The Dark Triad Of Personality And Work Efficiency Of Kindergarten Teachers

Conceptual framework

The Dark Triad of personality. In the study, the Dark Triad of personality is considered in accordance with the theoretical model in which the Dark Triad of personality is understood as a single complex, causing different levels of socially undesirable behavior and includes three interrelated, yet relatively independent traits – Machiavellianism, narcissism, and psychopathy (Paulhus & Williams, 2002; Jones & Paulhus, 2014).

The work of kindergarten teachers. This work is understood as a qualified professional activity aimed at the development and education of preschool children within the educational conditions and programs provided by preschool educational institutions. However, certain work efficiency indicators of kindergarten teachers are still debatable (e.g., Adewale, 2013; Araujo, Carneiro, Cruz-Aguayo, & Schady, 2016; Bassok, Latham, & Rorem, 2016). In this regard, the current study used expert assessment by professionals with extensive experience in the organization and execution of kindergarten activities. Moreover, the chosen indicators were not profession-specific in order to be able to compare various professional groups in the future.

Participants and procedure

Kindergarten teachers were invited to participate in this study aimed at studying the influence of personality traits on the effectiveness of professional activity. All kindergarten teachers who agreed to participate in the study were selected. Kindergartens in which the study was conducted were located in several regions of Russia (Moscow, Saint-Petersburg, Tver, Ulyanovsk and others).

After excluding cases containing missing data, the final sample consisted of 129 kindergarten teachers – all female, aged from 19 to 58 years ( M = 32.15, SD = 10.47), work experience as a kindergarten teacher ranged from 0,3 to 33 years ( M = 7.94, SD = 8.77). In addition, the study involved experts, who were highly qualified employees of the kindergartens where the study was conducted and performed expert evaluation of the work efficiency indicators. All research participants were volunteers. In accordance with ethics requirements, written informed consent was obtained from all participants.

The measured variables and the research design

In accordance with the main purpose of the study, its design was aimed primarily at identifying the association of the Dark Triad traits of personality with the work efficiency in kindergarten teachers. In this regard, we used three groups of variables:

• independent variables – – the traits of the Dark Triad considered as factors and in this respect associated with groups divided by levels of each factor;

• dependent variables – indicators of work efficiency of kindergarten teachers;

• other controlled variables – age and work experience.

The description of all variables and groups is given in Table 01 .

The measured variables and the research design

The Dark Triad. The personality traits of the Dark Triad were assessed using the "Short Dark Triad (SD3)" questionnaire (Jones & Paulhus, 2014) in its Russian-language adaptation by Egorova, Sitnikova, and Parshikova (2015). This adapted questionnaire:

• includes 27 items (9 items per subscale – Machiavellianism, narcissism, and psychopathy) which include short sentences that describe the behavior characteristic of the corresponding personality traits;

• provides for the evaluation of each statement on a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree );

• shows satisfactory psychometric properties; in particular, for Machiavellianism, narcissism, and psychopathy the Cronbach's alpha was equal to .74, .72, .70 respectively (Egorova, Sitnikova, & Parshikova, 2015).

The efficiency of the work. The work efficiency indicators of kindergarten teachers were assessed by experts on a 5-point Likert scale (from 1 = very low to 5 = very high ). The number of experts evaluating the work of one kindergarten teacher ranged from 3 to 8. Expert ratings for each indicator were averaged. We used five indicators and the total score (see Table 01 ).

Age and work experience. Age was measured as the number of full years at the time of the study. The work experience (see Table 01 ) was also measured as the number of full years with one decimal digit (if work experience was less than one year).

Data analysis

The main method of data analysis used to identify the association of the Dark Triad personality traits on work efficiency was a one-way analysis of variance (ANOVA). In addition, we used Pearson correlations and a number of well-known methods of statistical hypothesis testing (Mann-Whitney test etc.).

The level of the Dark Triad traits

Before the ANOVA, the whole sample ( N = 129) was divided into three groups corresponding to low, medium and high levels of each Dark Triad trait based on the frequency distribution. Descriptive statistics for the selected groups and the whole sample are shown in Table 02 for the Dark Triad traits, age and work experience and in Table 03 for work efficiency indicators.

For most of the studied variables and groups, the normality of distribution was confirmed by the test of Kolmogorov-Smirnov and/or Shapiro-Wilk test. The remaining cases allow us to conclude only relatively small deviations from normality, that based on evaluation of the ratio of skewness and kurtosis to standard errors (see, e.g., Field, 2009, p. 139). At the same time, one-way ANOVA is very robust to violations of normality (e.g., Field, 2009, p. 360), which allowed us to use this method (see section 6 .3).

For the comparison of the results of the Dark Triad levels, we failed to find any relevant data obtained on a sample of kindergarten teachers. Thus, we analyzed several studies conducted on professionally non-specific female samples (see Table 04 ).

The mean value for Machiavellianism obtained in our study was significantly lower compared to the one reported by Egorova, Sitnikova, and Parshikova (2015) and by Jones and Paulhus (2014) in Study 1, but did not differ from the mean values obtained by Jones and Paulhus in Study 2 and Study 3, and by Jonason, Wee, Li, and Jackson (2014). In turn, the mean value for narcissism in the current study was significantly higher compared to the one reported by Egorova et al. (2015), Jonason et al. (2014), and Jones and Paulhus in Study 1 and Study 3, but did not differ from the mean obtained in Study 2 by Jones and Paulhus. Finally, the mean value for psychopathy did not differ from the results of the four studies used for comparison (see Table 04 ).

Thus, based on these results we suggest that Russian female kindergarten teachers:

• have lower levels of Machiavellianism than on average in the Russian population, and lower or equal level in comparison to various foreign female samples;

• have higher levels of narcissism than on average in the Russian population, and a higher or equal level in comparison to various foreign samples of women;

• show the same levels of psychopathy as Russian women on average, as well as women in the USA and Canada.

It should be noted that the results on the level of narcissism were quite unexpected and might be related to an ambivalent role of narcissism in the professional activities of kindergarten teachers, we shall return to this in the discussion (section 6 .5). Another surprising result was on the levels of psychopathy: we expected that the professionals working with children would have shown lower levels than professionally non-specific groups.

Correlation between measures

We used the linear Pearson’s correlation coefficient to evaluate correlations between the variables. Although due to the above noted deviations from normality it would be more valid to use the nonparametric Spearman's correlation in some cases, we used the Pearson’s correlation to compare our results with other researchers. Table 05 presents the matrix of intercorrelations for the Dark Triad traits, work efficiency indicators, age and work experience.

For comparison of the correlations of the Dark Triad traits obtained in our study, we were unable to find any relevant data neither for kindergarten teachers nor for the female samples in general. Therefore, the results from studies using mixed sample (men and women) were compared:

• Egorova, Sitnikova, and Parshikova (2015) on a mixed Russian sample ( N = 571);

• Study 2 ( N = 279) and Study 3 ( N = 230) by Jones and Paulhus (2014) on a sample of men and women from Canada and the United States;

• Study 1 ( N = 422) by Jonason, Wee, Li, and Jackson (2014) on a sample of men and women from the USA.

We found a significant correlation between Machiavellianism and narcissism (see Table 05 ), which does not differ ( p > 0.5) from r = .31 reported by Egorova, Sitnikova, and Parshikova (2015) and r = .16 reported by Jonason, Wee, Li, and Jackson (2014), nor from the results of Study 2 ( r = .22) and Study 3 ( r = .29) by Jones and Paulhus (2014). The significant correlation found between Machiavellianism and psychopathy (see Table 05 ) did not differ significantly from r = .35 reported by Egorova et al. (2015), r = .47 from Study 1 of Jonason et al., (2014), nor from the results of Study 2 ( r = .40) and Study 3 ( r = .47) by Jones and Paulhus. At the same time, the non-significant correlation between narcissism and psychopathy (see Table 05 ) did not differ ( p > .05) from r = .31 from Study 2 by Jones and Paulhus, but was significantly lower than the correlations reported by Egorova et al. ( r = .42, p < .001), from Study 3 by Jones and Paulhus ( r = .42, p < .01), and reported by Jonason et al. ( r = .32, p < .05).

Based on these results, we suggest that for the sample of Russian female kindergarten teachers:

• the associations of Machiavellianism and narcissism, as well as Machiavellianism and psychopathy, do not differ from those characteristic for the Russian population (including both men and women), and mixed samples from Canada and the United States;

• in contrast, the correlation between narcissism and psychopathy is significantly lower compared with that characteristic for the Russian population, and is significantly lower or statistically not different from samples that include men and women from Canada and the United States.

The comparison was also performed among the correlations of the Dark Triad traits with age. In the present study, the correlations of Machiavellianism and narcissism with age were non-significant, unlike the correlation of psychopathy with age, which was significant and negative (see Table 05 ). None of these correlations were statistically different from the correlations of the Dark Triad traits with age reported by Egorova, Sitnikova, and Parshikova (2015) on the sample of Russian men and women: Machiavellianism ( r = -.009), narcissism ( r = -.026) and psychopathy ( r = -.031).

It should also be noted that the both age and work experience were found to be positively correlated with all work efficiency indicators, except Jointness ( Ef3 , see Table 05 ). This indicator might be to a greater extent mediated by relatively stable in relation to age and work experience personal characteristics such as extraversion and its components (sociability etc.).

We did not find any statistically significant correlations of Machiavellianism and narcissism with the work efficiency indicators. In contrast, psychopathy correlated negatively with Completeness ( Ef1 ) and Total efficiency ( TE , see Table 05 ). However, these correlations are very weak, this might be the result of a complex, nonlinear influence of psychopathy on the work efficiency, which is indirectly confirmed by the results of a one-way ANOVA (see section 6 .3 and Table 06 ).

The results of ANOVA

Table 06 presents the results of a one-way ANOVA meant to test the differences in work efficiency among the groups formed by the levels of the Dark Triad traits (high, medium, low). For all cases with a significant effect, the Levene's test showed homogeneity of variances, except for Competence ( Ef4 ) and groups with different levels of Machiavellianism (Mac 1, 2, and 3) with p =.48 (see Table 05 ). However, the violation of homogeneity of variances in this case was small, and an additional check using the Welch's test (Welch's robust test of equality of means) confirmed the significance of differences between groups ( p = .002).

Table 06 also presents the results of the post hoc tests with the Bonferroni's test. After correction for multiple comparisons, the differences remain statistically significant, except for the case of psychopathy (Psy) and Total efficiency ( TE ): the differences between the groups were only significant at p =.071.

Thus, we have to admit that the association between psychopathy and total work efficiency, which was originally identified by ANOVA, was not confirmed. However, we have found a number of statistically significant associations between Dark Triad traits and indicators of the work efficiency (see Table 06 ), which will be discussed in more detail in section 6 .4.

The differences in the associations between the Dark Triad traits and work efficiency

The results of our study show that the Dark Triad traits are significantly associated with the work efficiency. At the same time, these associations are specific for different traits, and these differences are manifested in two aspects.

First, each Dark Triad trait is associated with a particular set of the work efficiency indicators. Machiavellianism was only related to Competence ( Ef4 ), psychopathy – only to Completeness ( Ef1 ), and narcissism – to four indicators: Completeness ( Ef1 ), Creativity ( Ef2 ), Motivation ( Ef5 ) and Total efficiency ( TE ) (see Table 06 ).

Secondly, the type of the association differs among the Dark Triad traits. Thus, a higher level of Machiavellianism was associated with lower levels of Competence ( Ef4 ), but was only significant for the transition from medium to high level when the mean value of Competence dropped from 3.65 to 3.05 (see Table 03 ). When we merged the groups with low and medium Machiavellianism, the mean value of Competence in the combined group ( M = 3.58, n = 87) remained higher compared to the high-level group ( M = 3.05, n = 42) at p < .01 (here and further we are talking about the asymptotic 2-tailed significance from the Mann-Whitney test). Thus, the association of Machiavellianism and work efficiency cannot be called linear: rather, it is a specific quasilinear effect, which can be called "delayed quasilinear".

In turn, the association of narcissism with four efficiency indicators might be described as an inverted U-shaped curve. This pattern of association is clearly evident for the three indicators of efficiency (see Tables 03 and 06 ):

• Completeness ( Ef1 ) reaches a maximum at the medium level of narcissism ( M = 4.53) and drops significantly ( p < .01) with low ( M = 4.05) and high narcissism ( M = 4.08);

• Creativity ( Ef2 ) reaches a maximum at the medium level of narcissism ( M = 4.11) and drops significantly ( p < .01) with low ( M = 3.61) and high narcissism ( M = 3.68);

• Total efficiency ( TE ) reaches a maximum at the medium level of narcissism ( M = 4.06) and drops significantly ( p < .01) with low ( M = 3.67) and high narcissism ( M = 3.69).

The association of Motivation ( Ef5 ) with narcissism at the first glance has a different pattern close to the abovementioned quasilinear association with Machiavellianism. In fact, the relationship between narcissism and Motivation corresponds more to the inverted U-shaped curve and is significant ( p < .05) only at the transition from medium to high narcissism with mean of Motivation decreasing from 4.04 to 3.63 (see Tables 03 and 06 ). However, when we merged the low and medium narcissism groups, the mean of Motivation in the combined group ( M = 3.74, n = 86) was not statistically differ ( p > .05) from the high-level group ( M = 3.63, n =43) at p < .01. Thus, the association of narcissism and Motivation rather shows a pattern of an inverted U-shaped curve.

Psychopathy was found to be only associated with Completeness ( Ef1 ) showing a quasilinear pattern like in the case of Machiavellianism. However, there is a significant difference between the quasilinear patterns of associations of Machiavellianism and psychopathy with work efficiency: the nonlinear effect of Machiavellianism, as shown above, can be called "delayed quasilinear" (it is not present at the transition from low to medium level and is manifested only at the transition from medium to high levels) whereas the effect of psychopathy might be called "constrained quasilinear". While there were significant differences between low and medium psychopathy ( p < .01) with the mean of Completeness decreasing from 4.45 to 4.07, no differences were found between the medium and high-level groups ( p > .05), i.e. as if the decrease stops (see Tables 03 and 06 ). When we merged the medium and high level groups, the mean of Completeness in the combined group ( M = 4.11, n = 87) was significantly lower ( p < .01) than in the low-level group ( M = 4.45, n = 42). This confirms the specific quasilinear nature of the association of psychopathy with Completeness.

Thus, we identified three different types of nonlinear associations of the Dark Triad traits and work efficiency of kindergarten teachers:

• nonlinear pattern corresponding to an inverted U-shaped curve (characteristic of narcissism);

• "delayed quasilinear" pattern (characteristic of Machiavellianism);

• "constrained quasilinear" pattern (characteristic of psychopathy).

The ambivalent role of narcissism

The ambiguous role of the Dark Triad traits was noted by many researchers in terms of impact on the success of professional activities, and in other contexts (see, e.g., Egorova, Sitnikova, Parshikova, & Chertkova, 2016). In this respect, the ambivalence of narcissism found in our study contributes to the overall discussion about the particular role of this trait within the Dark Triad, which has been unfolding in recent years (see, e.g., Kowalski, Vernon, & Schermer, 2016). For example, on the one hand, narcissism is associated with low empathy and difficulties in identifying emotions (Jonason & Krause, 2013), and on the other hand, it is positively associated with emotional intelligence (Petrides, Pérez-González, & Furnham, as cited in Kornilova & Chumakova, 2016), success in interpersonal relationships (Foster & Campbell, as cited in Kornilova & Chumakova, 2016; Morf & Rhodewalt, as cited in Nübold et al., 2017) and morality in individuals with low self-esteem (Zuo, Wang, Wang, & Zhao, 2016). Some authors point directly to the ambivalent role of narcissism. For example, Kornilova and Chumakova (2016) revealed a positive correlation of narcissism with extraversion and negative emotional stability.

Our study also revealed an ambiguous association of narcissism and work efficiency of kindergarten teachers: both high and low levels of narcissism are associated with a decrease in the four indicators of work efficiency, with the medium level being optimal (see section 6 .4).

The first possible reason for this ambivalence lies in the specifics of this professional activity. The activity of the kindergarten teacher is largely demonstrative and relies on the external evaluation of outsiders – children, parents, colleagues, managers. There is often a need for a collaborative educational effect through visual, behavioral demonstration of referential reaction patterns. In this respect this activity is "inherently narcissistic", starting with the fact that for the overwhelming majority of actions the teacher is being watched by children.

It is also impossible to completely eliminate such a cause as shortcomings of the operationalization of the construct of narcissism in the Short Dark Triad (SD3) questionnaire, in which, for example, disagreement of a participant with the fact that he is “an average person” contributes to the level of narcissism (see Jones & Paulhus, 2014, p, 38). However, this disagreement can hardly be considered as an adequate correlate of extreme selfishness, self-love and self-admiration because it can be caused by other reasons: for example, very low self-esteem (a person considers himself a loser etc.).

At the same time, Kornilova and Chumakova (2016) come to the conclusion about the ambivalent role of narcissism after conducting empirical research using another popular questionnaire "The Dirty Dozen". Therefore, the problem may lie not in the shortcomings of specific questionnaire, but have a more general character.

Thus, the ambiguous role of narcissism is consistent with the findings of several other studies in which this personality trait also seems to be ambivalent. The results of our study show that, within the existing operationalization of the construct of narcissism, it might be beneficial leading to improvement of some performance indicators.