English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Literature shows that empathy has an innate or genetic component (García, González, & Maestú, 2011), an environmental component (Díaz-Narváez et al., 2014; García et al., 2011) or one of social and cultural environment (Delgado-Bolton, San Martín, Alcorta-Garza, & Vivanco, 2016). Therefore, empathy is a result of the interaction between these components. Empathy has a multidimensional internal structure formed by three components: perspective adoption, compassionate care, and ability to put oneself in the shoes of the other (Hojat, Gonnella, Nasca, Mangione, Vergare et al., 2002; Paro, Daud-Gallotti, Tibério, Pinto, & Martins, 2012). These attributes are involved in the physician-patient relationship, being associated with both improvements in patient care and good clinical outcomes (Batt-Rawden, Chisolm, Anton, & Flilckinger, 2013; Mercer & Reynolds, 2002), being an essential attribute for the practice of medicine (Vecchi, 2016). There is controversy regarding the measurement of results from the distribution of empathy between genders (Díaz-Narváez et al., 2015; Merino-Soto, López-Fernández, & Grimaldo-Muchotrigo, 2019), between educational levels (Silva, Arboleda, & Díaz, 2014) and between faculties of medicine, health sciences, and countries (Díaz-Narváez et al., 2014), existing several factors that affect it.
One of these factors is personality, which, beyond its multiple definitions, can be understood as a distinctive style of adaptive functioning that a member of a species tends to relate to it's typical spectrum of environments (Millon, 1994); being the personality styles of the individual strategies that the human being uses to adapt to the environment, which can be grouped into three major areas (Dresch, Sánchez, & García, 2005; Millon, 2003): motivational goals, cognitive modes, and interpersonal behaviors.
Motivational goals have to do with the desires and goals that guide people towards the end and lead them to act in a certain way. Cognitive modes allude to the way of transforming information and the sources used to acquire knowledge about life, importing what is taken into account when experimenting and learning, or what people do to make this knowledge meaningful and useful; it is the capacity for reflection to transcend the concrete and immediate, and symbolically represent the events and processes that are translated into potential for change and adaptation (Millon & Davis, 1996). Interpersonal behaviors refer to the preferred ways that people use to interact with others, a style of social behavior that derives in part from the interaction between motivational goals and cognitive modes (Millon, 2003).
Empathy is a multidimensional construct that includes all the processes of putting oneself in the place of the other and the associated affective and non-affective responses (Davis, 1983). In the context of health care, it involves cognitive aspects that encompass an understanding of the experiences, concerns, and perspectives of the patient being treated, as well as the ability to communicate such understanding to the person (Hojat et al., 2002), thus implying personality styles, such as information processing in a logical and analytical way (thought), the judgments conceived from their own affective reactions and personal values (feeling), the lack of emotion or social indifference (withdrawal), or the motivation to satisfy the needs of the other (protection), among other features and personality styles that necessarily participate in the expression of empathy (Dresch et al., 2005).
Current literature has established a relationship between empathy and personality in the field of health professionals, revealing a positive association between empathy and collaborative interprofessional work in physicians who are beginning their specialization (San-Martín et al., 2017). Other studies relate the model of the five big personality factors (Big Five) and empathy, finding a medium size effect when associating both variables, achieving predictive affective and cognitive empathy from the dimension of agility and consciousness (Melchers et al., 2016). In turn, personality differs according to gender and interacts with empathy and other socio-affective variables; in a study on personality in Chilean university students it was found that women have higher academic self-realization than men, but in turn a lower self-esteem and resolutive self-efficacy (Castellanos, Guerra, & Bueno, 2014).
In this context, we seek to analyze the relationship between empathy and its dimensions with the different personality styles presented by medical students, taking into account the differences by gender and educational level, improving understanding of the phenomenon of empathy, and providing backgrounds that contribute to the comprehensive training of doctors and other health science professionals.
Method
In order to establish the association between empathy, it's dimensions, and the different personality styles; a non-experimental, descriptive, cross-sectional study was designed.
Participants
The study included 278 students from the medical school of the University of Azuay, (Cuenca, Ecuador), corresponding to 98% of those enrolled during the academic year of 2016. 166 were women (59.7%) and 112 were men (40.3%), with an age range of 18 to 41 years (M = 20.88, SD = 2.78), distributed in the six levels of the career: first year (n = 72), second year (n = 92), third year (n = 24), fourth year (n = 45), fifth year (n = 17), and sixth year (n = 28). Only 2% of the students in the school did not participate, either because they were absent or did not give their informed consent to participate.
Instruments
The Jefferson Scale of Physician Empathy (JSPE-S) was applied, Spanish version for medical students, which measures the attitudes of medical students towards medical empathy in patient care situations (Hojat et al., 2001). The instrument consists of 20 items valued through a Likert-type scale, of seven points, from 1 (strongly disagree) to 7 (strongly agree). The scale gives scores ranging from 20 to 140 points and does not establish a cut-off point to constitute categories of high or low empathy. The instrument was previously subjected to cultural validation by judges, four experts whom examined the questions in order to adjust the concepts to the cultural environment and avoid confusion or subjective influences that distort the students' answers. A pilot study was performed prior to the final application that was carried out in March 2015.
The Millon Index of Personality Styles (MIPS) was also used (Millon, 1994), Spanish version, consisting of 180 items of dichotomous response (true/false) that allows the measurement of 12 pairs of poles that provide 24 features that determine the styles of personality, which in turn are grouped into three areas: motivational goals (openness, preservation, modification, accommodation, individualism, and protection), cognitive modes (extroversion, introversion, sensation, intuition, reflection, affectivity, systematization, and innovation), and interpersonal behavior (withdrawal, communication, hesitation, firmness, discrepancy, conformism, submission, control, dissatisfaction, and agreement). The scales measure personality in adults, older than 18 years, who function normally, thus moving away from a pathological perspective (Millon, 1994).
Procedure
This study was ethical approved by the Azuay University Faculty of Medical Sciences Research Ethics Committee. Written informed consent was obtained from each of the participants in accordance with research ethical principles, establishing a voluntary and anonymous participation. The measurements were made by a neutral investigator, using a paper form, in the regular class schedule of the students.
Data Analysis
A statistical analysis to describe the sample and the measured variables, central tendency, and dispersion statistics were calculated. Cronbach's alpha coefficient and McDonald's omega coefficient were used to estimate the internal consistency of the scales and their dimensions. The Students t-test for independent samples was applied to compare the mean empathy scores of men and women, by estimating the d of Cohen's formula as a measure of effect size (Cohen, 1992). One-way ANOVA was used to compare the empathy means according to educational level, calculating the square Eta (η2) as a measure of effect size. Prior to the calculation of the Student t-test and the ANOVA, the Shapiro-Wilk test was used to test the normality of the variables and the Levene test to evaluate homoscedasticity. The correlations between empathy scores, and its three dimensions, with personality styles, were calculated using Pearson's linear correlation coefficient (rxy), establishing the association for both the total sample and the sample segmented by gender. To predict empathy values and their dimensions, Hierarchical Multiple Linear Regression models were estimated, using the different measures of personality styles as predictor variables. A Multiple Group Confirmatory Factor Analysis was performed to analyze the factorial invariance of empathy measurement models, considering each gender as a sample to evaluate the goodness of fit of the models used for the Chi-square statistic, the Goodness of Fit Index (GFI) and the Root Mean Square Error of Approximation (RMSEA) as absolute adjustment measures, the Adjusted Goodness of Fit Index (AGFI), the Tucker-Lewis Index (TLI), Confirmatory Fit Index (CFI). and the Akaike Information Criterion (AIC). It is not possible to apply Confirmatory Factorial Analysis (CFA) to the measures of the MIPS since the factors are not independent when sharing weighted items in a differentiated manner between one factor and another. The significance level used in all cases was α ≤ .05. All statistical analyzes were performed with IBM SPSS Statistics V.22 (IBM Corp., 2013) and AMOS 21.
Results
Table 1 presents the descriptive statistics and the reliability coefficients of alpha of Cronbach and omega of McDonald of the empathy scale and its dimensions, considering the total sample (n = 278).
Table 1 Descriptive Statistics and Reliability of Empathy and its Dimensions
| Empathy | Min. | Max. | Mean | S.D. | Median | Alpha | Omega |
|---|---|---|---|---|---|---|---|
| Perspective adoption | 35 | 70 | 60.65 | 7.295 | 62 | .734 | .746 |
| Compassionate care | 18 | 49 | 40.94 | 6.580 | 43 | .627 | .646 |
| Putting oneself in the shoes of another | 3 | 21 | 10.53 | 3.472 | 10 | .364 | .439 |
| General empathy | 67 | 140 | 112.12 | 12.464 | 114 | .733 | .773 |
Regarding the measure of empathy, the invariance of the factorial structure between female and male students was analyzed, the goodness of fit indices allows acceptance of the equivalence of the measurement models between the two samples. Although the Chi-square value allows rejecting the hypothesis of invariance (X2 = 287.121), the other indices support the unrestricted model of invariance (GFI = .970, CFI = .983, RMSEA = .035, AIC= 416.102). The goodness of fit for each population is presented (Table 2) and the goodness of fit of the invariance models (Table 3).
Table 2 Goodness-of-fit Indices of the AFC of the Measure of Empathy for Women and Men, and Factorial Invariance
| Model | X2 | CFI | RMSEA | ACFI | TLI | CFI | CMIN/DF | AIC |
|---|---|---|---|---|---|---|---|---|
| 3 Factor model (Women) | 118.750 | .970 | .050 | .949 | .976 | .983 | 2.527 | 180.750 |
| 3 Factor model (Men) | 173.350 | .977 | .054 | .952 | .977 | .984 | 3.668 | 237.048 |
Table 3 Goodness-of-fit Indices for Factorial Invariance Models
| X2 | DF | CFI | NFI | CFI | RMSEA | AIC | |
|---|---|---|---|---|---|---|---|
| Model without Restrictions | 292.101 | 94 | .970 | .976 | .983 | .037 | 416.101 |
| Metric invariance | 350.388 | 104 | .965 | .971 | .979 | .039 | 454.388 |
When using a one-way ANOVA to compare the means of empathy and its dimensions according to educational levels (Table 4), there is no statistically significant difference between the means in the ability of the dimension to put oneself in the shoes of the other (p>.05, η2= .017), presenting differences only between the means of perspective adoption (p <.05, η2 = .059), compassionate care (p <.001, η2 = .073), and in the general empathy scale (p <.05, η2 = .042). Showing an average effect in the Eta square (η2) of the three variables, in which differences between the means are observed.
Table 4 Descriptive Statistics of Empathy and Comparison According to Educational Levels
| 1° | 2° | 3° | 4° | 5° | 6° | ||||
|---|---|---|---|---|---|---|---|---|---|
| Empathy | Mean (S.D.) | Mean (S.D.) | Mean (S.D.) | Mean (S.D.) | Mean (S.D.) | Mean (S.D.) | Total | F | p |
| Putting oneself in the shoes of another | 10.69 (3.58) | 10.71 (3.75) | 10.33 (4.37) | 9.71 (2.78) | 10.0 (3.12) | 11.29 (2.42) | 10.53 (3.47) | .939 | .456 |
| Perspective adoption | 58.62 (7.73) | 60.76 (6.40) | 61.67 (7.0) | 60.87 (6.70) | 62.76 (6.60) | 63.04 (9.47) | 60.65 (7.30) | 3.339 | .041 |
| Compassionate care | 38.60 (6.55) | 41.51 (6.64) | 45.0 (5.33) | 41.38 (6.46) | 40.35 (5.11) | 41.25 (6.65) | 40.94 (6.58) | 4.263 | .001 |
| General empathy | 107.92 (13.77) | 112.98(11.41) | 117.0 (11.89) | 111.96 (11.13) | 113.12 (9.45) | 115.57 (14.04) | 112.12 (12.46) | 2.394 | .011 |
When comparing the average scores of the empathy scale and its dimensions according to gender, it is evident that there are no statistically significant differences in perspective adoption and putting oneself in the shoes of the other (p>.05), which is confirmed by the confidence interval for the differences between the means and a small effect size (d), there being only differences with compassionate care and in the total empathy score, where women have higher means than men, although a small size is retained of the effect (Table 5).
Table 5 Comparison of Empathy Means and their Different Dimensions According to Gender
| Men | Women | IC 95% dif. | |||||||
|---|---|---|---|---|---|---|---|---|---|
| (n=112) | (n=166) | ||||||||
| Empathy | Mean | S.D. | Mean | S.D. | t | p | li ls | d | |
| Perspective adoption | 60.06 | 6.99 | 61.05 | 7.49 | -1.11 | .26 | -2.7 | .76 | -.13 |
| Compassionate care | 39.76 | 6.54 | 41.73 | 6.51 | -2.47 | .01 | -3.5 | -.41 | -.30 |
| Putting oneself in the shoes of another | 10.05 | 3.20 | 10.84 | 3.62 | -1.86 | .06 | -1.6 | .04 | - .23 |
| General empathy | 109.9 | 12.2 | 113.6 | 12.45 | -2.48 | .01 | -6.7 | -.79 | -.30 |
Note: d=Cohen's d (Effect size); S.D.= Standard Deviation.
The personality styles that stand out with the greatest presence in medical students are control, conformism, reflection, sensation, and firmness, with average prevalence greater than 68, of them, sensation is higher in women than in men (p <.001, d = -.408), and reflection is greater in men (p <.001, d = .739), there being statistically significant differences between men and women in slightly more than half of the personality styles, as seen in Table 6.
Table 6 Descriptive Statistics and Comparison of Personality Styles According to Gender
| MIPS | Total | Men (n=112) | Women (n=166) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | α | Mean | SD | Mean | SD | t | p | d | |
| Openness | 65.6 | 19.8 | .72 | 66.8 | 20.7 | 64.9 | 19.32 | 0.78 | .431 | .096 |
| Preservation | 33.6 | 20.1 | .83 | 29.9 | 20.6 | 36.1 | 19.41 | -2.53 | .012 | -.308 |
| Modification | 60.6 | 19.9 | .71 | 57.2 | 20.2 | 62.9 | 19.59 | -2.34 | .020 | -.286 |
| Accommodation | 39.9 | 20.7 | .67 | 44.8 | 22.2 | 36.7 | 19.06 | 3.14 | .002 | .390 |
| Individualism | 62.1 | 25.1 | .74 | 65.5 | 26.7 | 59.8 | 23.91 | 1.81 | .072 | .224 |
| Protection | 46.6 | 23.6 | .75 | 40.4 | 25.8 | 50.8 | 21.15 | -3.54 | .001 | -.442 |
| Extraversion | 49.8 | 24.5 | .74 | 51.9 | 24.8 | 48.5 | 24.33 | 1.13 | .257 | .139 |
| Introversion | 48.0 | 27.0 | .70 | 47.9 | 27.3 | 48.1 | 26.96 | -0.07 | .944 | -.008 |
| Sensation | 68.9 | 20.6 | .65 | 63.9 | 21.2 | 72.3 | 19.55 | -3.36 | .001 | -.408 |
| Intuition | 33.2 | 21.7 | .75 | 40.7 | 22.3 | 28.2 | 19.91 | 4.90 | .001 | .593 |
| Reflection | 71.1 | 21.9 | .70 | 80.2 | 20.2 | 64.9 | 20.93 | 6.03 | .001 | .739 |
| Affectivity | 37.5 | 24.1 | .79 | 32.2 | 25.2 | 41.2 | 22.73 | -3.04 | .003 | -.376 |
| Systematization | 62.9 | 21.2 | .68 | 58.7 | 24.0 | 65.6 | 18.68 | -2.55 | .011 | -.320 |
| Innovation | 35.6 | 22.6 | .77 | 43.9 | 21.6 | 30.0 | 21.70 | 5.23 | .001 | .640 |
| Retreat | 48.3 | 25.9 | .81 | 48.1 | 26.5 | 48.5 | 25.60 | -0.11 | .907 | -.014 |
| Communicability | 56.8 | 24.2 | .80 | 63.9 | 22.8 | 51.9 | 24.06 | 4.15 | .001 | .511 |
| Hesitation | 39.3 | 24.1 | .77 | 38.8 | 24.4 | 39.6 | 24.03 | -0.25 | .803 | -.031 |
| Firmness | 68.3 | 21.3 | .68 | 68.1 | 23.7 | 68.5 | 19.68 | -0.13 | .894 | -.017 |
| Discrepancy | 46.1 | 23.1 | .76 | 48.9 | 21.3 | 44.3 | 24.24 | 1.65 | .099 | .205 |
| Conformism | 71.2 | 19.6 | .75 | 74.3 | 19.3 | 69.1 | 19.60 | 2.19 | .029 | .269 |
| Subjection | 28.7 | 22.4 | .60 | 28.7 | 24.0 | 28.7 | 21.43 | 0.00 | .997 | .000 |
| Control | 71.5 | 20.6 | .61 | 71.4 | 23.4 | 71.6 | 18.65 | -0.08 | .936 | -.010 |
| Dissatisfaction | 44.2 | 24.9 | .74 | 48.5 | 24.2 | 41.3 | 25.01 | 2.37 | .018 | .291 |
| Agreement | 50.1 | 23.2 | .66 | 43.0 | 23.9 | 55.0 | 21.55 | -4.36 | .001 | -.527 |
Note: α= Cronbach alpha, d=Cohen's d (Effect size).
Table 7 shows the results of the Pearson correlation between personality styles and the total empathy score, considering the total sample and the sample segmented by gender. Showing low eminently associations, but significant, in most established relationships. From the perspective of motivational goals, total empathy presents a direct relationship with openness, modification, and protection and an inversely proportional relationship with preservation, accommodation, and individualism. Associations remain in the subsample of men, except for the scale of individualism and lack of association with empathy in the dimensions of openness, preservation, and accommodation in the subsample of women. From the perspective of the cognitive modes of medical students general empathy is not associated with intuition and innovation, both for the total sample and for the sample segmented by gender, maintaining a direct and significant association with extroversion (r = .130,p = .015), affectivity (r = .145, p = .008), and systematization (r = .110, p = .033) and an inverse relationship with introversion (r = -.164,p = .003), sensation (r = -.102,p = .045), and reflection (r = -.162,p = .003). When examining the correlation between general empathy and cognitive modes according to gender the association with extroversion, introversion, and systematization only turn out to be significant in men, with sensation being the only cognitive mode that is significant and inversely proportional in women (r = -.151, p = .007), indicating that women with greater empathy have less orientation to obtain knowledge of the concrete, tangible or observable and that men who are more empathetic tended to be extroverted rather than introverted, and tend to systematize and organize their experience and be more efficient. Likewise, in the different scales that reveal the styles of interpersonal behavior, there is an absence of correlation between empathy with conformism (p = .238), subjection (p = .078), and control (p = .412), showing directly proportional correlations with communication (r = .102,p <.05), firmness (r = .174, p <.01), and concordance (r = .110, p <.05), and inverse correlations with withdrawal (r = -.180, p <.01), hesitation (r = -.174, p <.01), discrepancy (r = -.186, p <.01), and dissatisfaction (r = -.193,p <.01), associations that are mostly confirmed in the subsample of men where control is added (r = .151, p <.05), and conformism (r = .166, p <.05), canceling the associations between empathy and most of the scales of interpersonal behavior in the subsample of women, where only a significant and inversely proportional association with discrepancy (r = -.180, p <.05) and dissatisfaction (r = -.212, p <.05), and a direct relationship with concordance (r = .164,p <.05).
Table 8 shows that, of the six personality styles referred to as the motivational goals, only protection shows an association directly proportional to perspective adoption (r = .163, p <.01), an association that is confirmed in the subsample of women (r = .217, p <.05) where in addition an inverse relationship with individualism is added (r = -.260,p <.01); associations that are not observed in men where the perspective adoption is positively associated with modification (r = .199, p = <.01). Perspective adoption is positively associated with affectivity in both men (r = .141, p <.05), and women (r = .157, p <.05), showing association with the bipolar dimension of extroversion-introversion only in men, where perspective adoption is positively associated with extroversion (r = .163,p <.05) and negatively with introversion (r = -.205,p <.01). In the styles of interpersonal behaviors, associations differentiated between men and women are presented, where perspective adoption presents an inverse association with discrepancy (r = -.121, p <.05) and dissatisfaction (r = -.098, p <.05), and a direct association with agreement (r = .061, p <.01) in the subsample of women; associations that do not present in men, in whom perspective adoption is associated directly with communicability (r = .216, p <.01),firmness (r = .223,p <.01), and conformism (r = .172,p >.05) and inversely with withdrawal (r = -.199, p <.01) and hesitation (r = -.178, p <.05).
Table 8 Correlation Between Personality Styles and the Dimension Perspective Adoption According to Gender
| Personality styles | Perspective adoption | ||
|---|---|---|---|
| Total | Women | Men | |
| Motivational goals | |||
| Openness | .096 | .081 | .112 |
| Preservation | -.041 | -.009 | -.081 |
| Modification | .076 | -.135 | .199** |
| Accommodation | -.097 | -.043 | -.118 |
| Individualism | -.055 | -.260** | .100 |
| Protection | .163** | .217* | .106 |
| Cognitive modes | |||
| Extraversion | .116* | .057 | .163* |
| Introversion | -.141** | -.043 | -.205** |
| Sensation | -.060 | -.151 | -.024 |
| Intuition | -.002 | .059 | -.012 |
| Reflection | -.068 | -.068 | -.031 |
| Affectivity | .157** | .157* | .141* |
| Systematization | .052 | .052 | .097 |
| Innovation | -.019 | -.019 | .029 |
| Interpersonal behaviors | |||
| Retreat | -.140** | -.140 | -.199** |
| Communicability | .113* | .113 | .216** |
| Hesitation | -.101* | -.101 | -.178* |
| Firmness | .116* | .116 | .223** |
| Discrepancy | -.121* | -.121* | -.079 |
| Conformism | .101* | .101 | .172* |
| Subjection | -.009 | -.009 | -.039 |
| Control | .008 | .008 | .118 |
| Dissatisfaction | -.098* | -.098* | -.015 |
| Agreement | .061 | .061* | -.037 |
Note:*p<.05; **p<.01
Table 9 presents the association of the dimension compassionate care with the different personality styles. From the perspective of motivational goals, an inverse association with preservation is observed in both men (r = -.276, p <.01) and women (r = -.172, p <.05), which implies that as care increases, attention decreases towards the pessimistic visibility of the problems of life. This also shows an inversely proportional association with accommodation and individualism, and a direct association with openness and protection. In cognitive modes, only reflection has an inverse association with compassionate care (r = -.147, p <.oi), which in turn occurs in the subsample of women (r = -.154,p <.05), an association that does not exist in the sample of men. From the perspective of interpersonal behaviors, an inversely proportional relationship of compassionate care with withdrawal stands out (r = -.120,p <.05), hesitation (r = -.183,p <.oi), discrepancy (r = -.151,p <.01), submission (r =-.116,p <.05), and dissatisfaction (r = -.156,p <.01), for which many of these associations are confirmed in the subsample of men, but not in women, where only dissatisfaction has an inverse correlation (r = -.167,p <.05), being the only interpersonal behavior in women that correlates with compassionate care.
Table 9 Pearson Correlation Between Personality Styles and the Dimension Compassionate Care According to Gender
| Personality style | Compassionate care | ||
|---|---|---|---|
| Total | Women | Men | |
| Motivational goals | |||
| Openness | .187** | .114 | .257** |
| Preservation | -.205** | -.172* | -.276** |
| Modification | .087 | -.136 | .211** |
| Accommodation | -.207** | -.117 | -.239** |
| Individualism | -.112* | -.220** | -.005 |
| Protection | .125* | .125 | .073 |
| Cognitive modes | |||
| Extraversion | .085 | .043 | .133* |
| Introversion | -.091 | -.023 | -.141* |
| Sensation | -.066 | -.202* | -.023 |
| Intuition | -.071 | .044 | -.088 |
| Reflection | -.147** | -.154* | -.071 |
| Affectivity | .023 | .020 | -.022 |
| Systematization | .069 | -.024 | .108 |
| Innovation | -.016 | -.030 | .071 |
| Interpersonal behaviors | |||
| Retreat | -.120* | -.088 | -.148* |
| Communicability | .062 | -.043 | .196** |
| Hesitation | -.183** | -.064 | -.272** |
| Firmness | .186** | .039 | .309** |
| Discrepancy | -.151** | -.146 | -.135* |
| Conformism | -.030 | -.132 | .069 |
| Subjection | -.116* | -.028 | -.185** |
| Control | .048 | -.119 | .190** |
| Dissatisfaction | -.156** | -.167* | -.118 |
| Agreement | .059 | .141 | -.066 |
Note: *p<.05; **p<.01
Table 10 presents the association of the different personality styles with the ability to put oneself in the shoes of the other, highlighting for the total sample a direct correlation with protection (r = .139, p <.01) and inverse with individualism (r = -.178,p <.01), being the only motivational goals that present statistical significance; those that are confirmed in the subsample of men, but not in women where no personality style, through the three great areas of personality, shows a significant association with putting oneself in the shoes of the other. From the perspective of cognitive modes, of the total sample, there is a direct correlation where affectivity is observed (r = .145, p <.01) and systematization (r = .157, p <.01), and inverse with introversion (r = -.118, p <.05), sensation (r = -.113, p <.05), and reflection (r = -.159, p <.01), which are eminently replicated in the subsample of men. Regarding interpersonal behaviors, inverse correlations with withdrawal are presented (r = -.123, p <0=.05), discrepancy (r = -.126,p <.05), and dissatisfaction ( r = -.192, p <.05), with a direct correlation with concordance (r = .154, p <.05).
Table 10 Pearson Correlation Between Personality Styles and the "Ability to put Oneself in Each Other's Shoes" According to Gender
| Personality styles | Ability to put oneself in each other's shoes | ||
|---|---|---|---|
| Total | Women | Men | |
| Motivational goals | |||
| Openness | .054 | -.057 | .135* |
| Preservation | -.055 | .065 | -.161* |
| Modification | .043 | -.024 | .060 |
| Accommodation | -.082 | .008 | -.113 |
| Individualism | -.178** | -.081 | -.227** |
| Protection | .139** | .059 | .164* |
| Cognitive modes | |||
| Extraversion | .063 | .038 | .092 |
| Introversion | -.118* | -.023 | -.178* |
| Sensation | -.113* | -.134 | -.142* |
| Intuition | -.032 | .068 | -.045 |
| Reflection | -.159** | -.092 | -.151* |
| Affectivity | .145** | .143 | .118 |
| Systematization | .157** | -.006 | .259** |
| Innovation | -.094 | .075 | -.146* |
| Interpersonal behaviors | |||
| Retreat | -.123* | -.089 | -.147* |
| Communicability | .013 | .062 | .030 |
| Hesitation | -.066 | .101 | -.170* |
| Firmness | .030 | -.066 | .098 |
| Discrepancy | -.126* | .078 | -.218** |
| Conformism | .001 | -.113 | .091 |
| Subjection | -.066 | -.025 | .095 |
| Control | -.061 | -.058 | -.067 |
| Dissatisfaction | -.192* | -.001 | -.282** |
| Agreement | .154* | .103 | .151* |
Note: * p<.05; ** p<.01
To determine the personality styles associated with empathy and its different dimensions and to predict the values of empathy, several models of Hierarchical Multiple Linear Regression were tested. The ANOVA of the regression model indicates that it is possible to predict the total empathy score from the personality styles ofprotection, sensation, and accommodation (F = 9.72, p <.001), which as a whole can explain 9.7% of the variance (R2 = .097). It is possible to predict perspective adoption from the styles of protection and firmness (F =6.83, p <.001), which as a whole explain 4.8% of the variance (R2 = .048). The styles of protection, reflection, affection, and firmness are variables that allow to predict compassionate care (F = 7.78, p <.001), which as a whole explains 10.3% of the variance (R2 = .103). The ability to put oneself in the shoes of the other is possible to be predicted from the dissatisfaction and affectivity (F = 8.17, p <.001), which as a whole can explain 5.7% of the variance (R2 = .057). The coefficients of the different models and their statistical significance are presented in Table 11.
Table 11 Hierarchical Multiple Linear Regression to Predict Empathy
| Model | β | EE | βe | t | Sr | |
|---|---|---|---|---|---|---|
| General empathy | Constant | 116.825 | 3.340 | 34.98** | ||
| Protection | .112 | .029 | .220 | 3.81** | .220 | |
| Sensation | -.075 | .034 | -.128 | -2.18* | -.126 | |
| Accommodation | -.113 | .034 | -.195 | -3.32* | -.191 | |
| Perspective adoption | Constant | 56.129 | 1.554 | 36.11** | ||
| Protection | .054 | .017 | .191 | 3.23** | .191 | |
| Firmness | .034 | .019 | .110 | 1.85* | .110 | |
| Compassionate care | Constant | 41.195 | 2.215 | 18.60** | ||
| Protection | .063 | .023 | .240 | 2,80** | .161 | |
| Reflection | -.064 | .020 | -.225 | -3.15** | -.181 | |
| Affectivity | -.075 | .025 | -.289 | -3.05** | -.176 | |
| Firmness | .064 | .017 | .219 | 3.74** | .216 | |
| Ability to put oneself in each other's shoes | Constant | 10.961 | .533 | 20.58** | ||
| Dissatisfaction | -.027 | .008 | -.190 | -3.23** | -.190 | |
| Affectivity | .020 | .009 | .138 | 2.34* | .138 |
Note: *p<.05; **p<.01; βe: Beta standardized coefficient; Sr: Semi partial correlation
Discussion and Conclusions
This study analyzes the association that exists between empathy and personality styles in medical students at the University of Azuay. In general, there are low to moderate correlations, which vary from + -.10 to + -.32 between the different personality styles and the general empathy score, depending on whether the general sample or the sample segmented by gender is valued, moderate associations concordant with those found in other studies with medical students (Song & Shi, 2017).
Perspective adoption is not homogeneous when comparing the first academic year with the students of the final academic year, however compassionate care, the ability to put in oneself in the shoes of the other, and the total empathy do not differ statistically between the academic levels of first to sixth year of students in medicine, these last two results concur with the conclusions of the jsPE-s applied to 1188 students of medicine (Quince et al., 2016). There are still contradictory results regarding the evolution of empathy along academic levels, where sometimes no differences have been found (Montilva, García, Torres, Puertas, & Zapata, 2015) or others in which there are differences between some levels (Wen et al., 2013).
The students present a mean score of total empathy of 112 points (SD = 12.46), higher than the measurements made in Japan, South Korea, China, Kuwait, India, Iran, UK, Australia, Colombia, and Dominican Republic, that vary between 101.4 and 110.09 (Roff, 2015) and lower than that found in Brazil (M = 114.95, SD = 12.41) (Paro et al., 2012), Portugal (M global average = 114.4, SD rank = 10.43-11.57) (Duarte, Raposo, Rodrigues, & Branco, 2016), USA average between years 2002 and 2012 (M = 114.3, SD = 10.4) (Hojat & Gonnella, 2015) and sample composed of students first and second years medical school of UK, New Zealand, and Ireland (M = 113.03, SD = 10.30) (Quince et al., 2016). In the present study men had a mean of empathy of 109.88 (SD = 12.20) and women a mean of 113.63 (SD = 12.45), quite similar to those found in Portugal and the United Kingdom (Alcorta-Garza, González-Guerrero, Tavitas-Herrera, Rodriguez-Lara, & Hojat, 2005) showing women had an average score slightly higher than that of men, as has already been observed in other samples of medical students (Leombruni et al., 2014; Wen et al., 2013), a trend consistent with 11 other countries in which the jspe-s has been used in medical students (Quince et al., 2016), which differs from that observed in professional doctors where empathy does not present significant differences according to gender (Delgado et al., 2016; Hojat et al., 2002; Montilva et al., 2015), diminishing the differences in the extent to which the profession is exercised.
From the perspective of personality styles, dissatisfaction, understood as a tendency to be passive-aggressive and ill humored, presents a negative association with general empathy, both at the level of the total sample and the sample according to gender, which is consistent with evidence that inversely relates neurotic traits to empathy and its dimensions (Song & Shi, 2017).
The results of the study reveal that different personality styles are significantly associated with empathy and that it is possible to predict the general empathy of the students from the styles of protection, sensation, and accommodation, which explains the 9.75 of the variance of the total score of empathy, predictions slightly lower than those reported with Chinese medical students, associating empathy and its dimensions with the five major personality factors (Song & Shi, 2017). There is a diversity of personality style respect to which the association they present with empathy and its components as well as gender differences is unclear, as occurs for example with the introversion-extroversion or firmness dimension what requiring more research about it.
It is observed that people with a high degree of empathy find it easy to accept the ups and downs of life, be optimistic about the future, trying to change their environment to achieve their goals and being motivated to satisfy others first; with a low orientation toward pessimism, lack of initiative, self-centeredness, and the satisfaction of their own needs. Female students present higher levels both in general empathy and in the dimension of compassionate care, which is inversely associated with the presence of an individualistic style, with sensation and preservation, an association that in general does not occur in the subsample of male students, who stand out for their firmness and hesitation.
It has been taken with great precaution to examine the invariance of measures of empathy according to gender, accepting that the factorial structure is invariant, which contributes to the validity of comparisons based on gender. This also yields a Cronbach's alpha of .733, similar to .74 reported in the original Spanish version (Alcorta-Garza et al., 2005) and reliability coefficients for the scale of personality styles slightly lower than those established by Millon (1994), which reports values which vary between .69 and .85 for the different dimensions; showing thus the measures used an acceptable reliability when estimated in the sample under study.
One limitation of this study was its cross-sectional design, which does not allow establishing the evolution of the development of empathy in students during their medical training, or establishing causal relationships between variables, requiring a longitudinal study that accounts for the relationship of both constructs. A second limitation is associated with the lack of comparability of the study with other studies that relate empathy with personality, since a different measure was used to establish the personality construct, and studies that used the Big Five (Song & Shi, 2017), the NEO-FFI inventory (Mooradian, Davis, & Matzler, 2011) or the MMPI, but not the MIPS that was used in the present study. Despite this, it is worth considering that the MIPS is an index that measures personality styles from a non-pathological perspective (Millon, 2003) and in that sense can be considered a more appropriate measure to be used in the population of university students, also providing a more detailed measurement to the limitations that the more global measurements based on the five big factors (Big Five) (Mooradian et al., 2011; Paunonen, Jackson, Trzebinski, & Forsterling, 1992). Existing in turn difficulty of generalizing the results or their conclusions to populations of medical students from other countries, requiring new studies to establish the regularities or singularities between countries. A final observation necessary to consider is related to the reliability coefficients that show some dimensions of the personality styles scale, which are lower than a desirable value of .70, which however is consistent with that reported in the literature regarding the reliability of personality tests, which do not respond adequately to the methods provided by the classical test theory. In addition, the assumption for estimating Cronbach's alpha is not fulfilled given that the personality style and empathy scales are multidimensional, violating the one-dimensional assumption of the construct, where the items are highly correlated with each other (Welch & Comer, 1988).
Despite the abovementioned, the study advances in unveiling the relationship between empathy and personality of students, that in spite of the ample literature on empathy, there are still very few studies that relate it to personality, expanding the information necessary to strengthen the various curricular measures, the formation or training that are being carried out to promote or enable the development of empathy of medical students, with the various benefits that this may bring to the doctor-patient relationship or to the effectiveness of medical treatments, or in methods of assessing the quality of care (Mercer & Reynolds, 2002), assessing empathy and personality styles as relevant elements in this context.
Some authors have argued that the attribute empathy is the result of a complex dialectical relationship between its components or dimensions (Calzadilla-Núñez et al., 2017; Díaz-Narváez, Estrada-Méndez, Arévalo-López, Calzadilla-Núñez, & Utsman-Abarca, 2017), all of which are influenced, in turn, by many different factors (Díaz-Narváez, Alonso-Palacios et al., 2017), of which, one of them is the personality. At the same time, the construction of this attribute has depended on the evolution of man (Decety & Harris, 2011) and the processes of ontogeny (Díaz-Narváez, Calzadilla-Núñez et al., 2017). The theoretical inference that is feasible to infer from the postulates of these authors is that empathy does not depend on a relevant variable, but on many variables acting at the same time and each of them contributes to its construction, but with different intensities due to different external means in which the subject develops. This situation is consistent with the results observed in several Latin American countries (Bilbao et al., 2015; Díaz-Narváez et al., 2014; Silva et al., 2014).
On the other hand, differences have been found in the levels of empathy among students of different dental schools within a country, as well as differences among students from different countries (Diaz-Narvaez, Erazo et al., 2017). In all these cases it is observed that the variables studied explain little about the behavior of the empathy. The aforementioned complexity, in the determination of empathy allows another complex inference to be made, consisting of the fact that the teaching-learning processes associated with it could not depend on a universal strategy and strictly standardized methods, but rather that such a teaching strategy (and the corresponding methods) must emanate from the specific characteristics of the empathic situation of each student population.
Finally, this study analyzes the relationship between empathy and its dimensions, and the different personality styles, in a sample of Ecuadorian medical students. The results show moderate to low relationships, but significant, of the different personality styles, showing total empathy as a positive association with protection and a negative association with sensation and accommodation. Achieving prediction for perspective adoption from the style of perspective protection and firmness; compassionate care from protection, reflection, affection, and firmness; and the ability to put oneself in the shoes of the other can be predicted from affection and dissatisfaction. From the perspective of the academic training of medical students, we agree with other authors who state that empathy can be improved and taught, that it is possible to approach and integrate different styles or personality factors that allow improving empathy in the medical education, setting pedagogical objectives oriented to the development of empathy or proposing individualized strategies of intervention based on personality traits, understanding that any intervention strategy must be comprehensive due to the complexity of the attribute intervened.
The results found are consistent with other works carried out in Latin America and allow visualizing a consistent tendency in which this attribute has a multifactorial character and it is important to study all the variables that influence it and determining its specific weight to correctly guide all possible pedagogical intervention.
