Temperament and character as determinants of well-being

Abstract

Objectives

We aimed to evaluate the effects of Temperament and Character Inventory (TCI) personality measures on well-being scores in a sample of 49–51 year old New Zealanders. Previous research has linked high self-directedness (SD) and low harm avoidance (HA) with well-being. We hypothesised that SD and HA would have predictive power for Warwick-Edinburgh Mental Well-being Scale (WEMWBS) well-being. We anticipated that character profiles with high SD and cooperativeness (CO) would be associated with higher well-being scores while high self transcendence (ST) scores would have less of an influence on well-being in a secular population such as New Zealand. Additionally we aimed to describe and assess the performance of a well-being measure, the WEMWBS and we intended to clarify the factors that underlie the questionnaire.

Methods

A cohort of 404 randomly selected participants in a study of ageing had WEMWBS means calculated and with principal component analysis used to investigate the factors that underlie the WEMWBS. Multiple hierarchical regression was used to predict WEMWBS scores from socio-demographic and TCI personality variables. ANOVA were used to explore the non-linear effects of personality profiles on well-being.

Results

Principal components analysis revealed one significant factor which explained 54.34% of the variance. After initial univariate analysis current depression, marital status and standard of living were entered at the first step of the multivariate regression. They explained 20% of the variance. In step two the seven TCI personality variables were added to the model, explaining a total of 49% (R² change = 0.29). Low HA, high SD and the absence of current depression were the main predictors of WEMWBS scores. Character profiles featuring high SD produced significantly higher scores on the WEMWBS. Total WEMWBS scores were normally distributed, and the mean was 52.83 (95% CI 51.96–53.70).

Conclusions

The key determinants of well-being were absence of current depression and personality variables, especially low HA and high SD. The WEMWBS has face validity and the New Zealand results were very similar to the UK results.

Introduction

In recent years, the focus of psychology and psychiatry has shifted from being almost exclusively interested with mental ill health to an emphasis on mental well-being and positive psychology. Well-being has been shown to have a positive effect on survival and future health [1]. Measuring and identifying the factors that influence well-being are important to achieve better understanding of this concept and may facilitate interventions that aim to improve well-being in the individual. There are many scales and questionnaires that have been developed to measure well-being; a recent addition is the Warwick-Edinburgh Mental Well-being Scale (WEMWBS) [2]. The WEMWBS was developed in the UK to assess mental well-being in the general population. It is a positively worded 14 item questionnaire that aims to measure both aspects of well-being; eudaimonic and hedonic [3]. Eudaimonic well-being is described as positive psychological functioning, self-realisation, autonomy and positive relations with others. Hedonic well-being is more concerned with the subjective experience of happiness or, increased pleasure and decreased pain. The WEMWBS was chosen for use in this research study because it is designed to measure positive mental health and it is validated for general population surveys. Notably, it has been shown to have high consistency and reliability across different English speaking cultural groups [4] and in a Spanish sample [5].

Research has indicated that well-being may be influenced by socio-demographic variables, such as sex, age and socio-economic status [6], [7], [8]. It is also influenced by depression [6], [9] and by personality variables [8]. DeNeve and Cooper (1998) [8] suggest that specific personality traits may have a strong influence on well-being because these traits determine whether we perceive life and life events in a positive or negative manner.

Cloninger’s psychobiological model of personality [10] is measured by the Temperament and Character Inventory (TCI) [11]. TCI scores have been shown to be strongly related to well-being, for example the Young Finns study (n = 1980, age 24–39 years) revealed that TCI character domains explained 65% of the variance in well-being with self-directedness (SD) alone explaining 40% of the variance [12]. An investigation of TCI personality and a health questionnaire revealed that health scores were predicted by harm avoidance (HA) and SD, with low HA and high SD indicating better health [13]. In a UK study satisfaction with life scores were associated with high SD and low HA when studying personality profiles characterised by high or low scores on these two dimensions of personality [14]. Results from TCI and depression research show a strong association between depression, SD and HA. The extensive research in this area has found that high scores on the dimension HA and low scores on SD are strongly associated with depression [15], [16], [17], [18].

Cloninger proposes that well-being and health are not necessarily linear in their relationship with personality and that multidimensional personality profiles may reveal more about the relationship of well-being and personality through non-linear analyses [19]. Research using TCI profiles [11] has confirmed a link to well-being by showing that a combination of high scores on all three of the character dimensions of self-directedness, cooperativeness (CO) and self-transcendence (ST) [12], [19] is associated with well-being. However, ST increased negative and positive affect in one of the studies [12]. Both of these studies used character profiles that grouped participants depending on their high or low scores for the three character dimensions thereby creating eight specific combinations of character to capture non-linear as well as linear effects. The eight character profiles range from ‘creative’ (mature and frequently feels positive emotion) to ‘depressive’ (immature and frequently feels negative emotion) [20]. However, there are limitations with these studies. One of the studies [19] was conducted in Israel which is spiritually and culturally quite different from more secular countries such as New Zealand and therefore results may not be generalised. The other study used a random sample, ranging from 24–39 years of age, from the Finnish population [12] which is a relatively young age group. Given that personality changes over time [21], what is not known is how TCI personality is associated with well-being in a middle age sample in a secular society like New Zealand.

Using cross-sectional data from a longitudinal study of 49–51 years old, we aim to describe and assess the performance of the WEMWBS. Utilising principal component analysis, we will clarify the factor or factors that underlie the questionnaire and how much variability in the data is explained. The second aim is to assess how well the seven dimensions of the TCI can predict WEMWBS well-being, when relevant socio-demographic factors and current depression have been controlled for. We hypothesise that SD and HA will have predictive power for well-being as measured by the WEMWBS because of the previously observed link between SD and HA with depression and well-being. Finally, we will examine the non-linear effects of TCI personality on well-being by exploring the association between eight different TCI character profiles and WEMWBS scores. Personality profiles characterised by higher scores on SD and CO have been associated with higher well-being scores. We expect that character profiles with high SD and CO will be related with higher well-being scores, while high ST will have less of an influence on well-being in a secular population such as New Zealand.