1The sociological research of surveys mostly compiles data on a multitude of variables related to attitudes and behavioural dispositions (less on behaviour itself) and a standard amount of socio-demographic data such as gender, age, education, etc. These data then serve to describe the relationships between the variables. It is also claimed that these can be interpreted as causal relationships. It is a fact, however, that in most cases, only a relatively small percentage of the variance in the dependent attitudes and behavioural patterns is explained (usually between 10 and 20%, one third at the high end). The larger proportion remains "unexplained"; they can be designated as coincidental variations, as Jencks (1973) did when he ascribed the unexplained proportions in income to the factor "chance". Based on the high significance and variability of external factors, this is not completely unjustified. On the other hand, one must admit that important variable complexes are not captured in sociological studies, irrespective of other limitations in their methodological design (most deal only with "snapshots" or surveys taken at a single point in time).
2Characteristics of personality and those of identity represent one such complex. The first are well-known as central areas of psychological research. The relationship between these variables and attitudes or behaviour seems thoroughly plausible, yet it is not adequately captured in either psychological or sociological studies. In the first, only a limited spectrum of social characteristics, attitudes and behavioural patters are observed and the examined samples are small and non-representative as a rule. Furthermore, personality dimensions are polled with very many items (often more than one hundred) in psychological studies, although recently efforts have been made to develop and validate short inventories to measure the "Big Five" (see Gosling et al. 2003; Rammstedt et al. 2004; Rammstedt & John 2006). In sociological studies and especially in survey research, the absence of personality and identity characteristics can be traced back to two causes. Firstly, they generally do not have the space or time for such comprehensive batteries of items, which means they simply have no viable instrument at their disposal to collect data on personality characteristics. Secondly, the area of personality research is seen as the domain of psychology. Interdisciplinary exchange is completely missing here. And this means a great scientific potential is also being missed as the different disciplines are often occupied with the same research questions, yet observe the topics from different perspectives. The combining of these different perspectives could contribute a great deal toward new insights.
3Why are characteristics of personality and identity also of great interest to the social sciences? The personality of an individual distinguishes him/her from others in how he/she thinks, feels and acts, allowing him or her to appear unique. At the same time, personality characteristics determine behaviour, attitudes and emotions to a great degree (see Pervin 2000: 24ff.). The different reactions of people to the same situations can be traced back to these personality characteristics. For example, one can study why some people have more social contacts, are more satisfied than others with their circumstances, or have a generally more optimistic view of life. Socio-demographic characteristics contribute to explanations here without doubt. It must be observed, however, that individuals belonging to the same social conditions can exhibit the most diverse personality characteristics. On the other hand, it is an interesting question why people disposing of similar incomes, professional status and educational level vary in their degree of happiness and satisfaction.
4The general intention of this contribution is to introduce several selected instruments to capture personality and identity, to discuss them and demonstrate how they can contribute to explain social attitudes. It deals with two questions in particular: the first is the delineation and theoretical classification of the two concepts identity and personality. The second question is about their contribution toward an explanation of life satisfaction.
5Psychology often uses the concept of the "self" rather than "identity" (Oerter et al. 1987: 297). One understands this to be "the unique personality structure in combination with the view that others have of this personality structure" (Oerter/ Montada 2002: 291, our translation). One's own perception of self should be added to this definition, namely the self-concept or self-schema. The self-concept can be divided into a cognitive and an affective component. The cognitive elements include knowledge of one's self and self-perception. The affective components refer to feelings of self-worth or inferiority and self-confidence.
- 2 For a detailed account of the development of the concept of identity see Weigert et al. 1986; Keup(...)
6Social psychologists and sociologists tend to use the concept of identity more frequently. This can be attributed to a long research tradition.2 In the last few decades this tradition has experienced a boom due to the work of the German-American psychoanalyst Erik H. Erikson (1980). The beginnings of identity research can be found already in the ancient world (De Levita 1976: 23) and extend through the early modern philosophers (Locke, Leibnitz, Hume) up to William James awakened an interest in this research from a socio-psychological and sociological perspective in his work "The Principles of Psychology" (1890). He divided the concept of the self into constituents including an empirical Self (Me) and a pure ego (I) – concepts which were then taken up by George Herbert Mead. Mead (1980: 242) sees the "I" as the spontaneous and creative element which makes our free will possible. In contrast, the "Me" represents an internalized conception of how we are seen through the eyes of others and of the expectations others have of us. All of the various Me's together form the social identity. Mead, who saw himself as a social psychologist, also uses the term "Social Self". The concept of social identity has also been used by Erving Goffman (1975: 73f.), relating to the different identity roles (mother, employee, customer, etc.) that a person takes on in his/her life. Personal identity, on the other hand, refers to the uniqueness of an individual that emerges on the basis of identity tags (name, appearance, characteristics) and the specific, individual biography.
7The sociological identity research, the socio-psychological self-concept research and the psychological personality research proceed mostly without reference to one another, despite the fact that they deal with closely related research questions (see Oerter/Montada 2002: 382). Mischel and Morf (2003: 22) refer to a natural connection between Self and personality. A separation of these two concepts is not to be recommended; integrating them should be the goal instead. To do so, the question must be investigated to what degree an individual's personality and his/her identity are interconnected. Answering this question has been hindered by the isolation of research of different (sub-) disciplines and the resulting conceptual heterogeneity.
8From the sociological perspective which forms the basis of this article, an integration of the concept of personality and the theory of identity seems to be worth pursuing. The following section develops a few considerations in this direction.
9We begin with the assumption that the overarching “Self” of a person includes – among other components – a social and a personal identity. The personal identity allows an individual to appear unique through a specific combination of personality characteristics, abilities, interests, physical attributes and biography. Personality and personal identity are bio-socially determined, which means that both are genetically pre-defined and socially reshaped. In connection to this, one can speak of a dialectic between nature and society (see Berger & Luckmann 1974: 192).
10How are characteristics of personality and identity determined and how can they be captured empirically? Personality characteristics can partially be traced back to genetic predispositions. The continuing controversy over the nature-nurture problem expresses the discord among scientists with reference to the degree of inheritance versus socialization. Pervin (2000: 30) attributes the major role in determining personality to genetic factors. Admittedly, this influence varies among different characteristics. Intelligence and temperament above all are strongly codetermined by genetic factors. On the other hand, sociologically relevant aspects such as values, ideals and convictions can hardly be traced back directly to a person's genetic make-up. These aspects will therefore not be counted here as personality characteristics in the narrow sense, but as parts of the concept of the Self. We conceive of the Self as a referential and ordering scheme, which strives for the resolution of intrapsychological divergences and conflicts between social and personal identity; it also supplies an orientation with regard to actions and attitudes (see also Haller 2003: 598f.). The great significance of the identity concept in explaining attitudes and behaviour arises out of this definition.
11Personality characteristics are a part of personal identity and therefore constitute a major component of identity as a whole. In principle, one can discern a variety of personality characteristics; more recent psychological research has shown, however, that these can ultimately be reduced to a limited number. In personality psychology today, the Big Five approach has been widely accepted (Oerter/Montada 2002:383). Despite its frequent application, capturing personality with a "five factor model" is not completely uncontroversial. Objections have been made with regard to the number of necessary descriptive dimensions. The question has been asked whether personality can be most adequately described using four, five, six or more dimensions. A further one is that important personality characteristics exist which can not be assigned to any of the five factors. This problem indicates a general dilemma of personality tests, which should capture personality as completely as possible, yet also be as efficient as possible in their application. With the "Big Five", a model of personality is available that sufficiently fulfils both requirements of empirical research practice with justifiable omissions. For this reason, the approach has been chosen for this contribution and will be dealt with in more detail in the following section.
12The Big Five approach has its origins in two different traditions of factor-analytic personality research – in the psycho-lexical approach and in the differential, clinical personality research (Lang & Lüdtke 2005: 18). Within the scope of relevant psychological research today, the five factor model is seen by many as the most suitable instrument for describing personality (Oerter/Montada 2002: 383).
13The basic assumption of this model is that there is a limited number of fundamental personality traits and that each person is characterized by a specific bundle of these. These characteristics are Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to new experiences. (A more detailed description follows in the next section.) They exhibit a relatively high stability over the life course. It is assumed that all significant inter-individual differences between people are present and anchored in language (sedimentation hypothesis). In terms of empirical method, this personality model is based on a linguistic-lexical approach, whereby the complete lexicon of a language is first examined for all of the adjectives it contains that serve to characterize people. The underlying "lexical hypothesis" states that all aspects of human personality significant for day-to-day behaviour can be found in the vocabulary of a language (see De Raad 2000: 16). Allport and Odpert (1936) searched through English dictionaries for words that describe individual dispositions and came up with a list of 17,953 words. Norman (1967) reduced this list step by step and systematically to 2800 concepts. Goldberg (1990) was able to compile 339 adjectives into 100 groups. After administering different tests of self-assessment and assessment of others with regard to these adjectives, a factor analysis was carried out, leading to five extracted factors. Through this process, the multitude of adjectives in day-to-day language could be reduced to only five character dimensions (see Asendorpf 1999: 127ff.). Borkenau and Ostendorf (1993) adapted the NEO Five Factor Inventory for the German-speaking world. Their version measures the "Big Five" with 60 items; the original English language version originated from Costa & McCrae (1989). Out of the Big Five Inventory with 44 items from John and Srivastava (1999) and in the style of the abbreviated scales from Gosling et al (2003), the "Short Big Five Inventory" was developed. This consists of only 15 items (see Lang & Lüdtke 2005).
14Three aspects distinguish the concept of identity from that of "personality characteristics": (1) The various facets of identity are not isolated from one another, but form a more or less integrated whole: (2) cognitive aspects, such as a person's self-awareness, target preferences and values, play a central role, whereby the emotional aspect (self-respect, pride or shame) is in no way disregarded; (3) this concept makes it possible to produce a more conclusive relation between individual characteristics and social environment. Therefore, a basic assumption of identity theory is that every person strives to create a positive picture of him/herself and in his fellow human beings in order to develop self-respect (see Scheff 1990).
15In this article, the relevance of the concepts of identity and personality is examined with regard to one specific attitude, namely life satisfaction. With reference to the Big Five personality characteristics, several studies (Diener et al. 2003: 406; Schimmack et al. 2004: 1062) show that the characteristics extraversion and neuroticism have a significant effect on life satisfaction. Moreover, the five factor model is widely applied to other questions in various sub-disciplines. Connections have been ascertained of the "Big Five" with voting behaviour, trust in institutions, extremism on the right, and religious conviction (Schumann 2005; Kunz 2005).
16The "Big Five" and characteristics of identity are assumed to be important predictors of general life satisfaction and this connection is worth examining (see Section 3). Life satisfaction represents a subjective comprehensive evaluation of a person's situation in life, which is derived from individual satisfactions in specific areas of life such as family, work, leisure time, etc. It is left to the interview subjects themselves to determine which aspects they view as particularly relevant and draw on in the evaluation of their personal situation. Life satisfaction is understood as the cognitive component of subjective well-being (Schimmack et al 2004: 1063). Older studies concentrated mainly on social characteristics (demographic variables) to explain life satisfaction, but these were able to clarify only a modest portion of the variance. Newer research results, however, substantiate a stronger effect of "inner factors", as they represent personality characteristics (Diener et al 2003: 406f). Our hypothesis is that extroverted, compatible, conscientious, and open people are happier, whereas introverted, mistrustful, and neglectful people who are not open to new experiences tend to be unhappier. From various studies, we know that sociability and social contacts just as trust in fellow human beings increase happiness in life (Haller/Hadler 2006; Argyle 1987; Bradburn/Caplovitz 1965). The fear of speaking in front of other people is a negative experience of a great part of the population; a considerable portion suffers from social phobias which limit the willingness to have contact with other people (André / Légeron 1995). Furthermore, it can be assumed that purposefulness and endurance as well as intellectual curiosity and an openness to the world lead to more enriching life experiences and therefore probably also to success. In this way they also increase life satisfaction.
17We begin with the assumption that, along with the "Big Five", primarily aspects of social identity and the self form the attitudes of a person to a decisive extent. The degree of an individual's subjective identification with a social group or institution influences his/her judgments, opinions, decisions and ultimately his/her actions. This thesis, that identities have a motivational effect on behaviour due to a need for presentation, is also a basic assumption of Structural Symbolic Interactionism. It theorizes that one fosters expectations concerning the behaviour of other people, and that one classifies oneself, which leads to internalized expectations of one's own actions (see Stryker 1976: 268f.; Burke et al. 2003: 2f.). Each person possesses several social identities (family, profession, circles of friends, community, country, etc.). Depending on their subjective importance, these are organized in a hierarchy of significance which mirrors the probability of a certain identity being activated in a specific situation. Identities that stand at the highest position in this order influence attitudes and modes of behaviour more strongly than those that are considered less important. In conjunction with this, we posit three hypotheses: (1) Those people will be happier who identify strongly with the areas of life that are important for every person – family and occupation. It is known from many studies that these areas are central for a successful identity. (2) A positive connection should also exist between life satisfaction and the degree of identification with gender and age roles. We base this thesis on the fact that people who are dissatisfied with these roles (e.g. would like to be younger than they are in reality) tend to be dissatisfied in general. (3) With reference to further aspects of the social identity (political and religious persuasions, ethnic origins, national affiliations, etc.), we expect no effect on life satisfaction, as these are not central for the identity of a person as a rule.
18Along with the social identity, the moral framework of the Self has a constitutive impact on attitudes and modes of behaviour. This framework forms the background for our moral judgments and confers meaning on actions (Taylor 1994). A successfully formed identity also implies the ability to find meaning in life (Frankl 1978). Our hypothesis is that people who are able to find a sense in their lives are more satisfied and happier than those who are not able to do so.
19The basis of the following analysis is the study "National Identity, Citizenship and the State", which was carried out as part of the International Social Survey Programme (ISSP) 20043. Throughout Austria, 1006 people were surveyed using a standardized questionnaire.4 The surveying was conducted by the Institute for Empirical Social Research (IFES) in Vienna. The random sampling consisted of adults 18 years or older. Elderly people of 60 years or more constitute 29% of the sample, 46-59 year olds 22%, 30-45 year olds 33%, and young adults up to 29 years old 16%. Somewhat more women (53%) than men (47%) agreed to an interview. 44% of those interviewed work fulltime, 8% part-time, and 3% less than part-time. Due to the large percentage of interviewees over 60 years old, 25% are already in retirement. Only 4% of the people were in school or training at the time of the interview. The remaining interviewees consisted of housewives/-husbands, unemployed people, or persons who are unable to work. In regard to education, the sampling is also balanced: 20% had completed compulsory education, 36% learned a trade, 15% had graduated from a vocational school, 19% had completed high school, and 10% had graduated from college / university. The interviewees are spread over all nine Austrian provinces and over communities of all sizes.
- 3 See http://www.issp.org/.(...)
- 4 Type of interview: oral, personal CAPI interviews at the place of residence. "Computer-Aided(...)
20The following section describes the instruments for capturing personality. The first instrument draws from the five central characteristics of personality as they were developed within the field of differential personality psychology. The further instruments were also worked out within the framework of personality psychology; they can also be seen, however, as means of operationalising the construct "Identity".
21The Big-Five personality characteristics consist of the dimensions "Extraversion", "Agreeableness", "Conscientiousness", "Neuroticism", and "Openness". On the positive side, characteristics such as sociability, activity, talkativeness and perseverance belong to Extraversion, and on the negative side, characteristics such as reclusiveness and taciturnity to Introversion. Agreeableness means interpersonal behaviour characterized by closeness, trust and cooperation – or alternately by distance and distrust. Conscientiousness differentiates reliable, goal-oriented and persistent people from careless, apathetic, and inconsistent ones. Neuroticism describes the emotional situation of the interviewees which can be either carefree and relaxed or fearful, nervous, depressed and insecure. The last dimension concerns the degree of Openness to new experiences, which can span from curiosity and inquisitiveness to traditionalism and conservatism (see Goldberg 1990; Rammstedt et al 2004; De Raad 2000; Pervin 2000; Asendorpf 1999).
22Rammstedt and John (2006) developed a greatly reduced instrument for economically measuring these five fundamental personality characteristics in population surveys. The 10-item short version of the Big Five Inventory (BFI-10) allows each dimension of the "Big Five" personality characteristics to be reliably and validly measured with only two items. Only the personality characteristic Agreeableness suffers in validity through the reduced number of items. Therefore, the authors suggest to capture this dimension through an additional item (and so 3 items in all are included here). In our empirical analysis, this scale also exhibits the lowest values in reliability and validity.
- 5 The exact formulation of the question was: "We now have some statements about how one judges(...)
23The questionnaire of the ISSP study "National Identity and Citizenship" contains an item battery developed by Rammstedt and John (2006) with ten items to capture the Big-Five personality characteristics (BFI-10). The interviewees could answer to a five stage Likert scale (from "completely true" and "true on the whole" to "completely untrue" and with an additional category "I can't say").5 In the analysis the last answer category was disregarded due to the very small number of entries. The response rate of all items together was nearly 100%. In order to test the scale's quality in capturing the five personality dimensions, a factor analysis was first performed. The Kaiser-Mayer-Ohlin criteria resulted in a low value of 0.58, which leads to the conclusion of a minimal suitability of the chosen analysis procedure for summarizing the variables or weak consistency of the respective item pairs in capturing personality dimensions.
Table 1: Factor analysis of the Big Five personality characteristics (Factors with eigenvalues over 1, factor loadings)
Notes: Extraction method - Major component analysis. Rotation method - Varimax with Kaiser-Normalisation. The items were grouped according to content affiliation. Factor loading over .30 are shaded gray. The fields in which a high factor loading was expected due to content considerations have bold outlines. Sampling: Representative population survey in Austria; N = 1006.
Source: ISSP Study 2004, Institute of Sociology at the University of Graz.
24The five extracted components with an intrinsic value above 1 together account for 64.9% of the total variance. However, the items pertaining to certain factors mostly do not match up in terms of content. Also, several variables can not be clearly attributed to a single dimension.
- 6 The first German sampling consisted of 457 students, the second of 376 students; the first two Ame(...)
25There is a similar dynamic in the formation of personality scales. Here it can be seen that the items for the Big Five personality characteristics also do not result in a good scale. Table 2 shows that the convergent validity is mostly very low, which means that only a weak correlation exists between the two items of a scale. Admittedly, the items of Rammstedt and John (2006) were intentionally selected heterogeneously and therefore they do not utilize the Cronbach Alpha measure of reliability. Notwithstanding, our empirical data (in particular the factor analysis) reveal that the formation of a scale of personality characteristics is possible only to a limited degree. In comparison to the study of Rammstedt and John (2006), this is a rather disappointing finding. But it must be noted that Rammstedt and John conducted a written survey whereas our data resulted from personal face-to-face interviews. These two different methodological approaches could be relevant for the results. It is possible that the interaction with the interviewer evokes distortions in the answers due to social acceptability or to the intention to present oneself in a positive light (see Költringer 1993: 15). Another reason why the Big Five can not be effectively reproduced in the Austrian study using the BFI-10 might be traced back to the different samplings. Whereas the majority of samplings in the study of Rammstedt and John (2006) consisted of students6, the interviewees in the ISSP study comprise a representative cross-section of the population.
26Why are the ten items unsuitable for reproducing the Big Five personality dimensions? One could surmise that the overall number of items is too low and that only two items are not sufficient to accurately capture the content of a dimension, and therefore they do not suffice to guarantee a high reliability of the resulting index. Notwithstanding, the selection of items by Rammstedt and John (2006) was made according to the extent they could be viewed as prototypical for each dimension. Due to the acceptable reliability and validity of the two scales Extraversion and Neuroticism, these two were selected for the ensuing multivariate analysis. Furthermore, other studies have shown that these personality characteristics have the strongest influence on life satisfaction (Diener et al. 2003: 406; Schimmack et al. 2004: 1062).
Table 2: Mean values, Standard deviance, Skewing, Cronbach’s Alpha and Correlations of the Big-Five Short Version
Note: Each scale can have values from 2 to 10. For example, a value of two by "Extraversion" means that the concerned person is very extroverted and sociable and likes to approach other people; a value of 10 means that he/she is socially withdrawn. Sampling: representative population survey in Austria; N = 1006
Source: ISSP Study 2004, Institute of Sociology at the University of Graz.
27All of the empirical mean values of the scales fell fairly close to the arithmetic mean of their scores. One exception is the value "Conscientiousness" which lies clearly under five (3.71) and also demonstrates the highest value in terms of skewing. This means that the interviewed people reported far less frequently that they are laid back and tend to be lazy than that they are thorough. This result can certainly be traced back to the strongly evaluative component of this item and the social acceptability of the answers.
28When persons fill out a questionaire about their personality characteristics they are confronted with declaring something very "personal". In the context of positive self-representation, everyone probably makes an effort to reveal him/herself in a good light and to "overlook" unfavourable qualities. It can also happen, however, that interviewees intentionally give seemingly negative answers out of a certain protest attitude against the study or a certain question. For this reason, it would be an advantage to also empirically measure or examine the degree of social acceptability of the items. In the answer tendencies of the studied subjects one can already draw conclusions about their personality characteristics. According to some studies, people who place a high value on social acceptability have a lower self-esteem and self-confidence as comparable subjects (see Mummendey 1995: 159ff.).
29A further battery of items in the ISSP survey “National Identity and Citizenship” contains questions to capture aspects of self-conception and attitudes toward life. Five dimensions of the self were collected; we do not claim, however, that these dimensions represent an exact operationalisation of the concept of identity. The dimensions are: private and public self-awareness, self-esteem, emotional self-concept and the feeling about the meaning of life. In the survey, 1006 interviewees were given 17 statements and asked if they applied or didn't apply to themselves according to a five stage Likert scale. On the basis of a pretest, we could assume that latent factors lie beneath the individual variables which allow for the formation of an identity scale. Taking also into consideration the factor loading and internal consistency, the scales were abbreviated in order to obtain an economical and efficient measuring instrument. Most of the items were taken from the handbook "Skalen zur Befindlichkeit and Persönlichkeit" (scales for mental state and personality) by Schwarzer et al. (1986). In the following, the scales are explained in more detail.
30"Private self-awareness" (Schwarzer et al. 1986: 3ff.) was originally measured in a survey with 14 items, which were reduced to two in our study. This scale captures dispositions regarding personal recognition of one's self. People who have a high degree of private self-awareness can reflect well about themselves and therefore know their own attitudes, emotions and behaviour relatively well.
31The "public self-awareness" (Schwarzer et al. 1986: 3ff.) was also elucidated with 15 items which were again reduced to two. This scale comprises items that capture a person's outward perception. Standing in the foreground is the impression one makes on others; great efforts are made to present oneself in a good light. The scale was tested in three samples in Berlin that were made up of 210 adults, 353 students and 450 youths in school or training.
32The "general self-efficacy" (see Schwarzer et al. 1986: 15ff.) is composed of the two scales of school-specific and study-specific self-efficacy. The first refers to the general disposition toward one's own actions in situations involving particular demands. The subjective cognition of the efficiency of one's own performance reveals one's stance toward certain demands of life and how one ultimately can handle them. The scale was implemented in Trier and Berlin in samples consisting of 549 youths in school or training, 115 students and 210 citizens. Originally this dimension was covered with 20 items – or in the shorter form with 10; of these, only three were used here.
33The scale of "self-esteem" (see Schwarzer et al. 1986: 47ff.) is derived from a self-description questionnaire for youths (mostly pupils and students). Along with self-esteem, other components of the self-conception were also elucidated, namely "physical self-concept", "social self-concept", "emotional self-concept", and "academic self-concept". It is assumed that these dimensions are hierarchically structured in that they are subordinated under "general self-esteem". Twelve items were used to capture this scale, which were again reduced to only three.
34As already mentioned, the scale of the "emotional self-concept" (see Schwarzer et al. 1986:15ff.) was taken from the self-description questionnaire. This dimension measures the emotions of the interviewees with 10 items out of which, again, three were selected. Both scales of self-esteem and of the emotional self-concept were applied to 507 students, high school students or pupils in Berlin. Eight answer categories were provided ranging from clearly applicable to clearly not applicable.
35In addition to these readily available and tested instruments, we formulated our own items to capture the feeling of the meaningfulness of life in the sense of the depth psychology theory of Viktor Frankl (1978). We judge this to be a central dimension of identity; the more clearly defined and solid the identity of person is, the more he/she should see a sense in life. An answer to the question of what the sense of life actually is can not be found so simply because meaning can not be provided from outside, it must be found (Frankl 1978). Every person must search for him/herself and construct a meaning that is individually valid. This meaning can be in reference to different areas and responsibilities in life: Family and children, community, self-actualization, health, support for others, the environment, work, ethical values, religion, or simply the enjoyment of life (see Schaeppi 2004: 74). A subjective theory of meaning is of great importance for the individual Identity: (1) such a theory defines the geographic, social, temporal, and intellectual-metaphysical space in which an individual lives; (2) it defines the subjective position or social role of one's own person in this space; (3) it supplies a justification for one's own existence; (4) it establishes individual goals and values (Schaeppi 2004: 272ff.).
36A total of 13 items were taken from the scale discussed above; four further items concerning the meaning of life developed by us were added. In order to check the final scale, a dimensional analysis was first conducted. The quality of the factor analysis was tested with the Kaiser-Meyer-Ohlin criteria (KMO). The resulting value of 0.81 confirms that the items are very well depicted by the chosen analysis method. The Bartlett test is also significant, which means that the correlation between the individual variables with a nearly 9% margin of error is not a matter of coincidence. The Anti-Image Correlation Matrix was then more closely examined. The diagonal of this matrix contains the MSA-values (Measure of Sampling Adequacy); the majority lie above 0.7 or 0.8; only a few fall in the area between 0.5 and 0.6, which does not require the elimination of any variables. The next step was the extraction of factors. Here we found five factors with an eigenvalue over 1; five factors were therefore extracted and these are listed in Table 3.
Table 3: Factor analysis of the Items on Aspects of Identity after Eliminating Two Items (rotated component matrix)
Notes: Extraction method: Major component analysis. Rotation method: Varimax with Kaiser-Normalisation. Sampling: representative population survey in Austria; N = 1006.
Source: ISSP Study 2004, Institute of Sociology at the University of Graz.
37In total, 59.5% of the total variance are explained by the five factors. The items "On the whole I have a high opinion of myself" and "On the whole I have a range of very good and valuable traits" were subsequently eliminated due to their low commonalities ( 0.5).
38The first factor contains the statements that imply thorough optimism about one's own actions. Three variables originate out of the scale "general self-efficacy" from Schwarzer et al (1986). General optimism goes hand in hand with a positive feeling about life that is not diminished when difficult times come. This is described especially by the item "Whatever happens, I will be okay", which loads highly almost equally in the first and second factors, but in content, it belongs to the first. To live life successfully according to personal intentions and perceptions while still maintaining equilibrium and composure does not lead to difficulties. This factor is given the name "Positive coping with life ".
39The second factor comprises variables of finding meaning in life. Here, meaning can be found in everything, even in suffering and difficult blows of fate: for this reason we refer to this factor with the concept "Feeling of meaningfulness in life".Such an attitude is certainly very helpful in solving life's problems. Therefore, it is an understandable fact that the item "Whatever happens, I will be okay" applies to the first as well as the second factor as indicated by a high loading.
40The third factor represents the opposite of the second and is referred to as "Feeling of meaninglessness”. In general it deals with a negative attitude toward life; one doesn't know what is worth living for and can not find any sense in life. For this reason, one is unable to cope positively with life's problems such as a serious illness; one would even prefer to die in such a case. It is not surprising that this personality disposition is coupled with emotional despondence and a lack of self-confidence.
41The fourth factor is characterized by a strong outward orientation. One observes one's self through a mirror represented by other people. The reactions of the environment are very meaningful. It is extremely important how others think of or perceive one. The concept of "public self-awareness" describes this constellation of variables (see Schwarzer et al. 1986: 3ff.).
42The last extracted factor is similar to the fourth except that the attention is not directed outward but inward. One desires to know and to observe one's self very precisely, which is expressed by the concept "private self-awareness". With these two scales, we are able to effectively reproduce the corresponding dimensions in the literature (Schwarzer et al. 1986: 3ff.).
43Subsequently five scales were created out of the variables in the five factors. The reliability analysis, however, did not produce very high values for Cronbach’s Alpha, which leads one to conclude that there is a low internal consistency of the items in the scales. One reason for the low values is surely the small number of adopted variables in the ISSP questionnaire. The personality scales of Schwarzer et al (1986) all consist of at least ten items and exhibit a considerably larger Alpha. It must also be considered here that the samples often contain a majority of students and pupils, i.e. they are more homogeneous than those of the ISSP study of 2004.
Table 4: Mean values, Standard Deviance and Cronbach’s Alpha of the Scales of Aspects of Identity
Sample: Representative population survey in Austria; N = 1006
Source: ISSP Study 2004, Institute of Sociology at the University of Graz.
44In light of the small number of items per scale, the reliability and validity display predominantely acceptable values. The first four scales are thoroughly suitable for further application in the data analysis. They depict dispositions toward dealing with one's life and match the way in which one lives and copes with problems. Therefore one can assume that they will also correlate with life satisfaction. This will be studied in the next section.
45In the ISSP Survey, a new instrument was developed to measure the different aspects of social identity. On this point the following question was asked: "We all have certain characteristics in common with others. A range of such characteristics are listed on this card. Please choose three characteristics that best describe you as a person. Which of them is the most important? And which is second most important? And which of them is third most important?" The interviewees could choose from the following answer categories: "a) my current or past profession (or being a homemaker); b) my national or ethnic origins; c) being a man or a woman; d) my age group; e)my religious convictions; f) my preferred political party, group or movement; g) my citizenship; h) my family status or my role in the family (husband/wife, widower/widow, son/daughter, father/mother, grandfather/grandmother, unmarried etc.); i) my social class (e.g. upper, middle, working, lower); j) the region of Austria I live in."
46The results show clearly that the family role is most often (41%) designated at first place for the description of one's own person. The profession is considered the most important characteristic for their personal description by 18% of those questioned; another 10% of the Austrians define themselves first and foremost in terms of their gender. The remaining social identities listed were designated less frequently as predominant. This, however, does not mean that they play a smaller role in people's lives. Looking at the designation at third place, it is noticeable that 18% of the Austrians consider the region they live in and 16% their citizenship as important components of their social identity. One can conclude from this that people are confronted with their family situation or their everyday occupational life most immediately and often, and therefore these parts are most present in their perception, but that also other social areas certainly shape their identities.
47The question of happiness or life satisfaction has been a longstanding concern for philosophers, writers and, even nation builders (the American Declaration of Independence of 1776 names the pursuit of happiness as a basic right of humankind). For several decades this theme – also treated under the term "subjective quality of life" - has become an important interdisciplinary field of research in which numerous contributions have been made by medical experts, psychologists, sociologists, economists and political scientists (see Argyle 1987; Michalos 1991; Myers 1993; Veenhoven 1999; Frey/Stutzer 2002; Haller/Hadler 2004, 2006). Being happy and satisfied is seen as an important goal in life by many people; in politics, the creation of the preconditions for happiness is also seen as an increasingly important goal, on a par with economic/monetary goals such as economic growth, full employment, etc. The social sciences primarily study the preconditions and consequences for happiness and satisfaction. In the process, it has become evident that life satisfaction also has positive consequences for other areas of social action (such as the willingness to get involved, the chances of finding a good job) and improves the further life expectancy in middle and senior ages. Here we are dealing primarily with the reasons why some people tend to be more happy and others more unhappy. In this context it is primarily of interest which characteristics of identity and personality influence life satisfaction.
- 7 It would naturally be better in terms of method to measure life satisfaction or happiness with a s(...)
48In the ISSP study, life satisfaction was measured with the following question: "When you consider your life as a whole, are you very satisfied, quite satisfied, rather satisfied, rather dissatisfied, or quite dissatisfied?"7 Somewhat less than a third (29%) of the 1006 people surveyed answered that they are very satisfied and as many as 52% reported that they are quite satisfied. When one now considers that another 16% are more satisfied than dissatisfied, one could believe that only a diminishingly small portion of the population struggles with life dissatisfaction. The data from the Social Survey Austria 2003 supplies us with a completely similar result. Here somewhat more (35%) of the 2000 people questioned reported being very satisfied and somewhat fewer (46%) quite satisfied. If one compiles these two answer categories, one comes to the exact same result for rather satisfied on the whole (81%). The remaining categories also differ by only one or two percentage points from each other. As a whole, therefore, a very high level of life satisfaction in the population is evident (see Schulz et al. 2005).
49Why do these answers emerge? One could assert that they contradict the booming demand in modern societies for psychologists, therapists, life advisors etc. (see Keupp 1994: 338f.; Bloch 1988: 13f.). Partly it can be argued that the offerings of consultation and therapy by these professions help "create" problems themselves (Andreski 1974) in that they make many people aware of their problems for the first time. Their offer, however, would not be accepted if it did not meet an actual and increasing need. Two explanations can be given for the generally high life satisfaction which is nevertheless detected. Firstly, based on the experiences of interviewers, people often answer such and similar questions with "One has to be satisfied" – as if according to the motto "you can't change anything anyway and so it is meaningless to complain". Secondly, it can be supposed that checking the alternative "rather satisfied" does not express a very positive feeling about life. Considered this way, one must assume that almost a fifth of the interviewees (18%) are only moderately satisfied with their current life situation.
50We will now examine which personal and social characteristics determine life satisfaction beginning with the demographic variables age and gender. With bivariate analyses hardly any differences can be ascertained. Of the interviewees under 50, the women are somewhat more satisfied than the men, among those who are older, the reverse is true. Yet one can detect a clear connection between education and life satisfaction with a very low probability of statistical error. Only 21% of the people with compulsory education are very satisfied with their lives as compared to 41% of those with a high school education.
51Using regression analysis, it will now be examined in more detail what influence the socio-demographic variables on the one hand, and personality characteristics on the other, have on life satisfaction. The results are presented in Table 5. It is confirmed here that the influence of neither gender nor age is significant. How satisfied a person is with life depends less on whether someone is a man or woman, old or young. The frequent and sweeping generalizations contrasting men and women (which usually postulate a discrimination against women) have no significance from the perspective of life satisfaction. The same is true for age. This is quite surprising as one could assume that the many hindrances and complaints of old age would be reflected in a reduced satisfaction in life. The fact that this is not the case leaves one to suppose that the expectations and demands in life are reduced at an advanced age, so that in this period of life, a balance between the two is possible. A large part of elderly people can also be happy about their good health. Many of them will have experienced the death of close family members and friends. On the other hand, the significant effect of the variable "highest educational attainment" is confirmed. Particularly the graduates of a compulsory school exhibit a significantly lower life satisfaction. These findings are very interesting and thoroughly plausible. In an epoch of a massive expansion in education, people who do not accomplish continuing education increasingly fall through the cracks because they are affected by unemployment to an over-proportional extent. They experience processes of social exclusion not only on the labour market but also in their private and personal social environment. For instance, men with no vocational training or education have a smaller chance of finding a marriage partner.
52Income represents a further important variable for life satisfaction; according to economic approaches, this should be the most important of all (see Frey/ Stutzer 2002). This supposition seems plausible as our society is very money and consumption oriented. It is however surprising that in our study it plays no role for life satisfaction how much money a person has at his/her disposal. This insignificance of financial conditions can certainly be traced back to one idea: that the subjective financial satisfaction is more important than objectively measurable income (see Haller & Hadler 2004: 26). Financial need is very different, depending on the household and family situation; this will certainly be considered in the subjective financial satisfaction. On the other hand, one must admit that "money (alone) does not buy happiness", and that close social relationships, a consolidated orientation in life and similar factors are more important.
53Also some further aspects of the personal and social life conditions seem important. Significant effects emerge out of the work and professional situation on the one hand, and the family situation on the other. It is well-known that work serves a range of functions beyond its economic importance, such as providing meaning, social integration and time structuring. The absence of these components is apparent in those affected by unemployment and therefore it has a significant negative influence on life satisfaction, as has already been shown in the classic Marienthal study (Jahoda, Lazarsfeld & Zeisel 1960). From the multiple regression analysis it becomes apparent, however, that being employed per se or living as a housewife has no influence on life satisfaction. The quality of the work plays a far more important role (see also Haller & Rosenmayr 1971). People in a leading position, for example, are significantly more satisfied than others. This type of employee has not only an interesting work as a rule, but also more freedom and latitude in disposition – an aspect that is very important for satisfaction (Haller & Hadler 2004). Not being employed, of course, has a negative effect on life satisfaction when – as in the case of unemployment – it is involuntary.
- 8 The independent variable (Life satisfaction) exhibits an ordinal scale measurement level. Fundamen(...)
Table 5: Multiple Regression analysis8 of general life satisfaction according to socio-demographic variables and characteristics of personality and identity
Notes: Significance: ***p 0,001, **p 0,01, * p 0,05, nsp 0,05; ns = not significant; N = 1006; missing values were replaced with mean values.
Source: ISSP Study 2004, Institute of Sociology of the University of Graz.
54The scales of social identity can accept values from 1 to 4. A value of 1 means that the corresponding social (partial) identity was named in the first position, the value 2 or 3 indicates that it was placed in the second or third place and a value of 4 indicates that it was not mentioned at al.
55Very clear effects are apparent with respect to the private/family situation. Married people and those living with a partner are satisfied above the average, divorced people are less satisfied with their lives (however the last effect is not significant). These findings on the central importance of a "significant other" for life satisfaction also emerged in a regression analysis of the World Value Survey data (Haller/Hadler 2004; 2006).
56Finally, it is remarkable that a significant connection exists between place of residence and life satisfaction. People in rural communities are more satisfied than those in urban areas. This effect even appeared in a study of children in a large city and in a small rural city (Zauner 1993). The causes for this could lie in the fact that cities – despite the variety of their professional and cultural offerings – exhibit a lower quality of life in many respects (for example, regarding environmental pressures, possibilities for spending free time in nature); despite the population density, people often remain more anonymous and their daily lives are flooded with irritations. In contrast, many studies show that people in the country or in small cities are more socially integrated (Haller & Janes 1996). There could be a selection effect, however, in that people with the desire for more freedom / less social control or those with certain problems tend to move to the cities.
57We now turn to the central thesis of this paper which states that, along with socio-demographic variables, personality and identity characteristics have an influence on life satisfaction. For this reason, we now examine the dependence of life satisfaction on the identity scales and the two Big Five personality scales (Extraversion and Neuroticism) previously described. The scale for "Positive coping with life" was eliminated from the analysis due to its high multi-collinearity.
58The two scales on meaning correlate with life satisfaction in a highly significant way. This is a clear confirmation of our thesis on the importance of the variable "Identity" in general, but also of the specific theory of Viktor Frankl in particular. In Frankls's view, the meaning of life is the most important component of identity. Our findings show that seeing a meaning in life implies a significantly higher satisfaction with one's own existence (Beta = 0.23). People who deny life this meaning tend to be more dissatisfied with their lives (Beta = -0.16). The effect of the variable "Meaningfulness of life" is the strongest of all 38 individual effects contained in the regression analysis in Table 5. This fact alone underscores how important this dimension is. People who see no meaning in life are more frequently despondent, depressed, and have less self-confidence. It might be mentioned here that in an analysis of the data from the World Value Survey, the variable "Feeling that one can make decisions freely in daily life" emerged as a very important determinant for happiness and life satisfaction (Haller/Hadler 2004). In our view, the consensus reached in these findings from various studies using different instruments clearly confirms the importance of the dimensions of personality and identity that are measured here.
59The two dimensions of self-awareness have only a minimal impact. The scale for private self-awareness – in other words, how well one knows one's self – has a weak but significant effect on life satisfaction: the more self-aware one is and the more one reflects about one’s self, the more satisfied one is with life in general. People with a low value on this scale reflect frequently on their own behaviour and therefore stand in a more balanced relationship with their social environment. In contrast, the desire to stand in a good public light has virtually no influence on satisfaction with one's own life. This is a highly interesting finding for social psychology and sociology. One could surmise that this desire is connected to the socially important but often destructive feeling of envy (Schoeck 1966). This emotion is often directly tied to personal discomfort as everyone knows from his/her own experience. People who continually register their effect on others in terms of attention might also frequently experience disappointment.
- 9 As stated above, three of the Big Five scales were not included in the regression for methodologic(...)
60Of the two "Big Five" personality characteristics measured in our study, only one has a statistically significant and relatively strong effect on life satisfaction, namely Neuroticism.9 If one has traits such as being reserved, nervous or insecure, then these tend to effect life satisfaction negatively. An insecure person has greater difficulty coping with daily life than someone with a "healthy" self-confidence. Furthermore, an insecure person will tend to avoid social contact; this, however, is absolutely crucial for a high quality of life (Haller/Hadler 2005). In contrast, an emotionally well-adjusted and stabile person reacts with more ease and composure to the behaviour of his or her fellow human beings and other external events; he/she is therefore happier seen on the whole. Contrary to expectation, the personality characteristic Extraversion had only a weak, insignificant influence. Whether a person is extroverted or introverted has little connection to his/her life satisfaction; our corresponding hypothesis could not be confirmed.
61Finally, almost all of the social identities measured in our study have a decisive and positive influence on life satisfaction. Our hypotheses here were considerably modified by the empirical findings. It is true – as we hypothesized – that people who identify with their family, professional, gender and age roles are more satisfied; identification with the family role has the strongest effect of those. Along with identification with one's professional role, identification with the gender role also has a quite strong and significantly positive effect on life satisfaction. Significant and quite clear effects also result from further aspects of the social identity, namely from ethnic/national, religious, regional, and class-specific identity. Only the political identity had no significant effect; this is not surprising considering the strong change of the Austrian political parties and the generally declining trust in the parties (see Plasser, Ulram & Sommer 2000). In general, it is revealed, however, that life satisfaction increases with the subjective importance of an identity facet. These can also be aspects of social identity that are important for only a few people today, such as religious or ethnic/national identity. The same is true for class and regional affiliation which are generally of diminishing importance. These findings that do not concur with our hypotheses can be very well explained with Viktor Frankl's Meaning Theory of life. This states that the meaning of life can be found in many various forms; one doesn't necessarily have to be married, employed etc. in order to find such a meaning. It can apparently also be found in active social participation – which surely is connected to a strong identification – in a religious community, a social class or an ethnic community.
62The regression analysis could clearly show the very important influence of characteristics of personality or identity on attitudes such as life satisfaction. As can be seen in Table 6, the model of personality and identity variables alone explains 28% of the variance while the model of socio-demographic variables alone explains only 17%. Furthermore, it is evident that the dimensions of identity exert a much stronger effect on life satisfaction than personality characteristics. This becomes apparent, although the fact must be considered that we measured the aspect of personality with far fewer variables than that of identity.
Table 6: Explained Variance in Life Satisfaction through different models (socio-demographic variables, personality characteristics, and total variables)
63It could be shown in this paper that personality characteristics and aspects of identity supply an important contribution to the explanation of life satisfaction. It can be assumed that these two variable complexes are also important for other social attitudes and modes of behaviour. Because life satisfaction itself already represents a very important variable, it seems generally necessary and promising for social research to intensify and expand efforts to develop instruments that can be implemented in standardized social science studies. The "economical" operationalising of personality characteristics and those of identity is admittedly a difficult undertaking. An important attempt in this direction was made by Beatrice Rammstedt and colleagues (Rammstedt et al. 2004; Rammstedt & John 2006). However, the positive result of a relatively high reliability of the BFI-10 that showed up in Federal Republic of Germany could not be reproduced in the same way in the Austrian study.10 This can probably be traced back to the different samplings as well as to the small number of items which was not sufficient to cover the broad content of the Big Five and to reproduce them comprehensively. Lang and Lüdtke (2005) also indicate that at least three items per personality characteristic are necessary to capture these reliably and with validity. It therefore seems required in further studies to consider not only efficiency and thrift, but also to pay attention that the content quality and the reliability of measuring instruments are ensured.
- 10 Based on many years of experience with international data from ISSP studies, we assume that there(...)
64Concerning the measuring of personality dimensions, it must be clarified whether the "Big Five" truly have the ability to capture a personality in all of its important facets (see Block 1995; Lang & Lüdtke 2005). A fundamental criticism of the model is certainly its high level of abstraction as well as the associated loss of content. For instance, artistic interest and tolerance are equally included in the factor "Openness", yet they have different relevance. Someone who is not occupied with art in his/her free time can be a thoroughly tolerant person and open for new experiences. Finally, the question of the temporal and intercultural stability of the Big Five personality characteristics requires further discussion. Wolfgang Jagodzinski (2005), for example, could not reproduce the Big Five very well in a study in East Germany. In order to take these context effects into account, his suggestion (2005) seems helpful, namely to differentiate between lasting components of personality (traits) and situation-dependent aspects (states). Even a person with a solid personality reacts differently in crises or in hostile environments than he/she usually does in everyday life. Therefore, it is necessary to examine personality characteristics also with regard to the context in which they appear.
65Along with personality characteristics, also aspects of identity were examined in our empirical research. Our study generated two important findings in this respect. Firstly, by means of items that we have developed ourselves, we were able to measure a central aspect of identity very well – the feeling of meaningfulness of life – which plays a central role primarily in the existential-analytical theory of Viktor Frankl. Secondly, we could show that this identity characteristic has a considerably stronger influence on life satisfaction than personality characteristics. Further facets of identity – such as identification with eleven different aspects of one's social identity (society, profession, family role, national identity, etc.) – almost all revealed significant and quite strong effects on life satisfaction. It is apparent, therefore, that the identity of a person is highly relevant for his/her attitudes – and presumably also his/her actions. For this reason, it seems meaningful and promising for further research to continue efforts toward an economical, practicable operationalisation of the concepts of identity and personality. The developed instruments should be integrated as explanatory variables in further social scientific studies; their ability to explain could be considerably increased in this way. The concept of identity seems particularly promising in this respect: it allows for a more conclusive alignment with the sociological theory of action than does the concept of (more or less) stable personality characteristics.