Value Profiles Among Finnish Adolescents and Their Associations With Subjective and Social Well-Being

Previous research on Schwartz’s theory of basic human values has mostly applied a variable-oriented approach. This study took a person-oriented approach and investigated how values co-occur and are manifested in individuals, that is, what kind of value profiles exist and how they differ in terms of subjective (i.e., life satisfaction) and social well-being (i.e., sense of belonging). In a sample of Finnish adolescents (N = 973; women 59.6%; Mage = 16.8, SD = .70), three value profiles emerged: personal-focus, growth-focus and self-protective, of which the personal-focus group scored highest on life satisfaction, the growth-focus group scored lowest on belonging to social media and the self-protective group scored highest on belonging to organizations. In all, subjective and social well-being were differently related to opposing values. The findings are discussed within a cultural and contextual framework.

Studies on values have typically adopted a variable-oriented approach where associations between individual values and a range of outcome variables have been examined and compared between various groups (e.g., gender, ethnicity, social class). Only recently have researchers turned to a person-oriented (i.e., typological) approach to investigate how values correlate within individuals, that is, what kind of latent configurations values form (e.g., Magun et al., 2015;Smack et al., 2017). This allows identifying groups of people with similar value profiles within a priori groups such as gender and nationality. It reveals how and to what degree a priori groups -often perceived as homogenous (e.g., men vs women) -are in fact internally divided. At the same time, though, it allows maintaining a sensitivity to what type of value combinations dominate (at a given time) in given social groups. More importantly, it allows comparing the profiles on various outcome variables in order to evaluate which value configurations (i.e., profiles) have a better or worse fit in their environment, that is, how they match the expectations of society.
To date there is a wealth of research on values and well-being but still little research on value profiles among adolescents and how they differ in terms of subjective (i.e., life satisfaction; Diener et al., 1985) and social well-being (i.e., sense of belonging; Baumeister & Leary, 1995). Understanding young people's value profiles can provide indications of the direction of societal development. Also, how values and well-being are linked is vital information for teachers, youth workers, health professionals and other practitioners who support the well-being and development of youth in their present circumstances. The current study aims to contribute to this gap in previous literature by examining 1) what value profiles emerge among Finnish adolescents using Latent Profile Analysis (LPA) and 2) how these groups differ in terms of subjective and social well-being. Schwartz's (1994) widely used theory of universal human values forms the basis of our study. Schwartz (1992Schwartz ( , 2012 has identified ten universal basic values: power, achievement, hedonism, stimulation, self-direction, universalism, benevolence, tradition, conformity, and security (see Fig. 1). According to him, the ten values form a circular continuum (i.e., circumplex) in which adjacent values such as universalism and benevolence are compatible, sharing the same motivational goals, while standing in stark opposition to values at the other end of the continuum, in this case power and achievement. Some of the core characteristics of values are that they are beliefs inextricably linked with affect, meaning that individuals feel (e.g.) happiness or despair depending on whether they can pursue their cherished values, or they are threatened. Further, people hold a hierarchy in values and their behaviour, at least in theory, is guided by the relative importance of multiple values. This means that although the content and structure of values are considered universal, the relative importance of each value may differ across groups and cultures (viz. regions, ethnicity, nationality) (Bardi et al., 2014;Schwartz, 1992). Recently, Schwartz's value theory was further 1 3

Schwartz' Value Theory
Value Profiles Among Finnish Adolescents and Their Associations… refined by subdividing the ten values into nineteen (e.g., Lee et al., 2019;Schwartz et al., 2012). For comparative purposes, however, in this study we will focus on the ten value dimensions.
The ten values have also been confirmed to organize around four higher-order value types along two bipolar dimensions (Cieciuch et al., 2014): openness to change vs conservation and self-transcendence vs self-enhancement (see Fig. 1). The former dimension (in the following referred to as openness) reflects the degree a person favours independence of thought, action and feelings, and readiness for change (i.e., self-direction, stimulation) over order, self-restriction, preservation of the past, and resistance to change (i.e., security, conformity, tradition), or vice versa (Schwartz, 2012). The latter dimension, in turn, reflects the degree a person favours concern for welfare and the interests of others (i.e., universalism, benevolence) over Schwartz quasi-circumplex model of value types and higher order dimensions. Note. From "Hierarchical Cluster Analysis of Human Value Priorities and Associations with Subjective Well-Being, Subjective General Health, Social Life, and Depression across Europe," by Â. Leite et al., 2021, Social Sciences, 10(2), p. 74. Copyright 2021 by MDPI the pursuit of one's own interests and relative success and dominance over others (i.e., power, achievement), or vice versa. Hedonism, by contrast, shares elements of both openness and self-enhancement. Schwartz (2012) has also proposed two additional organizing principles of values based on whether their focus is more personal or social (e.g., achievement vs tradition) and whether they are more self-protective or growth oriented (e.g., power vs self-direction). That is, the former dimension discriminates between values that concern self-expression (personal interests and characteristics) and how one relates to others. By contrast, the latter dimension discriminates between values that build on fear and perceived threats (i.e., so-called anxiety-based values), and values that build on "fearless" self-expansion and promotion of goals for gain (i.e., anxiety-free growth values).

Person-oriented Approach
Most studies on value diversity within or between countries have taken a variableoriented approach, meaning the investigation of values in terms of means, standard deviations, correlations, or other statistical measures (e.g., Fischer & Schwartz, 2011;Minkov & Hofstede, 2012;Verkasalo et al., 2009). Values have been shown to be meaningfully associated with gender, age, ethnicity, and personality traits, among other things (e.g., Parks-Leduc et al., 2015;Schwartz & Rubel, 2005). For instance, using European Social Survey data (ESS), Robinson (2013) showed that stimulation was negatively and tradition positively correlated with age. Similarly, with regard to the higher-order values, people tend to move from openness and self-enhancement towards conservatism and self-transcendence with age (see also Gouveia et al., 2015). Gender-wise, self-transcendence (and caring for others) is more common among women and self-enhancement more common among men (Beutel & Marini, 1995;Robinson, 2013). ESS data also shows that urban youth are more liberal than their peers who grow up in rural areas, a divide that has drastically sharpened during the last decades (Lassander, 2017).
Lately researchers have turned to typological, also called person-oriented approaches to identify how values latently co-occur and are manifested within individuals (Daniel et al., 2020;Leite et al., 2021;Magun et al., 2015;Moors & Vermunt, 2007;Smack et al., 2017;Ungvary et al., 2018). The person is here seen as an entity and individuals with similar configurations across a set of variables are then combined into homogeneous groups (also called profiles or classes; see Bergman & Trost, 2006). A purpose of the person-oriented approach is to detect non-linear associations (e.g., value hierarchies or systems) between the relevant attributes that can have a meaningful effect on several outcome variables. Further, the researcher is not bound by a priori groups in statistical comparisons (e.g., gender, SES, ethnicity) but is instead able to examine how latent value profiles cut across these groups (Osborne & Sibley, 2017). That is, the person-oriented approach instantly reveals what (latent) "subgroups" of values exist and dominate within social groups as well as how and to what degree members of these groups are divided and what they share with members of other groups. In this way it "deconstructs" traditional/dichotomous views on social groups as homogenous and opposite to other groups.
For instance, Magun et al. (2015) identified five value profiles in ESS data that map onto Schwartz's (1992Schwartz's ( , 2012 taxonomy of value configurations. ESS data is considered nationally fairly representative of European populations in terms of gender, SES and ethnicity. Three groups endorsed relatively conservative and self-transcendent values (labelled strong social-focus, weak social-focus, weak personal-focus), one group championed openness and self-enhancement (labelled strong personal-focus), and one group strongly favoured openness and self-transcendence (labelled growth). In general, the distribution of the profiles in each country was associated with the country's ranking on various indicators of resources for human development. That is, in Nordic and western European countries ranking high on these indicators, the growth profile dominated, followed by weak social-and personal-focus profiles. The situation was somewhat the opposite in lower ranked Mediterranean and post-communist countries, where the weak and strong social-focus profiles as well as the personal-focus profile dominated. Furthermore, both social-focus groups correlated positively with age. Taken together, which value profiles (i.e., value systems or hierarchies) dominate depends on individual opportunities and age in a given cultural context. However, it is worth noting that European countries were intrinsically diverse in their value profile compositions and the study did not examine the impact of (e.g.) SES or ethnicity on profile membership.
Further, Ungvary et al.s' (2018) study on the links between values, aggression and leadership found four value profiles among ethnically and socioeconomically diverse North American and Israeli adolescents: self-focused, anxiety-free, otherfocused, and undifferentiated. In contrast to Magun et al.'s (2015) results, these pertained to the personal-focus, growth and social-focus groups, as well as one group scoring in between on all value dimensions. The self-focused adolescents were the most aggressive and were viewed as leaders by their peers compared to the other groups, and together with the anxiety-free youth they reported more delinquency than the others (Ungvary et al., 2018). Also, girls were overrepresented in the anxiety-free (i.e., growth) and other-focused (i.e., social-focus) groups. Only small cultural differences emerged, with Israeli adolescents being somewhat more other-focused and less self-focused than their American peers. Four similar profiles were identified also in Daniel et al.'s (2020) longitudinal study of sub-urban middleclass Australian children aged 5-10 (i.e., self-focus, growth-focus, social-focus and undifferentiated). The social-focus profile was more common among girls and older children and predicted more prosocial and less aggressive behaviour over time. Finally, in a study with ethnically diverse North American undergraduate students aged 17 to 42 (representative of the public universities where the two studies were conducted), Smack et al. (2017) identified only a personal-and social-focus profile. The former was significantly smaller (16%), more common among men, and in terms of personality traits, showed stronger openness than the social-focus group. Agreeableness, in turn, was significantly stronger in the social-focus group than in the personal-focus group.
To sum up, value profiles present the hierarchical "mix" of values that people hold theorized by Schwartz (2012) and they can be used to better understand and predict behaviour. The few studies that yet exist today have linked value profiles with aggressiveness, leadership, and personality. However, how value profiles are related to well-being is still unexplored.

Subjective and Social Well-Being
Well-being has multiple dimensions and aspects such as subjective, psychological, material, social and physical (OECD, 2020;Ryff, 1989). Due to the characteristics of the data used in this study, we focus here only on the subjective and social. The former is assessed with life satisfaction and the latter can be assessed with sense of belonging (Diener et al., 1985;Keyes, 1998). Life satisfaction makes up the cognitive component (vs affective) in Diener's et al. (1985) model of subjective wellbeing (SWB). It is defined as an individual's global judgement (vs domain-specific) about the quality of her/his current life situation. The judgement is fully subjective, meaning the individual compares her/himself to others and what she/he thinks is an appropriate standard. In general, studies show that young people are somewhat more satisfied with their lives than older people (Baird et al., 2010) and boys more so than girls (Moksnes & Espnes, 2013). However, life satisfaction is relatively stable across the lifespan with fluctuations only around certain life events such as unemployment or bereavement, before returning to pre-event levels (Luhmann et al., 2012). Globally, national levels of SWB have become a tool to evaluate the progress of countries and to inform public policy (Diener et al., 2009). Northern and Western European countries typically score the highest on SWB (Helliwell et al., 2012). These countries provide a favourable context of high cultural egalitarianism and socioeconomic level. Countries with low levels of SWB tend to be low in cultural egalitarianism and socioeconomic level (Sortheix & Lönnqvist, 2014;Sortheix & Schwartz, 2017).
A sense of belonging (SB), in turn, is considered to be one indicator or dimension of social well-being (cf. social integration; Keyes, 1998;Rath & Harter, 2010). Building on Maslow (1943), Baumeister and Leary (1995) insist that humans have an innate core motivation or need to belong, that is, to form and maintain social attachments, and they define SB as the feeling of being accepted and liked by other group members, feeling connected to them and feeling like a member of a community. This was later made an integral part of self-determination theory (SDT) as a core need of relatedness (Deci & Ryan, 2012). This is because all personal achievements are recognized, validated, valued and meaningful only by and through other people. That is, no solitary self and achievements exist. Thus, it is only logical that SB strongly predicts meaning in life and firm mental health (Kitchen et al., 2012;Lambert et al., 2013;Sargent et al., 2002). Family, friends, school, nationality, and political associations make up examples of groups or communities that individuals typically affiliate with. Moreover, the importance of these affiliations varies according to the individual and they substitute for each other, meaning that lack of belonging to (e.g.) family or friends can be compensated by belonging to other groups (Baumeister & Leary, 1995). SB bears many similarities with social identity theory but they stem from different research traditions and the latter focuses on intergroup processes (Hogg, 2000).
In Finland the annual Youth Barometer study (YB; ages 15-29) has now for over two decades measured SB for a variety of social units such as family, peer group, school/work community, city/municipality, congregation/church, Finnish society, and the European Union (Myllyniemi & Kiilakoski, 2017). The YB sample is considered nationally fairly representative in terms of gender and SES. Overall, analyses show that family and friends are the most important affiliations, followed by Finnish society and school; girls feel more strongly tied to family and friends and boys more so to hobbies and the neighbourhood; higher educated youth have weaker SB to family and friends and stronger SB to Finnish society and the EU; the life circle seems to grow with age because the importance of larger communities (i.e., city, Finnish society, the EU) increases with higher age; and SB is generally stronger in rural than in urban areas. Longitudinal analyses, in turn, reveal that SB to Finnish society has increased since the global economic crisis began in 2008. And yet, to our knowledge, previous international studies have examined only one type of SB and rarely in relation to values (e.g., Akar-Vural et al., 2013;Newman et al., 2007).

Values and Well-Being
There is no straightforward link between values and well-being. Growth values (i.e., openness and self-transcendence) tend to be linked with higher subjective well-being around the world (see above) but these associations are very much mediated by personality traits on the individual level (Haslam et al., 2009) and societal context on the structural level (Sortheix & Lönnqvist, 2014;Sortheix & Schwartz, 2017). That is, people pursuing values congruent to those prevailing in their cultural context, whatever they be, generally report higher life satisfaction (and fewer depressive symptoms; Boer, 2017). In countries scoring high on the Human Development Index (HDI; UNDP, 2020) such as Finland, high life satisfaction goes hand in hand with growth values, whereas in low HDI countries marked by tougher individual competition, life satisfaction is positively linked with self-protective values (i.e., conservation and self-enhancement) (Sortheix & Lönnqvist, 2014;Sortheix & Schwartz, 2017). There seems to be a similar link between SB and well-being. An Iranian study by Joshanloo and Ghaedi (2009) found that social well-being (i.e., defined as a sense of integration; Keyes, 1998) was positively linked with conformity and negatively with self-direction. Schwartz and Sortheix (2018) suspected the finding to be an outcome of the cultural context, and Iran indeed ranks significantly lower on the HDI than most western countries (#70; UNDP, 2020). However, the result might also be due to differences in what the subjective and social well-being dimensions tap into. When compared, the former rests more on how individuals see themselves thriving in their private/personal lives, whereas the latter builds on how individuals see themselves thriving in their public/social lives (Keyes, 1998). In other words, social well-being might be positively associated with conservation values also in western countries.
Further, longitudinal studies show that causal relationships as well are not straightforward but reciprocal. That is, an increase in materialistic values such as power decreases well-being, in line with theory, but there is also a weak influence in the opposite direction (summarized by Schwartz & Sortheix, 2018). Recently Grosz et al. (2021) found a reciprocal causal relationship between subjective well-being and openness (but not the other three core values). Hence, existing studies suggest that the interlinks between values and well-being are moderated by individual and social factors but the evidence is not conclusive and more research on values and different dimensions of well-being is needed.

Current Study
Research on value profiles, that is, how values co-occur and are manifested within individuals is scarce. Moreover, the connections between youths' value profiles and their well-being is an under-studied area in the previous literature. To contribute towards filling this gap, our objective here is to investigate what typologies of values emerge in a sample of Finnish upper secondary school students (grades 10-12) and how they relate to both subjective (i.e., life satisfaction) and social well-being (i.e., SB). The affective side of SWB (i.e., depressive symptoms) was not considered due to restrictions of data. In terms of SB, we want to know specifically how various subdimensions of belongingness are related with values (cf. conservation; Joshanloo & Ghaedi, 2009). The subdimensions are picked from the Finnish YB (Myllyniemi & Kiilakoski, 2017) and they reflect core socialisation units in society: family, peers, social media, organizations (e.g., associations, congregation) and larger community (e.g., Finnish society). Our hypotheses were as followed: Hypothesis 1: Based on recent studies employing LPA in adolescent samples and cross-cultural profile comparisons in Europe (Daniel et al., 2020;Magun et al., 2015;Smack et al., 2017;Ungvary et al., 2018), four value profiles will emerge. These will be the growth-focus (i.e., high scores on openness and self-transcendence), the personal-focus (i.e., high scores on openness and self-enhancement), the social-focus (i.e., high scores on conservation and self-transcendence) and undifferentiated profiles (i.e., scoring around the mean on all values). Hypothesis 1a: The personal-focus profile will form the largest group because young people tend to endorse openness and self-enhancement more than older people (Gouveia et al., 2015;Robinson, 2013). Hypothesis 1b: Regarding gender, girls will be overrepresented in the growth and social-focus profiles and boys will be overrepresented in the personal-focus profile (see Ungvary et al., 2018;Daniel et al., 2020). Hypothesis 2: The personal-focus profile will score the highest on life satisfaction. This is mainly because self-enhancement values are important for young people in general (Gouveia et al., 2015;Robinson, 2013) but also because openness values are highly cherished in high HDI countries such as Finland, where openness has previously been linked with high life satisfaction (Schwartz & Sortheix, 2018).
We did not form hypotheses regarding social well-being. Although the Iranian study by Joshanloo and Ghaedi (2009) suggests a positive link between social well-being and conservation (and a negative link with openness), this might be due to a cultural variable (i.e., Iran is generally conservative and collectivistic, conformism is valued) but also because the study used a total score for various facets of social well-being (i.e., integration, contribution, coherence, actualization, acceptance). In our study we focus on one dimension of social well-being (cf. social integration; Keyes, 1998) -SB and its sub-parts.
Neither did we form hypotheses regarding profile composition by living area because all participants were from urban/suburban areas. Similarly, no hypotheses regarding language group were formed due to lack of previous studies.

Participants
The participants were 973 general upper secondary school students from several urban and sub-urban areas in Finland. General upper secondary school is equivalent to (American) high school and consists of grades 10-12 (age 16-18). Most of the respondents were from the Helsinki capital area (N = 552; 56.7%) and the rest from smaller provincial cities (N = 421; 43.7%). Most respondents were Finnish speaking (N = 575; 69.8%), but Swedish-speaking youth were also well represented due to research interests beyond this study (N = 398; 40.2%). Swedish-speaking Finns, the largest language minority in Finland, make up only 5.2% of the total population, so they were clearly overrepresented in the sample (Statistics Finland, 2020).
The data was gathered during spring 2018 with paper-and-pencil questionnaires, in classrooms with a teacher present. The study adhered closely to The Finnish Advisory Board on Research Integrity (2012) national guidelines for responsible conduct of research in Finland. Participation was based on informed consent from both students and their parents. All the data was gathered anonymously.

Values
Values were measured with the short 21-item Portrait Values Questionnaire (PVQ-21), developed for the ESS from the original 40-item measure (Schwartz et al., 2001). Respondents rate how well descriptions of people with different preferences and behaviour match themselves on a scale from 1 (not like me at all) to 6 (very much like me). For instance, achievement is measured with "It is important to her to show her abilities. She wants people to admire what she does" and benevolence with "It's very important to her to help people around her. She wants to care for their well-being". The degree of subjective similarity is then interpreted as an indicator of the corresponding value. The PVQ-21 has two items for each value, except for universalism, which is measured by three items. The higher-order values can be calculated with the same items (Conservation: security, conformity, tradition; Openness: stimulation, self-direction; Self-enhancement: power, achievement; Self-transcendence: universalism, benevolence). The scores for the two higher-order value dimensions are calculated by subtracting the individual score for Conservation from the Openness score and the score for Self-Enhancement from the Self-Transcendence score. Hence, the two value dimensions measure a preference for Openness over Conservation and for Self-Transcendence over Self-Enhancement. Although a popular scale in values research, it lacks proper validation and psychometric properties are seldom given in previous research (e.g., mostly western samples, values seem to overlap/difficult to form, see Sandy et al., 2017). Further, research has shown the scale to be measurement invariant across 15 European nations only regarding openness and self-enhancement (Cieciuch et al., 2018). Descriptives of PVQ-21 in this study are found in the supplementary material.

Satisfaction With Life
Life satisfaction was assessed with the five-item Satisfaction With Life Scale (SWLS; Diener et al., 1985), where respondents rate on a scale from 1 (strongly disagree) to 7 (strongly agree) items such as "I am satisfied with my life" (*alpha = 0.87). The scale has shown excellent psychometric properties. It is built firmly on a single factor, it has shown good internal consistency (0.79-89; e.g., Pavot & Diener, 2008) and a strong negative correlation with (e.g.) Beck's Depression Inventory (-0.72; Blais et al., 1989).

Sense of Belonging
Belongingness to different social groups and communities was measured with the belongingness scale developed for the Finnish consumption and lifestyle survey, conducted annually since 1999 (e.g., Koivula et al., 2018;Näsi et al, 2011). The measure has also been used annually in the Finnish Youth barometer since 2004 (Wilska, 2004). Respondents rate with one item on a scale from 1 (not at all) to 5 (very strongly) their sense of belonging with a particular group or community (i.e., "How strongly do you feel you belong to…?"). In this study respondents evaluated their belongingness to 11 social units, comprising five higher-order groups. The latter were formed based on explorative factor analysis (see supplementary material). Three items represented family (SBfam; immediate, household, extended; *alpha = 0.69), two items represented peers (SBfr; friends, study community; *alpha = 0.68), one item represented social media (e.g., facebook; SBsome), three items represented organizations (SBorg; hobby clubs (e.g., sports), organizational activities, congregation etc.; *alpha = 0.67) and two items represented belongingness to the larger community (SBlc; city, Finnish society; *alpha = 0.62). Although friends and study community as well as hobby clubs and organizational activities might appear as quite different domains with different implications each, they do overlap. For instance, school is the central place for meeting friends in adolescence (Ferguson et al., 2022) and hobby clubs represent one form of organizational or social activity beyond friends.

Data Analysis
We specified a similar Factor Mixture Analysis model (FMA, see e.g. Clark et al., 2013) as Magun et al. (2015), which takes into account individual response style by a method factor that affects all the value items. This has been recommended by Schwartz et al. (2012) too. We employed a combination of established fit indices to find the best fitting profile solution. See the supplementary material Statistical analyses for information on the analyses conducted, fit indices and criteria applied and the statistical programs used. No covariates were included in the analyses. The supplementary material (including All item/category probabilities) as well as data and scripts to reproduce the results are available online (https:// osf. io/ xrwhj/? view_ only= f0964 5a7b5 8446f 3bee7 00e4b 890f5 04).
We fitted models starting from one and kept increasing the number of profiles. The lowest score on the Consistent Akaike's Information Criterion (CAIC) was achieved with two classes and the lowest Bayesian Information Criterion (BIC) -score was achieved with a model estimating three classes (see Tables 1, 2). Additional profiles showed issues with instability and convergence, indicating that the data does not support extracting more than three profiles. The Bootstrapped Likelihood Ratio Test (BLRT) supported three profiles over two (p < 0.001) and the entropy value for three profiles was considerably higher than for two. Thus, we decided on the three-profile model and saved the Most Likely Class (MLC) for further use. In what follows, (latent) class and profile will be used synonymously.

Factor Mixture Analysis
In order to examine the value profiles content-wise we, first, situated all the profiles from our sample into the bigger picture of the European sample (Magun et al., 2015) by examining the location of the profiles in the two-dimensional value space (Fig. 2). Second, we inspected how the profiles were characterized by the four composite value dimensions (Fig. 3), and third, we examined the most accurate view of the profiles utilizing the individual item response probabilities for the two highest categories for each value item. These are presented in Fig. 4 (all item/category probabilities can be found in the supplementary material Table 1).
Our first hypothesis (H1) on the number of profiles and their specific patterns was partially supported. Only three profiles emerged, not four as we expected. However, the profile configurations were mainly in line with our expectations. The profiles were labelled personal-focus (56%), growth-focus (29%) and  self-protective (15%) based on their locations in the two-dimensional value space and their internal differences. The personal-focus group scored relatively high on openness and self-transcendence, the lowest on conservation, and close to the mean on self-enhancement. In terms of the individual value items, these individuals endorsed success, a good time and fun, and disliked rules, proper behaviour and traditions more than their peers. In turn, the growth-focus group scored the highest on openness and self-transcendence, and the lowest on self-enhancement. They put a high weight on modesty, helping and understanding other people as well nature, compared to their peers. Finally, the self-protective group placed near the center in the two-dimensional value space and could as such be labelled undifferentiated. However, because LPA is employed to identify relative differences between latent groups and this one scored significantly higher on conservation and self-enhancement than the two other groups, we considered the term undifferentiated incorrect if not misleading at least. Instead, as this group placed "opposite" to the growth-focus group in the value space and emphasized respect,  rules, and traditions more than the others, we labelled this self-protective according to Schwartz's (2012) taxonomy. For the above-mentioned reasons, our hypothesis (H1a) regarding the value preferences of the largest group was also supported. That is, although most respondents were located towards the "growth corner" in the two-dimensional value space, the personal-focus profile (the largest group) was more hedonistically and individualistically oriented than the growth-focus group. Thus, most respondents endorsed the combination of openness and self-enhancement in relation to the others.

Profile Differences
Our hypothesis regarding gender differences (H1b) was also partially supported (Fishers Exact p < 0.001). As expected, there were more girls than would have been expected by chance in the growth-focus profile (standardized Pearson residual 3.87). However, boys were not overrepresented in the personal-focus profile, as we expected, but instead in the self-protective profile (standardized Pearson residual 4.19).
Although no hypotheses were formed regarding profile composition by living area (i.e., capital metropolitan area vs the other) (Fisher's Exact p < 0.001), the results showed that the self-protective profile was more common in the capital area than elsewhere (standardized Pearson residual 2.25). Moreover, Finnish-speaking students were overrepresented (Fisher's Exact p < 0.001) in the growth-focus profile (standardized Pearson residual 2.44) and Swedishspeaking students in the self-protective profile (standardized Pearson residual 2.67). See Table 3 for profile differences by gender, living area and language.
Furthermore, our hypothesis (H2) regarding profile differences in terms of life satisfaction was also supported (see Table 4). As we expected, the personalfocus profile scored the highest on life satisfaction (ξ = 0.19). However, the difference was significant only in relation to the self-protective group. Finally, regarding SB, the profiles showed a small effect in relation to social media (ξ = 0.22) and organizations (ξ = 0.26). Individuals in the growth-focus profile showed the weakest SB to social media and individuals in the self-protective group the strongest SB to organizations. In turn, all groups felt equally affiliated with their family, peers, and larger community.

Discussion
This study employed a person-oriented approach and examined how values cooccur in individuals, that is, what value profiles emerge in a sample of Finnish upper secondary school students and how they differ in terms of subjective and social well-being. First, the three value profiles that were identified, personal-focus (56%), growthfocus (29%) and self-protective (15%), fit Schwartz's taxonomy (2012) and previous studies (Daniel et al., 2020;Magun et al., 2015;Ungvary et al., 2018). The personal-and growth-focus groups both scored relatively high on openness and self-transcendence. This is in line with previous studies on value profiles across Europe (Magun et al., 2015) which show that Finns in general are growth oriented in their values. However, the personal-focus group had a stronger preference for self-enhancement than the growth-focus group, hence the label "personal". As this group was the largest, our results were also in line with findings on values among adolescents in general (Robinson, 2013) and urban Finnish youth in particular which show that youths are comparatively self-centered and open-minded (Lassander, 2017). They typically endorse openness over conservation and self-enhancement over selftranscendence. The cultural context combined with age most likely also explains why neither a social-focus nor "pure" self-protective group emerged in this sample: In a high HDI country like Finland, with a high cultural egalitarianism and socioeconomic level, youth can afford to be comparatively anxiety-free and self-expansive. There is little need to fear uncertainties in the social and physical world, avoid conflict (conformity), maintain order (tradition, security), or actively control threats (power). Therefore, a combination of conservatism and self-enhancement/transcendence is very rare among young Finns. Accordingly, the self-protective group, endorsing a combination of more conservative and self-enhancement values (Schwartz, 2012) than their peers, was the smallest.
Second, girls were overrepresented in the growth-focus group and boys in the self-protective group. The former observation is in line with previous findings  (Ungvary et al., 2018) and theory on gender socialization, that is, girls in general endorse openness and self-transcendence (and care for others) more than boys (Beutel & Marini, 1995;Hitlin, 2006). Regarding the latter observation, we expected the boys to be overrepresented in the personal-focus group. This was not the case most likely because girls were overrepresented in our sample and most respondents placed towards the upper right corner in the two-dimensional value space, meaning everyone was relatively growth oriented. Instead, as the self-protective group formed the opposite to the growth-focus group, it was only logical that boys were overrepresented in this group (see Robinson, 2013). Third, our results supported previous findings on the relationship between values and subjective well-being (e.g. Schwartz & Sortheix, 2018): adolescents with a personal-focus profile scored highest on life satisfaction. Openness is common among Finns (and other high HDI-countries) and self-enhancement among youth in general. This allows again two interpretations: it reflects the fact that this particular combination of values is "objectively" the perfect balance for good mental health, or, it reflects the fact that adolescents who conform to contextual norms (i.e., country-and age-wise) indicate superior well-being (see Boer, 2017). This might be a combination of both but is for now left an open question. Also, we were not able to determine the impact of personality to values in this study (see Haslam et al., 2009).
Fourth, our finding that the self-protective group scored highest on SB to organizations is logical considering that organizations to some extent consist of hierarchies (e.g., congregation) and competition (e.g., hobbies). This is also partly in line with the Iranian study by Joshanloo and Ghaedi (2009), who found a positive link between conservatism and social well-being. Our results also showed that adolescents who endorse growth values had a significantly weaker SB to social media than the others. This is somewhat unexpected as social media is an information sharing tool and a medium for social interaction, suggesting a good fit with growth values. Then again, social media is also heavily focused on self-presentation and impression management, being perhaps more in line with self-enhancement. We found no significant differences between the value profiles on the other subdimensions of SB.
Thus, the link between values and social well-being (here indicated by SB) is not straightforward and it exists only in relation to organizations and social media. Nonetheless, this suggests that the association between conservative values and social well-being does not appear to be a mere contextual effect of the Iranian culture as suggested by Schwartz and Sortheix (2018) but a more general rule. More importantly, the associations observed here position social well-being somewhat in opposition to subjective well-being: the former co-occurs with conservatism and the latter with openness. Therefore, different values entail different types of well-being, either more individual or collective. However, what those who score high on either subjective or social well-being have in common, relatively speaking, are stronger self-enhancement values. Again, this fits previous findings on the importance of self-enhancement for adolescents, as they are in a sensitive, developmental phase that requires self-focus and autonomous decisions (Robinson, 2013). And yet, despite the statistically significant well-being differences between the profiles (both subjective and social), it is worth observing that the effect sizes were between small and medium. This means that the practical significance of these differences may be marginal.

Practical Implications
Methodologically, the study shows the strength of person-oriented approaches to values. For instance, an examination of individual variables would have revealed non-significant associations between (higher-order) values and life satisfaction, non-significant to moderate associations between values and SB, as well as moderate positive associations between life satisfaction and SB (see Table 1). However, this study shows that values form meaningful hierarchies (i.e., profiles) that are not uniformly connected to well-being. Instead, certain values go together with subjective well-being and others with social well-being. This adds another layer of complexity to the previous mixed findings on the cultural effects on values and subjective well-being (Sortheix & Lönnqvist, 2014;Sortheix & Schwartz, 2017). Moreover, despite the narrow sample and somewhat restricted effect sizes (i.e., life satisfaction and SB) in this study, the results were meaningful and expected. Future studies would benefit from a more diverse sample (e.g., ethnicity, age), other dimensions of well-being (e.g., selfesteem, depression) as well as socioeconomic status as a background variable. This should introduce more variation in values and therefore produce additional and more diverse profiles, as ethnicity, age and living area (i.e., region) have been linked with values (Lassander, 2017;Magun et al., 2015;Robinson, 2013).
Societally, this study provides vital information for teachers, youth workers, health professionals and policy makers on dominant values among Finnish adolescents and the potential future direction of Finnish society. As values are rather stable over time (Vecchione et al., 2016) and the overwhelming majority were here found to endorse growth values, it suggests that the Finnish society will stay egalitarian and inclusive in the foreseeable future. However, this general benevolence may to some extent be an effect of adolescence. A recent study by Mannerström et al. (2020) found Finnish young adults to be highly divided in political attitudes, with nearly half holding strong conservatist and anti-immigrant views. This is in line with the international trend on a deepening polarization in values and the rise of right-wing populism (Norris & Inglehart, 2019). Added with the covid pandemic, swift changes may take place. In uncertain conditions, people compensate their insecurity with firmer ties to social groups with more rigid and extremist views (Hogg & Adelman, 2013). This might also be a reason for why young people with right-wing conservatist views seem to be better off in terms of mental health than their more leftist-liberal peers (Mannerström et al., 2020). Hence, values might be more flexible and dependent on the societal situation (viz. crisis) than previously acknowledged. Yet, together with the current study, these findings show that values per se do not reveal whose well-being and development needs support. Instead, the interlinks with well-being are complex and in need of more research.

Limitations and Suggestions for Future Research
The results must be considered within some limitations. First, the sample was not randomly chosen or representative of all Finnish adolescents (the majority being urban/sub-urban youth), yet alone adolescents from other cultures. Any generalizations of the profiles and their links with well-being variables should be made with caution. Future research needs to address samples that are more diversified in terms of background. Second, the study used a cross-sectional sample, meaning it is not possible to examine development and causation in and between value profiles and well-being. Longitudinal studies would provide a deeper understanding of these interlinks. Third, the study focused only on the cognitive component of SWB. Future studies could benefit from including the affective component as well (i.e., depression). Also, interpretations regarding social media SB are problematic because only a single-item measure could be used in this study. Hence, multi-item scales of social media are recommended in the future to capture the complexity of social media engagement. Finally, data was collected with self-reports that do not reveal the subjective meaning attached to items and responses. Thus, the interpretations made by researchers and respondents may differ. Relatedly, social desirability effects typically play a role, meaning that survey responses reflect social expectations. That is, respondents indicate what they believe is expected of them (in terms of values) and not what they honestly think or feel. Future studies could employ interviews to scrutinize the subjective meanings attached to values.

Conclusions
Despite its limitations, the present study contributed to our understanding of values and well-being both methodologically and societally. First, adding to a growing body of research, the study shows the benefits of a typological approach. That is, although our sample of Finnish students in general scored high on openness and self-transcendence in the two-dimensional value space, examining combinations of values revealed the existence of three distinct value profiles or groups (i.e., personal-focus, growth-focus and self-protective) with some endorsing either more conservatism or self-enhancement or both, than the majority. Second, the findings revealed significant differences between the profiles related to subjective and social well-being (i.e., life satisfaction vs belongingness). Importantly, life satisfaction was the highest in the most popular value profile. Whether this is a consequence of the specific value combination or its normativity is left an open question. Third, associations between individual value dimensions and well-being need further research because they appeared somewhat conflictual. Whereas life satisfaction was positively related to openness, belongingness was positively related to conservatism. We suggest that future studies employ interviews to further scrutinize the subjective meanings of openness and conservatism and their relation to well-being.