top of page

The Winning Personality: The Role of Personality Traits in Educational Inequality

Evan Carron-Kee

Inter-disciplinary research is crucial for furthering our understanding of educational inequality. This article contributes to that effort by integrating trait psychology into cultural capital research. First, the article establishes the theoretical motivation for expanding Bourdieu’s concepts of habitus and cultural capital using trait psychology. Second, drawing on recent work in psychology, and education, the article proposes a novel pathway through which differences in personality traits might lead to educational inequality. Finally, the article presents results from an exploratory study of Irish students who sat the Leaving Certificate between 2016 and 2021. These results suggest that in higher level Leaving Certificate English and Mathematics, students with certain personality characteristics received better results under teacher-assigned grades compared to the traditional standardised and anonymised examinations. While this result cannot be generalised to the wider population, it provides some justification for the future investigation of personality traits in educational inequality and into the role of the teacher-selection mechanism. 

Literature Review 

Cultural Capital 

Cultural capital is a resource consisting of any collection of mannerisms, cultural tastes, and strategies of interactions with educators which is both associated with high status backgrounds and rewarded by the education system (Bourdieu, 1973; 1986; Lamont and Lareau, 1988).  Bourdieu (1973) argued that the behaviours and characteristics which make up cultural capital come as second nature to the children of high socio-economic status (SES) families. These families possess and inculcate in their children an upper class ‘habitus’, a mental framework of behaviours, dispositions, values, and tastes, which are shaped by the social background in which one grows up. Elements of this high-status habitus are then rewarded by the education system, meaning that the children of well-off families do better in school and maintain their advantage. Just as intergenerational wealth inequality is created by wealthy families passing on economic capital to their children, Bourdieu proposed that intergenerational educational inequality is created by well-educated and high SES families passing on cultural capital to their children.

Wildhagen (2009) outlines two mechanisms through which cultural capital might lead to better academic outcomes: teacher-selection and self-selection. Self-selection proposes that students with more cultural capital find school to be a more natural and welcoming environment due to the characteristics of their habitus, and thus put more effort into their studies and choose to study for longer. Teacher-selection proposes that students with more cultural capital better understand the implicit norms of education and use specific strategies of interaction with teachers. Therefore these students have better relationships with teachers, improving their outcomes relative to peers with less cultural capital. This might lead to a positive feedback loop where the student is rewarded better for their effort than a student with less cultural capital, and thus decides to put in more effort and pursue further education. Although Wildhagen (2009) found evidence only for the self-selection mechanism, it is possible that the teacher-selection mechanism only works for certain aspects of cultural capital. If this is the case, then the true relevance of the mechanism would be obscured by not specifying which aspects of cultural capital we expect to be relevant.  

There are two key issues which researchers must investigate to understand the role of cultural capital in educational inequality. First, the specific traits and behaviours which make up cultural capital vary between contexts and over time. Cultural capital is not defined as any specific trait or behaviour. Instead, any trait or behaviour which is associated with a high SES background and is rewarded by the education system may be considered cultural capital (Lamont and Lareau, 1988). Therefore, we must first identify what counts as cultural capital. Secondly, the definition of cultural capital does not specify how these traits or behaviours lead to better outcomes for students. To show that cultural capital is important for educational inequality, we must investigate the mechanisms through which it improves grades. This article argues that personality traits can be usefully considered an element of cultural capital, and proposes a model in which personality traits affect educational outcomes via the teacher-selection mechanism.

Personality Traits 

In order to understand the role personality traits play in educational inequality, it is important to understand how they relate to Bourdieu’s theoretical framework, to academic achievement, and to SES. I will discuss each of these issues in turn. 

Personality traits are defined as stable and enduring patterns of thought, behaviour, beliefs, and attitudes which vary across individuals (Mammadov, 2022). Habitus is defined as a mental framework of behaviours, dispositions, values, and tastes which are shaped by social structure. The similarities between these concepts and the potential of combining them seems clear. For instance, it is possible that personality traits might describe some characteristics of habitus. This would provide a useful way of quantifying the differences between the upper and working class habitus, and perhaps allow us to better identify which elements of habitus act as cultural capital. One of the most widely used personality taxonomies is the ‘Big Five’, which postulates five domain-level traits: Openness/Intellect, Conscientiousness, Extraversion, Agreeableness, and Neuroticism (McCrae & Costa Jr., 2008). The ‘Big Five’ is hierarchically ordered so that within each trait there are facets which represent more specific aspects of that trait. Kaiser and Schneickert (2016) argue that the opportunity to expand the concept of habitus is clear.  Integrating personality into a Bordieuan framework might help us identify the internal psychological aspects of habitus, measure their effects on inequality, and, given that personality traits manifest in observable differences in behaviour, identify new aspects of cultural capital.

If personality traits are to expand our conception of habitus, then they should both be shaped by social background and affect SES outcomes. Most literature discussing personality and SES treats personality as an ‘inner resource’, which is innate rather than socially determined (Shanahan et al., 2014). However, other researchers have investigated how personality is affected by, and serves to reproduce, social class. Leckelt et al. (2019) find that high-net-worth (<€1m) Germans have different personality characteristics compared to the general population, exhibiting the traits which are most associated with upwards social mobility (high extraversion, openness, conscientiousness, narcissism, low neuroticism, and an internal locus of control). Shanahan et al. (2014) finds that traits including agreeableness, intellect, and neuroticism partially mediate the effect of parents’ education on their child’s education and hourly wages as an adult. Kaiser and Schneickert (2016) also found it helpful to use personality to quantify habitus. Their results suggest promise in using personality traits to quantify elements of habitus. By showing that certain traits are associated with better outcomes, and that these traits are more common in well-educated and well-off people, they make a convincing argument to treat personality traits as a valuable form of capital which facilitates the reproduction of social class. It should be noted, however, that the relationship between SES outcomes and personality is less clear. Some large-scale studies have found a positive correlation between SES and conscientiousness and a negative correlation between neuroticism and SES (Jonassaint et al., 2009; Judge et al., 1999; Shiner et al., 2003). On the other hand, Hughes et al. (2021) found weak or non-existent relationships between ‘Big Five’ traits and SES outcomes.

In contrast to SES, the relationship between personality traits and academic achievement is well established (De Raad and Schouwenburg, 1996; Hill and Jackson, 2016; Poropat, 2009). A recent meta-analysis has confirmed that personality is a key predictor of academic achievement, and pointed to conscientiousness and openness/intellect as the two most important traits for academic performance (Mammadov, 2022). However, the way academic achievement is measured matters (Hübner et al., 2022). This study considers two distinct measures: standardized test scores, which are taken to be an objective measure of competence, and teacher-assigned grades, which may also measure competence but may be affected by teachers’ subjective judgements of their students. In the American context, where students take standardized tests but grades are often based on the teacher’s judgement, grades and test scores are only moderately correlated. This suggests that they may measure different aspects of a student’s competence and behaviour (Borghans et al., 2016; Willingham et al., 2002). Congruent with this view, personality differentially predicts outcomes across achievement measures, predicting teacher-assigned grades more accurately than standardized test scores (Borghans et al., 2016; Lena Roemer et al., 2022). It seems likely that personality contributes to different achievement measures through different mechanisms.

Roemer et al. (2022) distinguishes between two broad pathways through which personality affects academic achievement: competence-related mechanisms and competence-independent mechanisms. Competence-related mechanisms propose that personality leads to behaviours which directly affect achievement. For example, openness/intellect is significantly associated with critical thinking, and peer learning (Bidjerano & Dai, 2007). These behaviours may then foster greater mastery of the material, leading to higher grades and scores on achievement tests.  Competence-independent pathways are associated with better results on teacher-assigned grades but not necessarily on standardized tests, and generally propose that certain traits are associated with behaviours which are rewarded by teachers but not connected to actual competency. For example, the ‘invest and accrue’ model of conscientiousness holds that individuals with high conscientiousness tend to display behaviours which involve expending more effort in the present to receive greater rewards in the future (Hill & Jackson, 2016). Westphal et al. (2021) finds that students with high conscientiousness received better teacher assigned grades compared to peers with lower conscientiousness controlling for differences in standardised test scores. They suggest that teachers reward students who exhibit these behaviours in class, even if they do not actually contribute to mastery of the material, by assigning better grades to students who display conscientiousness-driven behaviours. Thus, conscientiousness appears to improve grades in ways that are not  related to increased competence.

Combining Two Disciplines 

To my knowledge, only one study has combined personality traits, habitus, and cultural capital. Kaiser and Schneickert (2016) found that three ‘Big Five’ facets—focus (a facet of conscientiousness), intellect and curiosity (facets of openness/intellect)—partially mediated the effect of parental education on teacher-assigned grades. Thus, these facets may be considered elements of cultural capital. However, a key limitation of Kaiser and Schneickert’s work is that they did not investigate the mechanism through which high SES personality traits led to greater academic achievement. It is this issue which I will consider next.

The Teacher-Selection Mechanism for Personality 

There is a striking overlap between the teacher-selection mechanism proposed to explain cultural capital effects and the competence-independent pathways proposed to explain why certain personality traits affect teacher-assigned grades more strongly than standardized tests. In both cases, characteristics of the student which are not directly related to competency, are preferred by teachers, leading to positive academic outcomes. It seems natural then to propose a teacher-selection mechanism to explain how personality works as an element of cultural capital. If teachers have more positive evaluations of the abilities of students with high SES personality characteristics, then they might overestimate the grades of these students. Based on the literature reviewed above, the most likely trait to fulfil this role is conscientiousness. It is associated with high-SES backgrounds (Jonassaint et al., 2009) and more strongly associated with positive results in teacher-assigned grades than standardised tests (Westphal et al., 2021). Kaiser and Schneickert (2016) also found that one facet of conscientiousness may be considered an element of cultural capital.

B3C26DA9-00C0-4635-9833-324BBC274A78_4_5005_c.jpeg

Figure 1 represents the hypothesised teacher-selection mechanism for personality. Pathway 𝑎1 represents the effect of SES on personality traits. Pathway 𝑎2 represents the effect of personality traits on academic achievement, measured in grades. Here, the personality trait operates as a mediator for the effect of SES on academic achievement. Pathway 𝑐 represents the effect of SES on grades that isn't mediated through personality traits. Finally, grade type—teacher-assigned or standardised test—operates as a moderator of the relationship between personality and grades. Its value affects the strength of the effect of personality on grades. This is represented by pathway 𝑏. The effect of personality on grades is therefore 𝑏×𝑎2. The key test of the model is whether this moderated mediation relationship exists. If the mediation of SES advantage on grades through personality is stronger under teacher-assigned grades than under standardised tests, this would provide evidence that the teacher selection mechanism is an important pathway for explaining how personality traits play a role in cultural capital.

Exploratory Analysis 

Background 

This model was originally intended to be tested on data collected from former Irish Leaving Certificate (ILC) students who were assessed between June 2016 and June 2021. However, the sample size (n=93) was too small to test the teacher selection model outlined above. Therefore, the analysis excludes the SES component and instead tests for the moderation effect of grade type on the relationship between personality traits and grades. The ILC usually consists of 6-8 subjects, mostly assessed through standardised and anonymously marked exams. There are other components, such as oral, practical, and project work, but the exams carry the most weight in calculating final grades. This research collected data from students who sat higher level Irish, English, and Mathematics, which have no project or practical examinations. As school-leaving exams, they play an important role in determining students’ opportunities in higher education. In 2020, due to the Covid-19 pandemic, ‘calculated grades’, which relied mostly on teacher judgement, were offered to students for the first time, as an alternative to in-person assessments. There are some peculiarities which make the ILC calculated grades distinct from the teacher-assigned grades referred to in the literature review. The school principal and the other subject teachers had to agree on the grades and class rankings of all students of a particular subject. A standardisation process was implemented centrally by the Department of Education, which adjusted school-assigned grades based on students' previous results in state exams. Finally, teachers were encouraged to base their teacher-assigned grades on class tests, mock exams, and their estimation of what the student would have achieved in a normal year. As such, they were actively discouraged from allowing their subjective judgements or perception of the student’s ability or potential to affect their decision. For the purposes of clarity, I will still refer to these as teacher-assigned grades, although it is important to be aware of the aforementioned differences.  

Hypotheses: 

      H1:    Certain personality traits partially mediate the positive effect of SES on grades 

      H2:    This mediation is moderated by assessment type, such that it is stronger under teacher-assigned grades than under exams 

Methods 

Data was collected through an online survey of people who took higher level English, Irish, and Mathematics at Leaving Certificate level between June 2016 and June 2021. As there is no available sampling frame of past Leaving Certificate students, the survey was promoted on social media and respondents answered through an online platform. To incentivise participation, eight €25 One4All vouchers were randomly distributed to participants. Women are somewhat overrepresented in the sample, while students who took exams are underrepresented. This can be attributed to the fact  that the survey was open to students from four years before calculated grades were introduced but only two years after calculated grades were introduced. The small and unrepresentative sample means that the results cannot be generalised to the wider population. The results should therefore be interpreted as exploratory in nature, and simply provide guidance on whether this hypothesis should be tested in a large-scale study of the calculated grades process. 

 

Descriptive Statistics 

______________________

 

 

78EA2CCE-0C9A-43E4-A673-4F32E44D75C6_4_5005_c.jpeg

Results 

An exploratory data analysis, involving fifteen linear regressions, yielded two notable results. This analysis investigated the relationships between each personality trait, subject, and an exam dummy (where 0 indicates teacher-assigned grades and 1 indicates examined grades), while controlling for gender and a simple measure of SES. First, students with higher levels of intellect/openness received slightly higher grades in English when they received teacher-assigned grades compared to when they sat examinations. Second, students with higher levels of conscientiousness received slightly higher grades in mathematics when they received teacher-assigned grades compared to when they sat examinations. The marginal effect of an increase in the relevant personality trait on teacher-assigned grades and examined grades is reported in Table 4. In both cases, an increase in the personality trait is associated with a small but significant reduction in examined grades, and a small but significant increase in teacher-assigned grades. It is important to note that after applying a Bonferroni correction to the significance level of 10%, neither of these results are significant.   

Note: Grades are presented on an 8-point scale, where 1 is the highest grade (H1 or A1) and 8 is the lowest grade (H8 or F). Therefore, negative coefficients imply a positive effect on grade achieved. Grades from 2016/17 were converted to the current grading intervals based on the closest approximate interval. Personality traits are presented on a 20-point scale, where higher values correspond to higher intellect/openness. Well Off is a simple measure of SES, where a higher value indicates that the respondents’ family was better off compared to their peers at the time they sat the Leaving Certificate. 

 

* p < 0.10, ** p < 0.05, *** p < 0.01, p<0.001****

0EC088D3-A79E-4743-BDBC-616FDAB0CAD2.jpeg
39B849A9-499B-4773-8B7E-F2DE24D2607C.jpeg
624650A2-AF37-472A-9025-35F202C54408_4_5005_c.jpeg

Discussion 

Any analysis of these results must take into account two serious limitations: the small and unrepresentative sample, and the high p-values (especially given the number of tests which were carried out). However, the results provide preliminary support for the claim that English and maths teachers of the students in the sample overestimated the grades of students with certain personality characteristics. This implies that teacher-assigned grading may result in different outcomes for students with certain personality characteristics compared to traditional examinations. To accept this explanation however, we must explain why openness/intellect was only associated with better teacher-assigned grades in English, and why conscientiousness was not associated with better teacher-assigned grades in English or Irish. Proposed explanations for these facts are offered below and should be examined in future research.

First, there is a plausible explanation for why intellect/openness to experience might be associated with grades in English but not Irish and mathematics. Those with high levels of intellect/openness, to experience, tend to be more imaginative, creative, and curious (McCrae & Costa Jr., 2008). These traits may be rewarded more heavily in English, which includes a creative writing component and rewards students for the aesthetic quality of their work, in contrast to Irish and mathematics which are less focused on these criteria. Hence intellect/openness may be seen as desirable by English teachers, and they may overestimate its role in predicting achievement. Teachers’ perceptions should be studied directly to test this explanation in future research.

Second, given the previous literature on teacher-assigned grades and personality traits, there is no clear reason why conscientiousness would have no association with teacher-assigned grades in Irish or English. It may be a result of the fact that the sample is older than the sample used by Kaiser and Schneickert (2016). At an older age, factors such as extrinsic motivators, social pressures, and access to resources may be more likely to determine outcomes, making the effect of personality traits harder to detect in a low-powered study. It is also possible that the importance of the exams played a role; perhaps students who aren’t naturally inclined to display conscientiousness-driven behaviours were motivated by the importance of the exams. Finally, teachers may have been more resistant to allowing subjective judgements regarding conscientiousness-driven behaviours to affect their grading decisions, given that they were specifically advised to assign the grade they think the student would have got in an exam (Department of Education and Skills, 2021). Future research should test the role of personality in academic achievement for older cohorts and in high-stakes assessments. Of course, it is also possible that the effect would have been detected with a large and representative sample.

Theoretical Limitations 

This research suffered from three significant limitations beyond the empirical weaknesses identified above. First, it used only one personality framework. Future research in this area would benefit from exploring other personality indicators, such as the dark triad (cf. Papageorgiou et al., 2020), and more detailed indicators such as personality facets, which are subsets of each of the ‘Big Five’ personality traits (McCrae and Costa Jr., 2008). Indeed, Kaiser and Schneickert (2016) note that personality effects on grades which are not significant at the level of traits may become significant when facets are used instead. Second, the research only considered a narrow range of subjects. Students usually receive grades in seven or eight subjects at either ordinary or higher level for their Leaving Certificate. This study only considered the three mandatory subjects, and only at higher level. A more comprehensive study would investigate effects across all subjects, and both levels, to provide a broader picture, especially concerning ‘lower’ achieving students who may take subjects at ordinary level or may have an exemption from studying Irish. Finally, the results lack external validity. The social and educational environment under which teacher-assigned grading systems were introduced in Ireland was vastly different to the pre- and post-pandemic environments, making it impossible to truly understand the effect of the new grading system in isolation from the effects of the pandemic.

Conclusion 

This article has made both empirical and theoretical advances on the existing literature. Despite significant limitations, the empirical findings suggest that the assessment method used in the Irish Leaving Certificate may have led to different outcomes for students depending on their personality characteristics. These results provide support for the inclusion of personality questions on future studies of teacher-assigned grading and for future large-scale studies on the fairness of the Irish calculated grades process. The theoretical contribution—the application of the teacher selection mechanism to personality traits—should also inspire future research on teacher-assigned grading, but with a focus on testing for the mechanisms through which such grading systems might lead to educational inequality. Further integrating trait psychology into cultural capital research appears to be a promising path forward for sociologists studying educational inequality.

Bibliography 

Bidjerano, T., & Dai, D. Y. (2007). The relationship between the big-five model of personality and self-regulated learning strategies. Learning and Individual Differences, 17(1), 69–81. https://doi.org/10.1016/j.lindif.2007.02.001.

 

Borghans, L., Golsteyn, B. H. H., Heckman, J. J., & Humphries, J. E. (2016). What grades and achievement tests measure. Proceedings of the National Academy of Sciences, 113(47), 13354–13359. https://doi.org/10.1073/pnas.1601135113.

 

Bourdieu, P. (1973). Cultural Reproduction and Social Reproduction. In Knowledge, Education, and Cultural Change (p. 42). Routledge.

 

Bourdieu, P. (1986). The Forms of Capital. In Handbook of Theory and Research for the Sociology of Education (p. (pp. 241-258)). Greenwood. 

De Raad, B., & Schouwenburg, H. C. (1996). Personality in learning and education: A review. European Journal of Personality, 10, 303–336. https://doi.org/10.1002/(SICI)1099-0984(199612)10:5<303::AID-PER262>3.0.CO;2-2. 

Department of Education and Skills. (2021). DES Calculated Grades Information. https://www.gov.ie/en/publication/fc8f7-a-guide-to-state-examinations-and-accreditedgrades-for-leaving-certificate-2021-confidentiality-and-data-protection/. 

Hill, P. L., & Jackson, J. J. (2016). The Invest-and-Accrue Model of Conscientiousness. Review of General Psychology, 20(2), 141–154. https://doi.org/10.1037/gpr0000065. 

Hübner, N., Spengler, M., Nagengast, B., Borghans, L., Schils, T., & Trautwein, U. (2022). When academic achievement (also) reflects personality: Using the 

personality-achievement saturation hypothesis (PASH) to explain differential associations between achievement measures and personality traits. Journal of Educational Psychology, 114(2), 326–345. https://doi.org/10.1037/edu0000571. 

 

Hughes, B. T., Costello, C. K., Pearman, J., Razavi, P., Bedford-Petersen, C., Ludwig, R. M., & Srivastava, S. (2021). The Big Five Across Socioeconomic Status: Measurement Invariance, Relationships, and Age Trends. Collabra: Psychology, 7(1), 24431. https://doi.org/10.1525/collabra.24431.

Jonassaint, C., Siegler, I., Barefoot, J., Edwards, C., & Williams, R. (2009). Low Life Course Socioeconomic Status (SES) is Associated with Negative NEO PI-R Personality Patterns. International Journal of Behavioral Medicine, 18, 13–21https://doi.org/10.1007/s12529-009-9069-x. 

Judge, T. A., Higgins, C. A., Thoresen, C. J., & Barrick, M. R. (1999). The Big Five Personality Traits, General Mental Ability, and Career Success Across the Life Span. Personnel Psychology, 52(3), 621–652https://doi.org/10.1111/j.1744-6570.1999.tb00174.x

Kaiser, T., & Schneickert, C. (2016). Cultural Participation, Personality and Educational Inequalities. Sociological Research Online, 21(3), 41–56. https://doi.org/10.5153/sro.4063. 

Lamont, M., & Lareau, A. (1988). Cultural Capital: Allusions, Gaps and Glissandos in Recent Theoretical Developments. Sociological Theory, 6(2), 153–168. https://doi.org/10.2307/202113. 

Leckelt, M., Richter, D., Schröder, C., Küfner, A. C. P., Grabka, M. M., Grabka, M., Grabka, M. M., & Back, M. D. (2019). The rich are different: Unravelling the perceived and self‐reported personality profiles of high‐net‐worth individuals. British Journal of Psychology, 110(4), 769–789. https://doi.org/10.1111/bjop.12360.

 

Lena Roemer, Lechner, C. M., & Rammstedt, B. (2022). Beyond Competencies:Associations between Personality and School Grades Are Largely Independent of Subject-Specific and General Cognitive Competencies. Journal of Intelligence, 10(2), 26–26. https://doi.org/10.3390/jintelligence10020026.

 

Mammadov, S. (2022). Big Five personality traits and academic performance: A meta-analysis. Journal of Personality, 90(2), 222–255. https://doi.org/10.1111/jopy.12663. 

McCrae, R. R., & Costa Jr., P. T. (2008). The five-factor theory of personality. In Handbook of personality: Theory and research, 3rd ed (pp. 159–181). The Guilford Press. 

 

Papageorgiou, K. A., Likhanov, M., Costantini, G., Tsigeman, E., Zaleshin, M., Budakova, A., & Kovas, Y. (2020). Personality, Behavioral strengths and difficulties and performance of adolescents with high achievements in science, literature, art and sports. Personality and Individual Differences, 160, 109917. https://doi.org/10.1016/j.paid.2020.109917 

 

Poropat, A. E. (2009). A meta-analysis of the five-factor model of personality andacademic performance. Psychological Bulletin, 135(2), 322–338. https://doi.org/10.1037/a0014996.

 

Shanahan, M. J., Bauldry, S., Roberts, B. W., Macmillan, R., & Russo, R. (2014). Personality and the Reproduction of Social Class. Social Forces, 93(1), 209–240. https://doi.org/10.1093/sf/sou050. 

Shiner, R. L., Masten, A. S., & Roberts, J. M. (2003). Childhood Personality Foreshadows Adult Personality and Life Outcomes Two Decades Later. Journal of Personality, 71(6), 1145–1170. https://doi.org/10.1111/1467-6494.7106010.

 

State Examination Commission—Exam Material Archive. (n.d.). Retrieved February 16, 2023, from https://www.examinations.ie/exammaterialarchive/. 

 

Westphal, A., Vock, M., & Kretschmann, J. (2021). Unraveling the Relationship Between TeacherAssigned Grades, Student Personality, and Standardized Test Scores. Frontiers in Psychology, 12. https://www.frontiersin.org/article/10.3389/fpsyg.2021.627440. 

 

Wildhagen, T. (2009). Why Does Cultural Capital Matter for High School AcademicPerformance? An Empirical Assessment of Teacher-Selection and Self-Selection Mechanisms as Explanations of the Cultural Capital Effect. The Sociological Quarterly, 50(1), 173–200. https://doi.org/10.1111/j.1533-8525.2008.01137.x.

Willingham, W. W., Pollack, J. M., & Lewis, C. (2002). Grades and Test Scores: Accounting for Observed Differences. Journal of Educational Measurement, 39(1), 1–37. https://doi.org/10.1111/j.1745-3984.2002.tb01133.x

18920782_1579443455408285_7419471100481888458_o_edited.jpg
bottom of page