. Additionally, it would be positive to perform a multi-center study and broaden the sample to other university faculties. Due to the preliminary nature of this research, future studies are needed to confirm the results with a confirmatory factor analysis (CFA).

#### **5. Conclusions**

The adaptation of the reduced versions of the Highly Sensitive Person (r-HSP) scale is defined as a reliable, valid, and agile replica of the original structure of sensitivity in Spanish university students. The present initial validation of the reduced HSP scale is adequate for its application to university students, as it can distinguish between the HSP students and those who are not. The prevalence was found to be greater than the general population.

**Author Contributions:** Conceptualization, A.P.-V., D.J.-R. and P.E.P.; methodology, A.P.-V., D.J.-R. and P.E.P.; formal analysis, A.P.-V., A.J.S.-M. and G.C.-L.; investigation, A.P.-V., D.J.-R., P.E.P. and J.I.G.-G.; data curation, A.P.-V., A.J.S.-M. and G.C.-L.; writing—original draft preparation, A.P.-V., A.J.S.-M., J.I.G.-G., D.J.-R., P.E.P. and G.C.-L.; writing—review and editing, A.P.-V., J.I.G.-G., D.J.-R., A.J.S.-M. and P.E.P.; supervision, D.J.-R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Bioethics Committees of the Catholic University of Murcia (Approval no. CE 061902).

**Informed Consent Statement:** Informed consent was obtained from all participants involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

