Personality and Individual Differences 127 (2018) 107–113

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Big-Five personality and aspects of the self-concept: Variable- and person- T

centered approaches
Aleksandra Pilarska
Department of Personality Psychology, Institute of Psychology, Adam Mickiewicz University, Szamarzewskiego 89AB, 60-568 Poznań, Poland

A R T I C L E I N F O A B S T R A C T

Keywords: This study used variable- and person-oriented approaches to examine the importance of Big-Five personality in
Self-control predicting aspects of the self-concept (i.e., self-control, self-esteem, and self-feelings). The Mini-IPIP scales (IPIP-
Self-esteem BFM-20), Self-Control Scale (SCS), Rosenberg's Self Esteem Scale (SES), and Test of Self-Conscious Affect
Self-conscious emotions (TOSCA-3) were administered to 357 Polish students (59% female). The variable-centered approach, based on
Big-Five traits
multiple regression analysis, revealed that the personality traits explained 5 to 45% of the variance in the self-
Personality types
Person-centered
variables, with the largest effect found on self-control. Two-step cluster analysis yielded three personality types,
Variable-centered which corresponded to the previously described Resilient, Overcontrolled, and Undercontrolled types, and were
meaningfully distinguished on self-variables of interest. However, this type approach showed weaker predictions
than continuous and even dichotomized Big-Five traits.

1. Introduction existential anxiety, monitoring the degree of social inclusion-exclusion,

and promoting goal achievement (see review by Kernis, 2006). Low
According to McCrae and Costa's (2008) personality model, basic, self-esteem is associated with various emotional/behavioral problems,
biologically-based tendencies, such as the Big-Five traits, are expressed including depression, aggression, and loneliness (e.g., Donnellan,
as characteristic adaptations, which represent all acquired psycholo- Trzesniewski, Robins, Moffitt, & Caspi, 2005; Leary, Schreindorfer, &
gical attributes that are, to some extent, shaped by contingencies in Haupt, 1995).
psychosocial contexts. In this model, basic tendencies and characteristic Self-conscious emotions constitute an important class of emotions
adaptations represent, respectively, the distal (indirect) and proximal that are theorized to be critically involved in social behavior regulation
(direct) determinants of behaviors and experiences. A particularly im- (Tangney & Dearing, 2002; Tracy & Robins, 2004). Prototypical self-
portant subset of characteristic adaptations is the self-concept. It con- conscious emotions include guilt, shame, and pride. Both shame and
sists of knowledge, views, and evaluations of the self, through which guilt arise from a perceived wrongdoing; guilt, however, is concerned
people understand themselves. The self-concept occupies a central po- with a negative evaluation of a specific act, while shame pertains to a
sition in many personality theories and has been associated with a wide negative evaluation of the whole self. Guilt-proneness has been linked
range of human actions and modes of response. The prominence of the with reparative and prosocial behaviors such as empathy, altruism, and
self-concept is amply demonstrated in psychotherapy, which is often caregiving. Shame-proneness, in contrast, has been found to be related
explained in terms of reconstruction of the self (e.g., Rogers, 1951). to social-psychological maladjustment (e.g., social withdrawal;
However, to the extent the self-concept is shaped by stable personality Tangney & Dearing, 2002). Pride is a positive emotion that promotes
traits, it cannot be expected to be affected by therapeutic interventions. sense of achievement and self-satisfaction. Some researchers distinguish
This investigation was designed to study the associations between these between pride in self (alpha/hubristic) and pride in behavior (beta/
two layers of personality. More specifically, it focused on how core authentic), with the latter being shown to lead to more positive out-
personality traits, i.e., extraversion, agreeableness, conscientiousness, comes than the former (Tangney & Dearing, 2002; Tracy & Robins,
emotional stability (reversed neuroticism), and openness, explain dif- 2007).
ferences in self-esteem, self-conscious emotions (or self-feelings), and Self-control refers to the executive aspect of the self. Defined as the
self-control. capacity to override one's responses (thoughts, feelings, impulses, and
Self-esteem and self-feelings both refer to the evaluative aspect of behaviors), self-control is posited as crucial for achieving personally
the self. Self-esteem is typically defined as one's attitude of personal valued goals (Baumeister, Vohs, & Tice, 2007). Consistent with this
worth (Rosenberg, 1965). It serves such vital functions as buffering view, research has linked self-control to many positive outcomes,

E-mail address: [email protected].

https://doi.org/10.1016/j.paid.2018.01.049
Received 17 July 2017; Received in revised form 30 January 2018; Accepted 31 January 2018
0191-8869/ © 2018 Elsevier Ltd. All rights reserved.
A. Pilarska Personality and Individual Differences 127 (2018) 107–113

including personal adjustment, academic performance, and social re- Overcontrolled, and Undercontrolled would emerge; (3) the clusters
lationships (De Ridder, Lensvelt-Mulders, Finkenauer, Stok, & would differ with respect to self-variables, with Undercontrollers re-
Baumeister, 2012). Poor self-control results in deteriorated perfor- porting lowest self-control, Resilients scoring highest on self-esteem,
mance, and has immense personal and societal repercussions as diverse guilt-proneness, and beta-pride, and Overcontrollers reporting highest
as procrastination, depression, obesity, violent crime, and drug abuse shame-proneness, and (4) the Big-Five traits would show a higher
(Tangney, Baumeister, & Boone, 2004). predictive power than the Big-Five types.
To date, most research has taken the variable-centered approach
(e.g., regression; structural equation modeling) to address the relation
between the Big-Five personality and self-variables. Several studies 2. Method
have reported the Big-Five correlates of self-esteem (e.g., Donnellan,
Oswald, Baird, & Lucas, 2006; Robins, Hendin, & Trzesniewski, 2001), 2.1. Participants and procedure
generally showing that self-esteem had positive associations with all
five personality dimensions, especially the two having a clear affective Participants were a convenient sample of 357 Poznan (Poland)
component, namely emotional stability and extraversion. Self-control university students (59% female), majoring in different academic dis-
has previously been found to correlate strongly and positively with ciplines (29% in professions and applied sciences, 27% in social sci-
conscientiousness, and, to lesser degrees, with emotional stability and ences, 14% in humanities, 10% in natural and formal sciences, and 20%
agreeableness (Marcus, 2003; Tangney et al., 2004). The scarce avail- in interdisciplinary academic areas). Participants' mean age was
able research has linked authentic pride to socially desirable and gen- 21.19 years (SD = 1.88, range = 18–31). Questionnaires were ad-
erally adaptive Big-Five traits (especially extraversion and emotional ministered, in a counterbalanced order, in classrooms during academic
stability), whereas hubristic pride has been negatively related to the class hours by trained research staff. Participation was voluntary, and
two prosocial traits of agreeableness and conscientiousness (Tracy & anonymity and confidentiality were guaranteed. No financial incentives
Robins, 2007). The only consistent finding concerning guilt and shame were offered.
has been that both correlate negatively with emotional stability (Abe, Based on the most complex analysis planned, the sample size was
2004; Einstein & Lanning, 1998; Harder & Greenwald, 1999). determined sufficient to detect a small effect size of f2 = 0.05, with 80%
This study aimed to extend previous research by using both vari- power and alpha set at 0.05.
able- and person-centered approaches. It not only focused on separate
traits in a nomothetic way, but also considered how individuals'
standings on each of the Big-Five traits might shape their self-concepts. 2.2. Measures
The person-centered approach has recently attracted considerable in-
terest in Big-Five personality research. Using Q-factor or cluster ana- The Big-Five traits (i.e., extraversion, agreeableness, con-
lyses, three personality types (known as RUO types) have most con- scientiousness, emotional stability, intellect) were measured by
sistently been identified: Resilient (i.e., well-adjusted), Undercontrolled Donnellan et al.'s (2006) Mini-IPIP scales (IPIP-BFM-20; adapted by
(i.e., dysregulated), and Overcontrolled (i.e., constricted; Asendorpf, Topolewska, Skimina, Strus, Cieciuch, & Rowiński, 2014).1 The in-
Borkenau, Ostendorf, & van Aken, 2001; Robins, John, Caspi, Moffitt, & strument consists of 20 items rated on a 5-point scale (1 = very in-
Stouthamer-Loeber, 1996). However, only two studies have reported on accurate, 5 = very accurate), with higher scores indicating that the trait
the effect of personality type on self-variables of interest (to be exact, describes the individual better.2
self-esteem). Both have found Undercontrollers and Overcontrollers to To assess self-control, the Self-Control Scale (SCS) developed by
have lower self-esteem than Resilients (Asendorpf et al., 2001; Tangney et al. (2004; adapted by Pilarska & Baumeister, in press) was
Pulkkinen, Männikkö, & Nurmi, 2000). Considering that the RUO ty- employed. It consists of 36 5-point scale items ranging from 1 = not at
pology refers back to Block and Block's (1980) proposal, which focused all to 5 = very much. These items pertain to control over thoughts,
on ego-resiliency and ego-control, it seems reasonable to expect that the emotions, impulses, performance, and habit-breaking, and yield a single
personality types would be distinguished by their self-regulation ten- total score, with higher values indicating higher self-control.
dencies and capabilities, as reflected by self-control and self-feelings. Self-esteem was evaluated using the Rosenberg's Self Esteem Scale
This study, thereby, could provide evidence on the suitability of Block (SES; Rosenberg, 1965; adapted by Łaguna, Lachowicz-Tabaczek, &
and Block's model as a reference framework for interpreting the Big- Dzwonkowska, 2007). The SES is a 10-item scale in a 4-point format
Five types. (1 = strongly agree, 4 = strongly disagree), with higher scores in-
The point is also worth noting that despite the advantage of pre- dicating greater self-esteem.
serving information on individuals' personality structure, the type ap- Self-conscious emotions were assessed with Tangney, Dearing,
proach suffers from the disadvantage of losing information on inter- Wagner, and Gramzow's (2000) Test of Self-Conscious Affect (TOSCA-3;
individual within-type variation, which makes its predictive power adapted by Adamczyk & Sobolewski, 2014). The measure uses 16 sce-
questionable. Several studies have evaluated the extent to which this narios followed by responses indicating shame-proneness, guilt-prone-
approach can compete with the variable-centered approach, and most ness, alpha-pride, beta-pride, and defenses such as externalization (i.e.,
of them found the dimension prediction outperforming the type pre- blaming others) and detachment (i.e., minimizing problems or emo-
diction (e.g., Asendorpf, 2003; Costa, Herbst, McCrae, Samuels, & Ozer, tionally distancing oneself). Responses are rated on a 5-point scale,
2002; Roth & von Collani, 2007). However, because the outcome of ranging from 1 (not likely) to 5 (very likely), with higher scores in-
such head-to-head comparisons may depend on different factors, like dicating a greater proneness to that reaction.
number and intercorrelations of the predictors, type of the criterion
variable (dimensional or type), study design (cross-sectional or long-
itudinal), a definite conclusion about the relative predictive power of
types versus traits awaits further investigation. 1
The IPIP-BFM-20 measures the five basic traits as identified in the lexical approach.
Based on the literature review and research objectives, the following There is sufficient overlap between the lexically- and psychometrically-derived models to
hypotheses were proposed (1) self-variables would be predictable from assume intellect and Costa and McCrae's openness refer to the same personality domain.
the Big-Five traits, with self-control being predicted mainly by con- Moreover, the items on the IPIP Intellect and the NEO-PI Openness scales possess similar
content.
scientiousness, self-esteem, pride, and shame-proneness being related 2
A person-mean substitution was used to replace missing values for participants
most strongly to emotional stability, and guilt-proneness being pre- missing up to 20% of a (sub)scale's items. The (sub)scale was unscored for those missing
dicted mainly by agreeableness; (2) the clusters for Resilient, more items.

108