Relationship between Self-Esteem and Personality Traits of Big Five and Dark Triad

This report evaluates the relationship between self-esteem and the personality traits of the big five and the Dark Triad. A sample size of 209 undergraduate students completed the series of measures online where Likert scale was used in the questions for them to individually do self-evaluation. IBM SPSS statistic software was used in the analysis to identify the connection between RSES and the personality traits. Results revealed the level of internal consistency of the used instrument in relation to measure of the RSES and the personality traits of the Dark Triad and the big five framework.

A study of the relationship between; self-esteem and personality traits of The Big Five and Dark
Triad.

Abstract
The purpose of this report was to evaluate the relationship between self-esteem and the
personality traits of the big five ad the Dark Triad. A sample size of 209 undergraduate students
completed the series of measures online where Likert scale was used in the questions for them to
individually do self-evaluation. IBM SPSS statistic software was used in the analysis to identify
the connection between RSES and the personality traits. Results revealed the level of internal
consistency of the used instrument in relation to measure of the RSES and the personality traits
of the Dark Triad and the big five framework.

Results
This part will comprise of the analysis results and the discussion of the results from the primary
data collected for the Rosebery Self-Esteem Scale (RSES) in the measurement of self-esteem.
Factor analysis
This is the analysis that group similar cases to form dimensions. Descriptive statistics for the
means of the 10 items were as in the below table;
Table 1: Descriptive Statistics
N Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error
means 209 1.90 9.00 5.2665 .86131 .228 .168 4.872 .335
Valid N (listwise) 209
The variable means was achieved by reducing the number of ten items to one by computing the
means of the variables of those who felt they were worthy people and at least on equal plane with
others (rse1), those with the feelings that they had a number of good qualities (rse2), those with
the general feelings of being failures (rse3), those with the feelings that they can do most of the
things other people do (rse4) up to (rse10) those with the feelings that at times they are not good
at all. The minimum achieved mean was 1.90 and the maximum of 9.00. The general mean for
all the ten items was 5.2665 and the standard deviation of 0.8613. The data values were slightly
skewed to the right as shown with skewness of (0.228).

Table 2: KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .910
Bartlett's Test of Sphericity
Approx. Chi-Square 1947.885
df 105
Sig. .000
The KMO and Bartlett’s test of sphericity was conducted to check for the analysis to be carried
out. As a result, the test significance was greater than the minimum 0.06 i.e. (0.910) which was a
superb value for go ahead to use factor analysis. The analysis that was used in the extraction of
factors was the principal components analysis. In conjunction with oblimin rotation, Eigen
values were used to determine the number of factors from the 10 items.
Table 3: Communalities
Initial Extraction
rse1 1.000 .839
rse2 1.000 .840
rse3 1.000 .750
rse4 1.000 .749
rse5 1.000 .621
rse6 1.000 .748
rse7 1.000 .748
rse8 1.000 .616
rse9 1.000 .733
rse10 1.000 .817
Extraction Method: Principal
Component Analysis.
The communalities table above show the representation of variables with the factors. Almost all
the variables were well represented in the common factor space since they had values (greater

than 0.5) as in the extraction column. This means the all the items were well represented and
suitably used in the research instrument and need not to be removed.
Table 4: Total Variance Explained
Component Initial Eigenvalues Extraction Sums of
Squared Loadings
Rotation Sums of Squared Loadings
Total % of
Variance
Cumulat
ive %
Total % of
Varianc
e
Cumulati
ve %
Total % of
Variance
Cumulative %
1 5.861 58.612 58.612 5.861 58.612 58.612 4.015 40.150 40.150
2 1.600 15.995 74.607 1.600 15.995 74.607 3.446 34.457 74.607
3 .592 5.920 80.527
4 .482 4.822 85.350
5 .437 4.368 89.718
6 .257 2.569 92.287
7 .224 2.237 94.524
8 .199 1.988 96.513
9 .181 1.807 98.320
10 .168 1.680 100.000
Extraction Method: Principal Component Analysis.
From fifteen variables used, there were 10 components with only two factors retained as in the
first two variables shown in the extraction sums of squared loadings in the above table. From the
total column, the first factor has the highest Eigenvalue since it accounts for the most variance
then followed by the second factor and the trend continues reducing until the least variance
accounted for by the last component (component 10) in this case. 58.61% of the variance are
accounted by the first factor, 15.995% by second component, and 1.81% by the third component
and so on in the decreasing sequence.

Figure 1
The scree plot was plotted for Eigenvalue against components from (RSES) test where from the
first component to the third component there was great change in the shape of the graph and the
graph seems to flatten after the third component showing the least variance accounted for by the
remaining components.

Reliability test using ten RSES items
Table 5: Reliability Statistics
Cronbach's Alpha Cronbach's Alpha
Based on
Standardized
Items
N of Items
.448 .452 10
For the nomological network of RSES, the Cronbach’s alpha was used to test for the reliability
of the ten items used in the online questionnaire. There was internal inconsistency since the
Cronbach’s alpha (0.448) was lower than the considered minimum (0.7) hence not fully reliable
as from the test.
Big five reliability analysis
Table 6: Reliability Statistics
Cronbach's
Alpha
Cronbach's
Alpha Based on
Standardized
Items
N of Items
.144 .257 5
Reliability of the instrument used was as well tested for the big five factors where general
Cronbach’s alpha of 0.144 was achieved which further confirmed that the online instrument was
not reliable for use in collection of data.
Table 7: Item Statistics
Mean Std. Deviation N
Extraversion 3.2608 .70030 209
Agreeableness 3.6699 .64365 209
Conscientiousness 3.4258 .51883 209
Neuroticism 2.9264 .74157 209
Openness 3.3258 .55902 209

The item statistics table standard deviation column show that standard deviation values are not
uniformly close to one another which further confirms that the used instrument was not reliable
and had internal inconsistency.
Discussion
Referring to the dimensionality, results from this report seem consistent with those in other
previous researches including those where RSES was tested and validated Li, Delvecchio, Di
Riso, Salcuni & Mazzeschi, (2015); Schaefer et al (2015). This is also in line with British and
China studies about Rosenberg Self-Esteem Scale and use of big five index to measure self-
esteem (Sariyska et al., 2014; Dufner et al., 2012).In the analysis full extraction of the correlation
of factors in relation to Rosenberg Self-Esteem Scale (RSES) factor analysis, the data was found
to be partially skewed to the right hand side as it was reflected by the skewness value (0.228).
The item wording effect made RSES a two dimension otherwise it was believed to be a one
dimension (Kong, Zhao & You, 2012). The psychometric evaluation of personality scale from
the factor analysis through principal component extracted two major factors where most of the
variance were accounted for. The choice of factor analysis was ended up with from the KMO
value (0.91) which was most suitable for the analysis to be conducted. All the items in the Likert
scale belonged to their respective scales as the extraction values were all greater than (0.5). The
Cronbach’s alpha from the reliability test of the instrument used was less than 0.5 i.e. (0.448).
This showed that the instrument had lower internal consistency thus could not be relied on. The
big five factors’ standard deviation also confirmed that the instrument was not reliable since the
standard deviation values were not too close or onto one another.

Limitations and future directions
From the results in the factor analysis, it is not quite easy to decide on the number of factors to
be included and used in the process. Further, the factors that were used could be difficult to
assume that they represent the data. The ten point scale in the data was not suitable as the
Rosenberg Self-Esteem Scale is best for four scale in the ten items. In the future therefore, the
appropriate scale should be considered and the ten items.
Conclusion
The construct validity and reliability test of the online instrument used was tested using the
Cronbach’s alpha. The principal component was used in the extraction of factors out of which
two factors were resulted with. RSES was found to be one of the suitable measure of self-esteem
since it is easy to understand. The big five framework i.e. openness, conscientiousness,
extraversion agreeableness and neuroticism were in this report included in the test of internal
consistency which agreed with the RSES ten items five point Likert scale.

References
Dufner, M., Denissen, J. J., Zalk, M., Matthes, B., Meeus, W. H., van Aken, M. A., & Sedikides,
C. (2012). Positive intelligence illusions: On the relation between intellectual self‐
enhancement and psychological adjustment. Journal of Personality, 80(3), 537-572.
Kong, F., Zhao, J., & You, X. (2012). Emotional intelligence and life satisfaction in Chinese
university students: The mediating role of self-esteem and social support. Personality
and Individual Differences, 53(8), 1039-1043.
Li, J. B., Delvecchio, E., Di Riso, D., Salcuni, S., & Mazzeschi, C. (2015). Self-esteem and its
association with depression among Chinese, Italian, and Costa Rican adolescents: A
cross-cultural study. Personality and Individual Differences, 82, 20-25.
Sariyska, R., Reuter, M., Bey, K., Sha, P., Li, M., Chen, Y. F., ... & Feldmann, M. (2014). Self-
esteem, personality and Internet addiction: a cross-cultural comparison study. Personality
and Individual Differences, 61, 28-33.
Schaefer, L. M., Burke, N. L., Thompson, J. K., Dedrick, R. F., Heinberg, L. J., Calogero, R. M.,
... & Anderson, D. A. (2015). Development and validation of the Sociocultural Attitudes
Towards Appearance Questionnaire-4 (SATAQ-4). Psychological Assessment, 27(1), 54.