LIVING IN COHOUSING COMMUNITIES Personality Traits and Trait Emotional Intelligence

Cohousing communities are characterized by an increased amount of exchanges in social support. Since this circumstance could be attributed to certain characteristics of their inhabitants, the aim of this study was the evaluation of personality traits and trait emotional intelligence. To this purpose, a group difference study was performed in Germany between a sample of residents in cohousing communities (n = 180) and inhabitants of common neighborhoods (n = 104). Significant differences were found that support the idea that residents of cohousing communities have higher levels of well-being and minor levels of detachment and psychoticism, as well as a lower overall score for maladaptive personality traits. We have concluded that further research is needed to examine the possible causal relationships between these findings, and to verify whether living in a cohousing community can operate as a moderator of these traits or if their inhabitants had already bore them before moving into such communities. study has shown that residents of cohousing communities have, from various points of view, lower levels of maladaptive personality traits and a higher level of well-being. Although these results are favorable, further research is necessary to verify whether living in a cohousing community can operate as a moderator of these traits or if its inhabitants had already enjoyed them before moving into the community. This question could be clarified through future investigations that deepen the study of such communities and evaluate the causal relationships through longitudinal studies and/or mediation analysis, as well as their effects on psychological symptoms and/or psychological well-being.


Introduction
Cohousing consists of a cooperative life that emerged in Denmark in the early 1970s due to social changes, especially the aging of the population and the changed role of women in society (Fedrowitz & Gailing 2003). In subsequent years, it has spread not only throughout Denmark, Sweden and the Netherlands, but also in the United States of America (Cummings & Kropf 2020). In these communities, each family unit has an apartment or house while enjoying a common area for group meetings (Luscombe 2019). Hence, cohousing includes the fusion of two opposing aspects: on one hand, individual autonomy and freedom, and on the other hand, a traditional community life marked by the feelings of security and belonging (Berghäuser 2013). Due to this, cohousing communities are, in a certain way, the antithesis of the current process of individualization (Andritzky 1999). Several studies have been able to show that, in these communities, social support exchange is substantially higher than in ordinary neighborhoods (Fedrowitz 2016;Margolis & Entin 2011;Markle et al. 2015) and this is precisely the reason why Parke (2017) considers cohousing communities as extended families.
Considering that this circumstance could be traced to certain characteristics of their inhabitants, the need for a closer analysis of personality traits arose. On the one hand, Goldberg's (1990) Big Five model describes five personality traits: Openness, conscientiousness, extraversion, agreeableness and neuroticism. On the other hand, the DSM-5 establishes a taxonomy of 25 maladaptive personality facets that are ascribed to five higher-level dimensions. These are negative affectivity, detachment, antagonism, disinhibition, and psychoticism (Krueger et al. 2011).
Of particular interest is that the first four domains of the DSM-5 (disinhibition, detachment, antagonism, and negative affectivity) clearly correspond to the negative (that is, socially undesirable) poles of the following four dimensions of Goldberg's (1990) Big Five: conscientiousness, extraversion, agreeableness and neuroticism. The connection between DSM-5 psychoticism and Goldberg's (1990) openness to experience, on the other hand, is less clear (Tyrer 2012).
Likewise, it is important to bear in mind that exploratory factor analysis has shown that trait emotional intelligence (or trait EI) is an independent factor of the Big Five (Petrides et al. 2007). As its components are well-being, emotionality, sociability, and self-control (Petrides 2009), it is conceivable that living in a cohousing community, which is characterized by an increased amount of social support exchange, could also promote well-being.

Present study
Although most of the research about cohousing communities maintains a sociological perspective and focuses on the aforementioned social support, it should be noted that there are some studies with a psychological approach. In a comparative study conducted in the United States, Waxman (2005) evaluated a sample of 31 residents in cohousing communities and an equivalent sample of conventional residents, and found no differences concerning personality traits, preferences, and values. Sanguinetti (2014) assessed residents of cohousing communities and their attachment to place, and Pereira and colleagues (2019) conducted a study on their connection to community and nature. Neither of these two investigations performed a comparative study with residents in traditional neighborhoods. To conclude, Schetsche and colleagues (2020) carried out a group difference study and found that, in comparison with residents of cohousing communities, habitants of traditional neighborhoods have higher levels of compulsive, anxiety, depressive, and eating disorders, as well as more use of coping strategies that are based on social withdrawal, problem avoidance, and emotional concealment. In line with previous studies (Fedrowitz 2016;Margolis & Entin 2011;Markle et al. 2015), they also found that the habitants of cohousing communities exhibited more use of social support seeking.
These investigations and their empirical evidence of greater social support in such communities were the starting point from which the present study sought to evaluate the personality traits and the trait emotional intelligence of its residents, thus comparing its results with those of a sample of inhabitants in common neighborhoods. Although Waxman (2005) already conducted a study on personality traits, it should be mentioned that the sample size was relatively small and that this study was conducted in the United States. To our knowledge, no study has assessed the personality traits of residents in German cohousing communities.
Taking into account the increased amount of social support exchange in cohousing communities, the hypothesis was that their residents have lower levels of detachment and antagonism and, on the other hand, higher levels of sociability and well-being.

Sample and procedure
The non-probability sample was collected online in Germany between April 14 and May 5, 2020. To ensure anonymous participation according to the General Data Protection Regulation that came into force within the European Union on May 25, 2018, we used a Google Form that was created on a G-Suite account. Through the acceptance of the Data Processing Amendment, this account guarantees the confidential treatment of information, including fingerprints that participants leave when they fill out a form located on the servers of the aforementioned company. To corroborate the satisfactory completion of the survey, a pilot test was conducted with over 30 individuals.
To recruit residents in traditional neighborhoods, the form was spread through Facebook, and to access the sample of residents in cohousing communities, we consulted the website of the Trias Foundation, or Stiftung trias in German, due to its inclusion of an index of approximately 1,000 cohousing communities and their corresponding contact details. Randomly, a total number of 150 cohousing communities were selected and contacted, to inquire about their interest in taking part in this study. Thirty-five administrators committed to forward the form link to their adult residents so that they could voluntarily participate in the survey. Table 1 shows the sociodemographic data of the two samples, including the 284 individuals who participated: 180 of them living in cohousing communities and 104 residing in ordinary neighborhoods. In the sample of cohousing community inhabitants, the mean age was 55.14 years (SD = 13.67) and the mean age of the non-residents in cohousing communities was 39.48 years (SD = 11.74). Of all the participants who reside in cohousing communities, 123 were female (68.33%) and 56 male (31.11%) whereas 85 participants from common neighborhoods were female (81.73%) and 16 male (15.38%). Thirty-two of the cohousing participants studied between 11 and 15 years (17.78%) and 126 of them studied between 16 and 20 years (70.00%). On the other hand, 40 residents in common neighborhoods studied between 11 and 15 years (38.46%) and 51 between 16 and 20 years (49.04%). The sense of community increases with years of residence (Pretty et al. 1996) and tends to be higher among inhabitants of small cities than among residents of large cities (Prezza et al. 2001). Since the lack of sense of community is associated with the perception of health problems (Farrell et al. 2004), it was decided to include years of residence in the analysis. On average, residents of cohousing communities lived in the same place for 8.68 years and residents of traditional neighborhoods for 7.86 years.

Personality Inventory for DSM-5-Brief Form -PID-5-BF
The Personality Inventory for DSM-5-Brief Form, or PID-5-BF, consists of the short version of the Personality Inventory for DSM-5, or PID-5, by Krueger and colleagues (2011). This instrument includes the five dimensions which are proposed by the DSM-5: detachment: disinhibition, antagonism, psychoticism, negative affectivity. For each of these dimensions, five items have to be answered on a 4-point Likert scale (0 = very false to 3 = very true). Zimmermann and colleagues (2014) undertook the German translation of the long version and the short version but published only the psychometric values of the former. In a study by Wissing and Reinhard (2017), who used this brief version of the inventory, internal consistencies were found between .77 ≥ α ≥ .65.
During the pilot test of the present study, which was carried out on 30 individuals, a large number of participants expressed their desire to mark a neutral or intermediate response, for which a 5-point scale was introduced.
In the Argentinian translation and validation process of the same instrument, Sanchez and colleagues (2020) decided to use a 5-point Likert scale due to a better adaptation to the cultural context. In this study, the psychometric properties were not impaired and the authors yielded internal consistencies between .74 ≥ α ≥ .68. Based on these results, it was decided to use a 5-point Likert scale (1 = does not describe me at all to 5 = describes me as I am) and not the 4-point scale, as proposed by the authors of the German translation and validation.

TEIQue
The Trait Emotional Intelligence Questionnaire (TEIQue) was developed by Petrides and Furnham (2001), and in the original English version, it comprises 153 items that are used to record four general dimensions in 13 sub-scales: (1) Well-being: Self-esteem, trait optimism, trait happiness; (2) Self-control: emotional regulation, stress management, impulsivity (low); (3) Emotionality: Perception of emotions, expression of emotions, trait empathy, relationships; (4) Sociability: Assertiveness, social conscience, handling of emotions. Two other facets, adaptability and self-motivation, are also included in the calculation of the global trait EI, and items are responded to on a 7-point Likert scale (1 = disagree to 7 = agree).
In addition to the original instrument, there is also a short version that only consists of 30 questions (TEIQue Short Form or TEIQue-SF). This version evaluates the same dimensions as the longer questionnaire, and Cooper and Petrides (2010) have shown its adequate psychometric properties. In the present study, we used the German short version, translated and validated by Freudenthaler and colleagues (2008) who obtained internal consistencies between .94 ≥ α ≥ .86.

Data analysis
To determine the internal consistencies, Cronbach alphas (α) were calculated and the group differences were determined, on the one hand, through the Mann-Whitney-U test (according to Rosenthal and DiMatteo (2001) and Fritz and colleagues (2012), with effect size estimates for the Mann-Whitney-U test being computed with the Pearson correlation coefficient) and, on the other hand, through the multivariate analysis of variances, or MANOVA, with Bonferroni's adjustment for several comparisons (the effect size with Eta 2 ). As outliers might reduce the representativeness of the sample (Hair et al. 2019), extreme outliers were eliminated. In each group, we recognized six outliers that gathered more than three standard deviations from their respective mean values, causing these values to be removed from the sample. Homogeneity of variance was verified through the Levene test. For all the calculations described above, SPSS 25 was used with the probability value p ≤ .05.

Demographic data and their group differences
We resolved to use control variables since apparent differences had been observed concerning the demographic data of both samples. Due to the presence of data that did not follow a normal distribution and/or data with an ordinal scale, several Mann-Whitney-U tests were performed (Dinneen & Blakesley 1973). Regarding educational levels, the distribution of both groups differed, Kolmogorov-Smirnov p < .05 and a statistically significant difference was found between residents of cohousing communities (M Rank = 157.82) and non-residents in these communities (M Rank = 115.99), U = 6,603.00, Z = -4.803, p < .001, r = .285. As to age, the distribution of both groups did not differ, Kolmogorov-Smirnov p = .304 and a statistically significant difference was found between residents of cohousing communities (Mdn = 54.50) and non-residents (Mdn = 39.00), U = 3,691.00, Z = -8.505, p < .001, r = .505. Concerning the number of years resident in the place, the distribution of both groups differed: Kolmogorov-Smirnov p < .05, and a statistically significant difference was found between residents of cohousing communities (M Rank = 149.99) and non-residents (M Rank = 129.54), U = 8,012.000, Z = -2,028, p = .042, r = .120.
Due to the reasons mentioned above, age, years resident in place and educational level were used as control variables to determine group differences concerning all the dimensions that are analyzed below.
Concerning the nominal demographic data, a chi-square test was used to compare both groups regarding gender. Two expected cell frequencies stood below 5 (33.3%), so we used a Monte Carlo simulation based on 10,000 sampled tables to compute the p value. Results showed a significant difference concerning gender, χ²(2) = 8.29, p = .012, φ = 0.18. A second chi-square test was conducted to compare both groups regarding civil status. Two expected cell frequencies were below 5 (25.0%), so we used a Monte Carlo simulation based on 10,000 sampled tables to compute the p value. Results showed a significant difference regarding civil status, χ²(3) = 9.33, p = .020, φ = 0.19.

Personality traits, trait emotional intelligence, and their group differences
First, we calculated Cronbach's alphas for reliability analysis. Although some internal consistencies were below those obtained in their original studies, we can observe in Table 2 that all the values are between moderate and high (Hinton et al. 2014).
Due to the robustness of MANOVA against violating Normality assumptions (Blanca et al. 2017;Glass et al. 1972;Harwell et al. 1992;Lix et al. 1996;Schmider et al. 2010), we only performed the Levene test and its results showed that all dimensions had homogeneity of variances.
Among the significant differences found between residents in cohousing communities and those living in common neighborhoods, we can highlight detachment, psychoticism, the PID Overall score, and well-being. Well-being should be particularly emphasized since it has shown the largest effect size. Notes. *The mean difference is significant at the .05 level and is based on estimated marginal means; b, Adjustment for multiple comparison: Bonferroni; M, mean; SD, standard deviation; CI, confidence interval; LL, lower limit; UL, upper limit.

Correlations
As data did not follow a normal distribution, Spearman correlations were computed. Table 3 shows the associations between the PID and TEIQue dimensions. Numerous significant correlations were found, so that only those moderate and strong correlations will be highlighted in which significant differences have been found between both samples. Of these dimensions, detachment stands out, since it contains the factor with the highest correlations, especially with well-being, emotionality, and the TEIQue Overall score. Second in relevance is the PID Overall score due to its correlations with well-being, self-control and the TEIQue Overall score. Finally, psychoticism showed a moderate negative correlation with self-control and the TEIQue Overall score (Cohen 1988).

Discussion and conclusion
To determine the existence of discrepancies between the inhabitants of cohousing communities and residents of common neighborhoods, 11 dimensions were evaluated. In four of them, we found significant differences showing that the sample of cohousing residents had higher levels of well-being and lower levels of detachment, psychoticism, and the overall score of maladaptive personality traits. In the remaining seven dimensions, no significant differences were found.
Although the relationship between lower levels of psychoticism and residence in a cohousing community is not so conclusive, the parallel between cohousing and decreased levels of detachment is in line with the hypothesis based on the increased number of social support exchanges (Markle et al. 2015). This correspondence is more evident when considering Krueger and colleagues' (2011) definition of detachment, characterizing it through its facets of social withdrawal, avoidance of closeness, anhedonia, depression, limited emotional experience, and mistrust. Accordingly, it is striking that no significant differences have been found concerning sociability. Furthermore, it could be shown that detachment is the dimension with the most significant negative correlation with well-being, thus forming a key trait that can hinder the achievement of increased psychological well-being.
On the other hand, it should also be mentioned that the correlations, although statistically significant, do not allow us to establish a causal relationship between the facts previously described. It is arguable, therefore, whether living together in a cohousing community stands as a key factor in the development of lower levels of detachment, or if the residents of such communities already enjoyed these traits before moving into them.
In the present study, it has been possible to go one step further towards understanding a particular characteristic of cohousing communities' inhabitants, and the decreased levels of detachment may be a factor that not only helps to increase psychological well-being but also reduces vulnerability to certain syndromes, such as avoidant, schizoid, and histrionic disorder (Kotov et al. 2017).
Despite the results found, it should be noted that these are not completely in line with previous research: Waxman's (2005) study revealed no significant differences concerning extraversion, which could be because the sample size of her investigation was considerably smaller. Likewise, it is important to bear in mind that, in the present study, a non-probability sampling was carried out and that the sample size can be considered relatively small, making it difficult to generalize its results. In addition, it should be noted that the sample was drawn online using non-probabilistic sampling, thus future studies should replicate its results in field studies.
In summary, the present study has shown that residents of cohousing communities have, from various points of view, lower levels of maladaptive personality traits and a higher level of well-being. Although these results are favorable, further research is necessary to verify whether living in a cohousing community can operate as a moderator of these traits or if its inhabitants had already enjoyed them before moving into the community. This question could be clarified through future investigations that deepen the study of such communities and evaluate the causal relationships through longitudinal studies and/or mediation analysis, as well as their effects on psychological symptoms and/or psychological well-being.