Symptoms of depression and the discrepancy between implicit and explicit self-esteem

https://doi.org/10.1016/j.jbtep.2018.12.001Get rights and content

Highlights

  • Investigate how damaged self-esteem is related to depressive symptoms.

  • Damaged self-esteem was positively related to the severity of depressive symptoms.

  • The link between explicit self-esteem and depression may be driving the finding.

  • Relative contributions of explicit and implicit self-esteem need further research.

Abstract

Background and objectives

While the role of explicit self-esteem (ESE) in depression has been well established, the relationship between implicit self-esteem (ISE) and depression is more uncertain. Recently, it has been suggested that a simultaneous consideration of both ESE and ISE may provide a more comprehensive understanding of the development and maintenance of depression than considering either ESE or ISE in isolation. The present paper tested whether the absolute discrepancy between ESE and ISE and the direction of the discrepancy are important factors to consider in relation to depressive symptoms.

Methods

87 university students from the northeastern U.S. were recruited for the present study. The Beck Depression Inventory, the Rosenberg Self-Esteem Scale, and Implicit Association Test were used to assess participants’ severity of depressive symptoms, explicit self-esteem, and implicit self-esteem, respectively.

Results

Results revealed a negative association between ESE and symptoms of depression. In addition, a positive relationship was found between the degree of discrepancy between ISE and ESE and depression. However, this relationship was only found among participants with higher ISE than ESE (i.e., damaged self-esteem), but not among participants with higher ESE than ISE (i.e., defensive or fragile self-esteem).

Conclusions

While damaged self-esteem may be a meaningful marker of depressive symptoms, it should be noted that the association between symptoms of depression and the interaction of ISE and ESE may be driven primarily by the strong link between ESE and depression. As prior studies of ESE and ISE have not looked, specifically, at the relative contributions of both constructs, this issue needs further investigation in future research.

Introduction

Lower self-esteem is associated with higher levels of depression in both adults and adolescents (e.g., Creemers, Engels, Prinstein, & Wiers, 2012; Ingram, Miranda, & Segal, 1998; Sowislo, Orth, & Meier, 2014). Moreover, a significant body of research has identified low self-esteem as a causal risk factor for the onset and maintenance of depression (vulnerability model; Beck, 1967; Sowislo & Orth, 2013). However, the overwhelming majority of studies have focused on the relationship between explicit self-esteem and depression.

Explicit self-esteem (ESE) is measured using a self-reported questionnaire, such as the Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965). However, explicit measures of self-esteem rely on individuals’ cognitive resources, ability to reflect on their self-worth, willingness to reveal this information. Therefore, it is possible that explicit measures may not be capable of fully evaluating important aspects of self-esteem. These limitations have led researchers to develop implicit measures of self-esteem. They are assumed to be accessing self-knowledge or representations that are introspectively inaccessible, preconscious, automatic, stable, and less susceptible to social desirability (Gawronski, LeBel, & Peters, 2007).

Despite the large literature linking ESE and depression, the research examining the link between implicit self-esteem (ISE) and depression is very limited and presents disparate results. While a few studies have shown a negative relationship between ISE and symptoms of depression (e.g., Risch et al., 2010), others have failed to replicate this effect (e.g., De Raedt, Schacht, Franck, & De Houwer, 2006; Franck, De Raedt, & De Houwer, 2007).

However, more recently an interesting line of research has emerged looking at ESE and ISE together, focusing on both the direction of difference (i.e., which assessment of self-esteem is greater) as well as its magnitude (Creemers, Engels, Prinstein, & Wiers, 2013; Creemers, Scholte, Engels, Prinstein, & Wiers, 2012; Leeuwis, Koot, Creemers, & van Lier, 2015). This seems to be presenting more converging results. When ESE and ISE do not agree, such discrepancy can take two different forms: higher ESE than ISE, where people are described to hold “defensive self-esteem,” and higher ISE than ESE, where people are believed to possess “damaged self-esteem.” A number of studies have demonstrated that, compared to individuals with congruent self-esteem, people with defensive self-esteem tend to display more self-enhancement tendencies, narcissism, defensive behaviors, in-group bias, dissonance reduction, and aggression (Bosson, Brown, & Zeigler-Hill, 2003; Jordan, Spencer, Zanna, Hoshino-Browne, & Correll, 2003; Sandstrom & Jordan, 2008; Zeigler-Hill, 2006). On the other hand, researchers have revealed that individuals with damaged self-esteem exhibit higher levels of anger suppression, maladaptive perfectionism, and symptoms of internalizing disorders (DeHart, Pelham, & Tennen, 2006; Franck, De Raedt, & De Houwer, 2008; Schroder-Abe, Rudolph, Wiesner, & Schutz, 2007).

Creemers et al. (2012) demonstrated that the absolute discrepancy between ESE and ISE was positively associated with depressive symptoms, suicidal ideation, and loneliness, only in participants with damaged self-esteem using the Name-Letter Task as a measure of ISE (NLT; Nuttin, 1985). Creemers, Scholte, Engels, Prinstein, and Wiers (2013) replicated their earlier findings using another measure of ISE, the self-esteem Implicit Association Test (self-esteem IAT; Greenwald & Farnham, 2000). While the replication of the findings using two different measures of ISE serves to illustrate the reliability of the unique association between damaged self-esteem and symptoms of depression, both of the previous two studies used the same sample that consisted solely of young adult women from the Netherlands. More recently, Leeuwis et al. (2015) conducted a longitudinal study and examined the influence of self-esteem discrepancies on the development of internalizing problems in Dutch children within the context of peer victimization. Their findings revealed that peer victimization was positively related only to damaged self-esteem at age 11 and showed that damaged self-esteem at age 11 predicted an increase in internalizing problems at age 12. Additionally, their study demonstrated that only damaged self-esteem mediated the relationship between the experience of victimization at age 11 and the development of internalizing problems at age 12. Thus, Leeuwis et al. (2015) not only replicated the findings of Creemers et al. (2012; 2013), but were also able to speak to a causal relationship where the interaction of ISE and ESE temporally precedes internalizing problems. In sum, these studies suggest that the discrepancy between ESE and ISE, more specifically where ISE is higher than ESE, may be an important factor of psychological problems, such as depression. Nevertheless, the few available studies examining the discrepancy between ESE and ISE were largely conducted by Creemers and his affiliates using European samples. This underscores the need for a replication experiment using a different sample.

Hypotheses are as follows:

  • 1.

    ESE will be negatively associated with symptoms of depression.

  • 2.

    The absolute discrepancy between ESE and ISE will be:

    • a.

      positively associated with symptoms of depression for participants with damaged self-esteem

    • b.

      not significantly associated with symptoms of depression for participants with defensive self-esteem.

Section snippets

Participants

A total of 87 undergraduates (67 female) at a private university in the northeastern U.S. participated in the study in exchange for course credit. Their mean age was 18.84 years (SD = 1.93, range 18–29). The sample was predominantly Caucasian (57%).

Severity of depressive symptoms

The Beck Depression Inventory, second edition (BDI-II; Beck, Steer, & Brown, 1996) was used to measure symptoms of depression. It is a self-report questionnaire consisting of 21 items. The BDI-II has well-established psychometric properties as a

Data Analyses

A hierarchical multiple regression analysis (HMR) was conducted to investigate how ESE and ISE as well as their interaction were related to the severity of depressive symptoms. As predictor variables, ESE and ISE were entered in step one and their interaction in step two. Subsequently, we performed a second HMR to examine the relationships among the discrepancy between ESE and ISE, the direction of the discrepancy, and their interaction with respect to the severity of depressive symptoms using

Results

Gender failed to moderate any results reported below, which are reported collapsing across genders. Descriptive statistics are reported in Table 1. The intercorrelations among the variables are presented in Table 2. Consistent with expectations, ESE was negatively correlated with symptoms of depressive and ISE was correlated with neither symptoms of depression, nor ESE. However, a significant positive correlation was observed between the discrepancy between ESE and ISE and symptoms of

Discussion

As expected, our results show a negative relationship between ESE and symptoms of depression. Our results replicate the findings by Creemers et al. (2012; 2013) and by Leeuwis et al. (2015) and demonstrate that damaged self-esteem was uniquely positively associated with depressive symptoms in a non-clinical college-aged population in the U.S. However, additional analyses suggest that the association between symptoms of depression and the interaction of ISE and ESE may be driven primarily by the

Conflicts of interest

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. There is no interest to be declared.

References (30)

  • D.M. Creemers et al.

    Damaged self-esteem is associated with internalizing problems

    Frontiers in Psychology

    (2013)
  • S.D. Farnham et al.

    Implicit self-esteem

  • M.B. First et al.

    Structured clinical interview for DSM-5 research version (SCID-5 for DSM-5, research version; SCID-5- RV, version 1.0.0)

    (2015)
  • E. Franck et al.

    Activation of latent self-schemas as a cognitive vulnerability factor for depression: The potential role of implicit self-esteem

    Cognition & Emotion

    (2008)
  • B. Gawronski et al.

    What do implicit measures tell us? Scrutinizing the validity of three common assumptions

    Perspectives on Psychological Science

    (2007)
  • Cited by (17)

    View all citing articles on Scopus
    View full text