Understanding Mental Health Vulnerability Through Intersectionality: Racial Discrimination Among African Ethnic Minority Groups in Belgium

Understanding Mental Health Vulnerability Through Intersectionality: Racial Discrimination Among African Ethnic Minority Groups in Belgium To what extent do specific socio-demographic characteristics amongst Sub-Saharan African ethnic minorities, moderate the relationship of microaggressions and harassment on mental health outcomes? Word count: 6437 Paloma Brent Moerenhout Student number: 02212014 Supervisor: Prof. Dr. Sorana Toma Co-supervisor: Prof. Dr. Els Clays A dissertation submitted to Ghent University in partial fulfilment of the requirements for the degree of Master of Science in Health Promotion Academic year: 2024-2025 Understanding Mental Health Vulnerability Through Intersectionality: Racial Discrimination Among African Ethnic Minority Groups in Belgium To what extent do specific socio-demographic characteristics amongst Sub-Saharan African ethnic minorities, moderate the relationship of microaggressions and harassment on mental health outcomes? Word count: 6437 Paloma Brent Moerenhout Student number: 02212014 Supervisor: Prof. Dr. Sorana Toma Co-supervisor: Prof. Dr. Els Clays A dissertation submitted to Ghent University in partial fulfilment of the requirements for the degree of Master of Science in Health Promotion Academic year: 2024-2025 INHOUDSTAFEL FOREWORD .............................................................................................................................. 6 English ............................................................................................................................ 6 Dutch............................................................................................................................... 7 ABSTRACT ................................................................................................................................ 8 English: ........................................................................................................................... 8 Dutch:.............................................................................................................................. 9 LITERATURE STUDY .............................................................................................................. 10 Racial discrimination..................................................................................................... 10 The impact on Mental health ........................................................................................ 10 Current gap in literature ................................................................................................ 11 Discrimination and Mental Health: Group-Level Variations ......................................... 11 Gender ................................................................................................................ 12 Legal status ......................................................................................................... 12 Household income: ............................................................................................. 12 Language proficiency .......................................................................................... 13 Research question........................................................................................................ 13 METHODOLOGY ..................................................................................................................... 14 Survey Design & sampling ........................................................................................... 14 Variables ....................................................................................................................... 14 Independent variables of Racial discrimination: microaggression and harassment ............................................................................................................................. 14 Dependent variables of Mental health well-being .............................................. 15 Confounding variables: ....................................................................................... 15 Statistical Analysis ........................................................................................................ 16 RESULTS ................................................................................................................................. 18 Descriptive statistics: .................................................................................................... 18 Multivariate regression analysis: .................................................................................. 22 DISCUSSION ........................................................................................................................... 27 Key findings .................................................................................................................. 27 Limitations ..................................................................................................................... 29 Future directions ........................................................................................................... 30 Conclusion .................................................................................................................... 31 APPENDIX ............................................................................................................................. 32 A1: ................................................................................................................................. 32 Directed Acyclic Graph (DAG) ............................................................................ 32 A2: ................................................................................................................................. 32 Table 1.3: ............................................................................................................ 32 Table 2.1 ............................................................................................................. 33 Table 2.4: ............................................................................................................ 34 Table 2.5: ............................................................................................................ 34 Table 2.6: ............................................................................................................ 35 A3: ................................................................................................................................. 35 Table 3.13.8: ..................................................................................................... 35 A4 : ................................................................................................................................ 40 Table 3.3-3.4 ....................................................................................................... 40 REFERENCES ......................................................................................................................... 42 FOREWORD English Writing this thesis and exploring this field of research has been an exceptionally enriching experience for me. It has undoubtedly provided me with many skills that I will apply and use throughout my future career. From reading to writing, analysing to reporting, and from understanding to articulating complex scientific concepts to the outside world. I continue to exceed my own expectations by accomplishing this alongside my main profession as a therapist in psychiatry and the artistic ambitions I pursue as a secondary occupation. I would first and foremost like to express my gratitude to my supervisor Sorana Toma, who guided me exceptionally well throughout this entire process. She consistently maintained a critical perspective and read my work with great attention. Her feedback and insistence on clear core arguments have greatly enriched my academic skills. The career she has built and the way she delegates all the work while remaining involved in mine, shows a remarkable commitment to both academic excellence and her mentees. Besides, I received tremendous support in the final stages of my thesis from her PhD researcher Sarah Derveeuw, who was always quick and thorough in assisting me. My cosupervisor, Els Clays, also provided a critical perspective on my work and supported me whenever needed, for which I am grateful. Lastly, I want to thank my partner, Dana Sprung, who consistently helped me manage the divergence of my mind and therefore the realisation of this complex topic. She is an enormous contribution to my academic development. Paloma Brent Moerenhout Dutch Het schrijven van deze thesis en het verkennen van dit onderzoeksveld is voor mij een bijzonder verrijkende ervaring geweest. Het heeft mij ongetwijfeld veel vaardigheden bijgebracht die ik in mijn toekomstige carrière zal toepassen en benutten. Van lezen tot schrijven, van analyseren tot rapporteren, en van het begrijpen tot het verwoorden van complexe wetenschappelijke concepten naar de buitenwereld toe. Ik blijf mijn eigen verwachtingen overtreffen door dit te realiseren naast mijn hoofdberoep als therapeut in de psychiatrie en de artistieke ambities die ik in bijberoep nastreef. Allereerst wil ik mijn dank uitspreken aan mijn promotor, Sorana Toma, die mij gedurende dit hele proces op uitzonderlijke wijze heeft begeleid. Zij hield steeds een kritische blik en las mijn werk met grote aandacht. Haar feedback en haar nadruk op duidelijke kernargumenten hebben mijn academische vaardigheden enorm verrijkt. De carrière die zij heeft opgebouwd en de manier waarop zij al het werk delegeert terwijl zij toch betrokken blijft bij mijn traject, tonen een opmerkelijke toewijding aan zowel academische excellentie als haar ondergeschikten. Daarnaast kreeg ik in de laatste fase van mijn thesis enorm veel steun van haar doctoraatsonderzoeker, Sarah Derveeuw, die mij steeds snel en grondig bijstond. Mijn copromotor, Els Clays, bood eveneens een kritische blik op mijn werk en ondersteunde mij waar nodig, waarvoor mijn oprechte dank. Tot slot wil ik mijn partner, Dana Sprung, bedanken, die mij telkens weer hielp om de divergentie van mijn denken in goede banen te leiden, en daarmee bijdroeg aan de realisatie van dit complexe onderwerp. Zij is een enorme bijdrage voor mijn academische ontwikkeling Paloma Brent Moerenhout ABSTRACT English: Objective: Racial discrimination in both overt and covert form remains a critical but underexamined determinant of mental health, especially among ethnic minority immigrant populations. The distinct impacts of microaggressions and harassment and their interplay with intersecting social identities, remain under-examined. This study investigates how microaggressions and harassment relate to mental health outcomes among sub-Saharan African minorities in Belgium, and whether gender, legal status, income, and language proficiency moderate these associations. Method: Data is analysed from the 2024–2025 ROAM-BE cross-sectional survey (N=923). Discrimination experiences were measured using multi-item scales for microaggressions and harassment. Mental health outcomes included general wellbeing, anxiety, sleep quality, and life satisfaction. Multivariate regression models tested the main effects of discrimination and two-way interactions with each moderator, adjusting for key covariates. Results: Both forms of discrimination were significantly associated with increased anxiety, reduced sleep quality, and lower life satisfaction, following a dose-response pattern. Interaction models revealed that moderate language proficiency intensified the negative impact of harassment on sleep and life satisfaction, while low or high proficiency showed weaker associations. Financial insecurity predicted poorer mental health across all models, though its moderating role was limited. Legal insecurity and female gender strengthened the effect of microaggressions on anxiety. In contrast, harassment had a stronger impact on sleep and life satisfaction among women, suggesting heightened vulnerability to overt threats. Conclusion: These findings emphasize that discrimination’s psychological impact cannot be fully understood in isolation but emerges from its entanglement with intersecting identities and structural inequalities. Dutch: Doelstelling: Raciale discriminatie, zowel in openlijke als subtiele vormen, blijft een cruciale maar onderbelichte determinant van mentale gezondheid, vooral onder etnische minderheidsgroepen met een migratieachtergrond. De verschillende effecten van microagressies en intimidatie, en hun wisselwerking met elkaar overlappende sociale identiteiten, zijn tot op heden onvoldoende onderzocht. Deze studie onderzoekt hoe microagressies en intimidatie samenhangen met mentale gezondheidsuitkomsten bij SubSahara Afrikaanse minderheden in België, en of gender, verblijfsstatus, inkomen en taalvaardigheid deze verbanden modereren. Methode: De gegevens zijn afkomstig uit de ROAM-BE cross-sectionele survey van 2024–2025 (N=923). Discriminatie werden gemeten met behulp van meerdere items voor microagressies en intimidatie. De mentale gezondheidsuitkomsten omvatten algemeen welzijn, angst, slaapkwaliteit en levenstevredenheid. Multivariate regressiemodellen toetsten zowel de hoofdeffecten van discriminatie als tweezijdige interacties met elke moderator, gecontroleerd voor relevante covariabelen. Resultaten: Beide vormen van discriminatie waren significant geassocieerd met verhoogde angst, verminderde slaapkwaliteit en lagere levenstevredenheid, volgens een duidelijk dose-respons patroon. Interactiemodellen toonden aan dat matige taalvaardigheid het negatieve effect van intimidatie op slaap en levenssatisfactie versterkte, terwijl lage of hoge taalvaardigheid zwakkere verbanden vertoonde. Financiële onzekerheid voorspelde consequent slechtere mentale gezondheid, hoewel het slechts beperkt als moderator fungeerde. Een onzekere verblijfsstatus en vrouwelijk geslacht versterkten het effect van microagressies op angst. Daarentegen had intimidatie een sterkere impact op slaapkwaliteit en levenstevredenheid bij vrouwen, wat duidt op verhoogde kwetsbaarheid voor openlijke bedreigingen. Conclusie: Deze bevindingen benadrukken dat de psychologische impact van discriminatie niet los kan worden gezien van sociale identiteiten en structurele ongelijkheden, maar daar juist uit voortkomt. Aantal woorden artikel: 6437 (exclusief woord vooraf, inhoudstafel, inleiding, abstract, bijlagen, tabellen en referentielijst) • Aantal woorden inleiding: 1495 • Aantal woorden methodologie: 1593 • Aantal woorden resultaten: 1515 • Aantal woorden discussie: 1843 covariates were tested as moderators to assess how aspects of identity shape mental health outcomes after racial discrimination. Gender was measured as a binary variable (1=male, 2=female). Legal status was derived from responses to “How long is your current residence permit valid for?” and recoded into three categories: 1=no residence permit or less than 1 year; 2=temporary residence permit of more than 1 year; 3=Belgian nationality or exempt from needing a permit. Household income was measured by perceived financial ease: 1=living comfortably on current income, 2=coping on current income , 3=finding it difficult on current income and 4=finding it very difficult on current income with categories 3 and 4 merged for analytical balance. Education level was based on the highest diploma obtained in or outside Belgium, originally ranged from 1 (primary) to 6 (doctoral degree) but recoded into four categories: 1=primary education or less, 2=secondary education, 3=short-cycle higher education, and 4=long-cycle higher education (master’s/doctoral).This recoding allowed for simplified comparisons across diverse educational systems. Language proficiency was assessed for French, Dutch, and English based on self-rated speaking and reading/writing skills. For each language, speaking and literacy scores were combined and categorized as: little to none, intermediate, good and very good proficiency. To reflect regional language realities, a final variable was created: for participants living in Dutch-speaking regions, Dutch proficiency was used; for those in French-speaking regions, French proficiency was used; and for participants in Brussels, the best score across French and Dutch was selected. Generation of immigrants is categorised as 1, 1.5 and 2nd generation. Age is categorised into tertiles: 1829= young adults; 30-44= middle-aged adults ; 45+ =older adults. Francophone countries are those where French is an official or widely spoken language, including parts of Europe (e.g., France, Belgium) and Africa (e.g., Senegal, Ivory Coast). The OIF defines its members as states sharing the French language and common values (Glasze, 2007) Religion is categorised as Catholic, Protestant, Revival/Evangelical/Pentecostal churches; Muslim; No Religion and Other. Statistical Analysis All analyses were conducted using IBM SPSS 29. Post-stratification weights were applied to align the sample with the known population distribution of individuals of sub-Saharan African origin in Belgium, based on country of origin, age, sex, region, generation (first, 1.5, second), and nationality, ensuring greater representativeness. Cases with missing values on the relevant variables were excluded from the analyses. The analytic strategy proceeded in three stages. First, preliminary analyses summarized key variables using descriptive statistics (means, SDs for continuous; frequencies, percentages for categorical), and Pearson correlations examined associations. Distributions were assessed via boxplots and histograms to identify skewness or outliers and finalize covariate inclusion. Second, one-way ANOVAs and chi-square tests identified which covariates predicted lower mental health scores and greater exposure to microaggressions or harassment (categorised format). Post-hoc Tukey tests assessed differences across exposure levels. For harassment, Mann–Whitney U tests were used due to unequal group variances (violated Levene’s test for homogeneity of variances). Bonferroni correction (p <.009) adjusted for multiple comparisons. For microaggressions, group sizes were relatively equal and large (N>200), and given ANOVA’s robustness under such conditions, parametric testing remained appropriate for this variable. This combined approach of parametric and non-parametric test ensures a transparent and robust analysis. Third, bivariate and multivariate regression models estimated the impact of microaggressions and harassment on all mental health indicators, controlling for possible confounding (gender, age, generation, subregion, legal status, francophone origin, education, income, regional language proficiency, and religion). All models were checked for assumptions, including multicollinearity, linearity, homoscedasticity, normality of residuals, independence of observations, and the absence of influential or outlying cases. Harassment and microaggression scores were treated as continuous variables to preserve superior statistical power. Further, separate regressions with two-way interaction terms tested whether gender, legal status, language proficiency, and household income moderated the effects of microaggressions and harassment. These moderators were selected based on theoretical relevance and prior links to mental health outcomes. Each model includes the main effect of harassment or microaggression, the main effect of the moderator, and the corresponding interaction term (e.g. harassment × gender). Interaction terms were entered one at a time to enhance statistical power, reduce multicollinearity, and allow for isolated interpretation. Residual diagnostics confirmed that assumptions for linear regression were sufficiently met. The most socially advantaged category (men, Belgian nationality, high language proficiency, living comfortably) served as the reference group in all dummy coding. Interaction terms indicate whether this association significantly differs across other subgroups. RESULTS Descriptive statistics: Table 1.1: Socio-demographic characteristics (unweighted), ROAM-BE survey (n=923) Sample size N Sample size % Age (in years) 18-29 (Young adults) 314 34% 30-44 (Middle-aged adults) 416 45,1% 45+ (older adults) 193 20,9% Gender Male 463 50,2% Female 460 49,8% Generation 1st 636 68,9% 1,5nd 106 11,5% 2nd 181 19,6% Duration of stay in Belgium Les then 5 years 268 29% 5 to 10 years 142 15,4% 10 years or more 332 36% Born in Belgium 181 19,6% Subregion living in Belgium Brussels 276 29,9% Flanders 333 36,1% Wallonia 314 34% Francophone origin Francophone country 751 81,4 Other 172 18,6 Legal status / Residence permit No residence permit or permit of 1yr or less 263 28,5% Residence permit of > 1 yr 176 19,2% Belgian nationality 479 52,2% Missing 5 Household standard of living Living comfortable 148 16,2% income is sufficient 385 42,1% Very difficult to copy with income 381 41,7% Language proficiency per region Very good 494 53,6% Good 178 19,3% Moderate 88 9,5% Little to none 161 17,4% missing 2 Educational degree Primary or less 129 14 Secondary 208 22,5 Higher short education 255 27,6 Higher long education 331 35,9 Religion catholic 301 32,6% protestant 133 14,4% Revivalist, evangelical, Pentecostal 133 14,4% Muslim 187 20,3% No religion 101 10,9% Other 67 7,3% missing 1 Table 1.2: Descriptives of dependent variables of mental health (unweighted) total N =923 N Mean / % SD Min Max Missing WHO-5 wellbeing (100 = highest score) 921 54.96 23.45 0 100 2 Poorer well being (≤50) 417 45.2 - - - Greater well being (>50) 504 54.6 - - - - PHQ-2 anxiety (0 = Lowest anxiety ; 6 = Highest level of anxiety) 921 1.24 1.57 0 6 2 Sleep Quality (0= "very bad" and 10= "excellent") 918 7.57 1.52 0 10 5 Life satisfaction (0 ="not at all satisfied" and 10= "totally satisfied") 920 6.96 1.94 0 10 3 Table 1.1 presents the ROAM-BE sample (N=923), which was gender-balanced and mostly composed of young to middle-aged adults. The majority were first-generation migrants (69%) with francophone backgrounds (81%), and 71% had lived in Belgium for over five years. Over half (52%) held Belgian nationality. Financial hardship was common, with only 16% living comfortably. Most rated their language proficiency as good or very good (73%) and 64% had higher education. Catholics (33%) and Muslims (20%) were the largest religious groups. Table 1.2 shows moderate overall mental health: WHO-5 well-being averaged 54.96, with 45% scoring below the cut-off; anxiety was low (M=1.24), sleep quality high (M=7.57), and life satisfaction relatively high (M=6.96). Missing data were minimal. Among those who experienced recent discrimination (N=641), around one-third reported stress (28.2%), frequent rumination (27.6%), and avoidance behaviours (29.3%). Fewer reported sleep disruption (7.3%) or loss of confidence (9.3%). Notably, 43.9% mentally prepared for such situations, while 55.2% said they didn’t worry (Appendix A2, Table 1.3). Table 2.1 (Appendix A2) showing that sociodemographic factors such as gender, legal status, household income, migration generation and language proficiency were strongly associated with variations in life satisfaction, anxiety, sleep quality, and wellbeing scores. In contrast, education and age showed weaker or non-significant associations. Table 2.2: Associations between frequency of discrimination (unequal treatment, harassment, and microaggressions) and mental well-being indicators: Life satisfaction, sleep quality, anxiety, and WHO-5 wellbeing Life satisfaction average (Max score = 10) Quality of sleep (Max score = 10) Anxiety (max score = 6) Wellbeing WHO5 (max score = 100) Lower wellbeing (0-50) % Higher wellbeing (50<) % Unweighted frequencies Unweighted percentages Experienced unequal treatment/ discrimination in Belgium F= 11,570 ; p=<,001*** F = 7,302 ; p= <,001*** F = 17,034 ; p= <,001*** F = 1,558 ; p=,198 X² = 0.948 ; p=,814 N=906 ; missing = 17 100% Never 7,93 7,28 0,7973 53,6557 46,60% 53,40% N=367 40,50% Once 7,65 7,05 1,5013 58,5154 41,90% 58,10% N=144 15,90% Sometimes 7,33 6,73 1,4364 54,363 45,60% 54,40% N=285 31,50% Often 6,98 6,48 1,767 54,567 46,10% 53,90% N=110 12,10% Experienced unequal treatment in last 5 years F=5,972 ; p=<,001*** F = 5,691 ; p=<,001*** F= 3,023 ; p= .029* F = ,222 ; p=,881 X² = 0.636 ; p=,888 N=534 ; missing=384 100% Never 8 7,34 1,3081 54,5652 46,60% 53,40% N=55 10,30% Once 7,59 6,98 1,367 56,7945 41,90% 58,10% N=136 25,50% Sometimes 7,22 6,73 1,49 54,929 45,60% 54,40% N=233 43,60% Often 6,94 6,19 1,9307 55,7272 46,10% 53,90% N=110 10,60% Harassment (overt forms of discrimination): F = 20,348 ; p<,001*** F = 23,409 ; p<,001*** F = 25,910 ; p <,001*** 54,75 ; F = ,918 ; p = ,432 X² = 1,697 ; p=.638 N=860 ; missing=63 100% No harassment experience in the last 5 years or never 7,95 7,47 0,78 54,94 45,50% 54,50% N=419 47,10% 1 type of harassment exp 7,36 6,7 1,48 56,91 44,90% 55,10% N=205 23,00% 2 types of harassment exp 7,2 6,2 1,71 52,28 49,40% 50,60% N=179 20,10% 3 or more types 6,92 6,55 1,92 55,94 41,10% 58,60% N=87 9,80% Microaggressions (covert forms of discrimination): F = 17,324 ; p<,001*** F = 17,255 ; p<,001*** F = 18,320 ; p <,001*** 54,62 ; F = ,923 ; p = ,398 X² = 3,193 ; p=.203 N=860 ; missing=63 100% No micro-aggressions in the last 5 years or never 8 7,5 0,77 52,71 51,20% 48,80% N= 211 24,50% 1 or 2 types of microaggression 7,58 7,03 1,25 55,08 44,30% 55,70% N= 307 35,70% 3 or more types 7,23 6,53 1,6 55,39 44,00% 56,00% N= 342 39,80% *p < .05. **p < .01. ***p < .001 (two-tailed test) In Table 2.2, experiences of unequal treatment or discrimination showed significant main effects for life satisfaction (F=11.57, p<.001), sleep quality (F=7.30, p<.001), and anxiety (F=17.03, p<.001), indicating that more frequent experiences of discrimination were associated with worse outcomes across all three indicators. Respondents who reported being treated unequally “often” had the lowest life satisfaction (M=6.98), poorest sleep quality (M=6.48), and highest anxiety (M=1.77), compared to those who reported “never” experiencing unequal treatment. However, no significant differences were found for WHO-5 well-being scores (F=1.558, p=.198) or its categorical classification (χ²=0.948, p=.814). A similar pattern was observed for recent discrimination in past 5 years (I4), with significant effects for life satisfaction (F=5.97, p<.001), sleep quality (F=5.69, p<.001), and anxiety (F=3.02, p=.029). Those with frequent recent experiences (“often”) scored lower on life satisfaction (M=6.94), sleep (M=6.19), and had higher anxiety (M=1.93), though no differences were observed for WHO-5 well-being (F=0.222, p=.881; χ²=0.636, p=.888). Harassment was also strongly associated with worse psychological outcomes. Life satisfaction (F=20.35, p<.001), sleep quality (F=23.41, p<.001), and anxiety (F=25.91, p<.001) declined significantly with increasing types of harassment experienced. Those reporting three or more types of harassment had the lowest scores for life satisfaction (M=6.92), sleep (M=6.55), and the highest anxiety (M=1.92), though no significant effects emerged for WHO-5 (F=0.918, p=.432; χ²=1.697, p=.638). Finally, microaggressions were significantly associated with declines in life satisfaction (F=17.32, p<.001), sleep quality (F=17.25, p<.001), and anxiety (F=18.32, p<.001), with outcomes worsening progressively across exposure levels. However, WHO-5 well-being remained unaffected (F=0.923, p=.398; χ²=3.193, p=.203). Table 2.3: Chi-square test looking at frequency in types of microaggressions and harassments across sample characteristics (covariates) Variable Microaggression (3 cat) χ² p-value (microagression) Harassment (4cat) χ² p -value (harassment) Gender 3.189 0.203 5.095 0.165 Age 18.034 0.001** 9.566 0.144 Generation 16.146 0.003** 8.307 0.216 Subregion of residence 21.888 <0.001*** 9.377 0.153 Legal Status 5.163 0.271 4.810 0.568 Francophone Origin 9.388 0.009** 1.104 0.776 Highest educational degree 5.099 0.531 18.183 0.033* Household income 13.642 0.009** 9.317 0.157 Language Proficiency 17.638 0.007** 16.944 0.050* Religion 26.455 0.003** 11.760 0.697 *p < .05. **p < .01. ***p < .001 (two-tailed test). The chi-square test results in table 2.3 indicate that experiences of microaggressions significantly vary by age (p=0.001), generation (p=0.003), subregion of residence (p<0.001), Francophone origin (p=0.009), household income (p=0.009), language proficiency (p=0.007), and religion (p=0.003). In contrast, experiences of harassment show significant variation only by highest educational degree (p=0.033) and language proficiency (p=0.050). Findings from table 2.4 and 2.5 (Appendix A2) revealed robust and statistically significant associations between increased exposure to both microaggressions and harassment and poorer outcomes across anxiety, life satisfaction, and sleep quality (all p<.001). The tables demonstrate a clear dose-response relationship with mental health outcomes, with individuals exposed to three or more types of microaggressions or harassment reported the poorest mental health indicator outcomes (p<.001). No significant differences were observed for WHO-5 wellbeing scores across any exposure levels in either tables. In table 2.6 (Appendix A2.6) The non-parametric test for harassment aligned with the original findings, reinforcing the validity of observed effects. Even one type of harassment is linked to a significant, immediate decline in mental health: participants reporting one type had higher anxiety (Z=-6.633, p<.001), lower life satisfaction (Z=-5.246, p< 001), and poorer sleep (Z=-5.256, p<.001) than those with no harassment. the largest gap consistently lies between those with no harassment and those with just one type. No significant group differences were found for WHO-5 well-being, confirmed by non-significant ANOVA and post hoc tests. Multivariate regression analysis: Table 3.1: Multivariate regression analysis estimating effect among microaggression and harassment on primary mental health outcome: WHO-5 Wellbeing (N=908) Model 1a (Unadjusted) Model 1b (Adjusted) b SE p-value b SE p-value Wellbeing Microaggressions 0,07 0,12 0,575 0,12 0,1 0,35 Harassment 0,12 0,19 0,541 0,19 0,2 0,32 Life satisfaction Microaggression -0,04 0,01 <.001*** -0,03 0,01 <.001*** Harassment -0,08 0,01 <.001*** -0,07 0,1 <.001*** Sleep Quality Microaggression -0,06 0,01 <.001*** -0,05 0,01 <.001*** Harassment -0,1 0,02 <.001*** -0,09 0,1 <.001 *** Anxiety Microaggression 0,04 0.01 <.001*** 0,04 0,01 <.001*** Harassment 0,09 0,01 <.001*** 0,1 0,1 <.001*** Note: Only microaggressions and harassment coefficients are shown; Full model outputs available in Appendix Table A3. Model 1b adjusted for age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. *p < .05. **p < .01. ***p < .001 In the unadjusted models (Model 1a), neither microaggressions (b=0.07, p=.575) nor harassment (b=0.12, p=.541) were significantly associated with wellbeing. After adjusting for sociodemographic and socioeconomic covariates (Model 1b), these associations remained non-significant for both microaggressions (b=0.12, p=.350) and harassment (b=0.19, p=.322). As a result, no moderation analysis was conducted for the wellbeing variable. Microaggressions and harassment were significantly associated with all three secondary outcomes. In both unadjusted and adjusted models, higher levels of microaggressions were significantly associated with lower life satisfaction (adjusted: b=–0.03, p<.001), poorer sleep quality (b=–0.05, p<.001), and higher anxiety (b=0.04, p<.001). Similarly, harassment was consistently associated with decreased life satisfaction (adjusted: b=–0.07, p<.001), poorer sleep (b=–0.09, p<.001), and greater anxiety (b=0.10, p<.001). Some violations of assumptions were observed: particularly small heteroscedasticity due to positive skewness and influential observations. Models were tested excluding influential observations (Std. residuals ≥–3/3) and showed the same significance. No multicollinearity or non-linearity was found across all models (full adjusted models in Appendix A3, Table 3.1–3.8). Across all bivariate and multivariate regression models, household income (living comfortably), gender (female), and legal status (short/no permit) emerged as the most consistent predictors of mental health outcomes when controlling for microaggression, harassment, and other covariates. These factors significantly explained variance in wellbeing, anxiety, sleep quality, and life satisfaction. Individuals reporting higher income consistently showed better outcomes across all domains, with large positive coefficients (B=6.95 for wellbeing, p<.001; B=0.85 for sleep, p<.001). In contrast, female participants and individuals with short legal status consistently reported worse mental health indicators (B=–7.47 for wellbeing among females, p<.001; B=0.57 for anxiety among those with no/short-term permits, p<.001). Strong language proficiency and 1.5 generation status were also associated with better mental health in several models, though the strength and consistency varied. Given the strength and consistency of these associations and their theoretical relevance based on prior literature, these four variables were included in the two-way interaction models to explore whether they moderated the relationship between discriminatory experiences and mental health. Table 3.3: Two-way interaction multivariate regression models of mental health outcomes (anxiety, sleep quality, and life satisfaction) by microaggression exposure and moderators (gender, legal status, language proficiency per region, and household income). (N=908) Anxiety Sleep quality Life satisfaction All two-way interactions for microaggressions: b SE p-value 95% CI b SE p-value 95% CI b SE p-value 95% CI Microaggression x female -0.04 0.02 0.007** [-0.08, -0.01] -0.01 0.02 0.583 [-0.05, 0.03] -0.01 0.02 0.399 [-0.04, 0.02] Microaggression x male (reference) — — — — — — — — — — — — Microaggression x No work permit or work permit of 1 year or less -0.05 0.02 0.023** [-0.08, -0.01] -0.01 0.02 0.633 [-0.06, 0.04] -0.03 0.02 0.15 [-0.06, 0.01] Microaggression x Work permit of more than 1 year -0.02 0.02 0.446 [-0.06, 0.03] 0.02 0.03 0.495 [-0.04, 0.07] 0.00 0.02 0.972 [-0.04, 0.04] Microaggression x Belgian nationality / no need for permit (reference) — — — — — — — — — — — — Microaggression x little to none -0.03 0.02 0.221 [-0.07, 0.02] 0.03 0.03 0.306 [-0.03, 0.08] 0.01 0.02 0.585 [-0.03, 0.06] Microaggression x moderate -0.04 0.03 0.132 [-0.09, 0.01] 0.02 0.03 0.525 [-0.04, 0.09] -0.03 0.03 0.281 [-0.08, 0.02] Microaggression x good 0.03 0.02 0.179 [-0.01, 0.07] 0.01 0.03 0.623 [-0.04, 0.07] -0.01 0.02 0.765 [-0.05, 0.03] Microaggression x very good (reference) — — — — — — — — — — — — Microaggression x Difficult or very difficult with current income -0.04 0.02 0.067° [-0.09, 0.00] 0.03 0.03 0.322 [-0.03, 0.08] 0.001 0.02 0.948 [-0.04, 0.04] Microaggression x Sufficient income -0.03 0.02 0.249 [-0.07, 0.02] 0.003 0.03 0.927 [-0.05, 0.06] 0.003 0.02 0.885 [-0.04, 0.05] Microaggression x Living comfortable (reference) — — — — — — — — — — — — Note: Interaction effects are interpreted at conventional and marginal levels of statistical significance. Specifically, results are considered: p < .001: highly significant*** ; p < .01: significant ** ; p < .05: moderately significant* ; p ≤ .10: marginally significant (trend-level)°. Findings with p ≤ .10 are discussed when theoretically relevant or consistent across models. All models were adjusted for potential confounders, including age, generation, gender, legal status, subregion of residence, Francophone origin, highest educational attainment, language proficiency, household income, and religion. Each two-way interaction was tested in a separate model. This table shows a comprehensive overview. Full table with main effect for microaggression, interaction effects by interaction term in appendix (A3 table 3.3) Table 3.4: Two-way interaction multivariate regression models of mental health outcomes (anxiety, sleep quality, and life satisfaction) by harassment exposure and moderators (gender, legal status, language proficiency per region, and household income). (N=908) Anxiety Sleep quality Life satisfaction All two-way interactions for harassment: b SE p-value 95% CI b SE p-value 95% CI b SE p-value 95% CI Harassment x female 0.00 0.03 0.994 [-0.05, 0.05] -0.06 0.03 0.054° [-0.12, 0.00] -0.04 0.02 0.070° [-0.09, 0.00] Harassment x male (reference) — — — — — — — — — — — — Harassment x No work permit or work permit of 1 year or less -0.10 0.03 <0.001*** [-0.16, -0.04] 0.02 0.04 0.599 [-0.05, 0.09] -0.04 0.03 0.144 [-0.10, 0.01] Harassment x Work permit of more than 1 year -0.05 0.04 0.158 [-0.13, 0.02] 0.04 0.05 0.432 [-0.06, 0.13] 0.03 0.04 0.435 [-0.04, 0.10] Harassment x Belgian nationality / no need for permit (reference) — — — — — — — — — — — — Harassment x little to none language proficiency -0.01 0.03 0.850 [-0.07, 0.06] 0.02 0.04 0.626 [-0.06, 0.10] -0.06 0.03 0.083 [-0.12, 0.01] Harassment x moderate language proficiency 0.06 0.05 0.187 [-0.03, 0.16] -0.12 0.06 0.036** [-0.24, -0.01] -0.14 0.05 0.002** [-0.22, -0.05] Harassment x good language proficiency - 0.05 0.03 0.111 [-0.01, 0.12] -0.01 0.04 0.711 [-0.10, 0.07] -0.05 0.03 0.108 [-0.11, 0.01] Harassment x very good language proficiency (reference) — — — — — — — — — — — — Harassment x Difficult or very difficult with current income -0.06 0.04 0.094° [-0.13, 0.01] -0.06 0.03 0.050* [-0.12, 0.00] 0.03 0.03 0.381 [-0.04, 0.09] Harassment x Sufficient income -0.05 0.04 0.190 [-0.12, 0.02] -0.06 0.03 0.050* [-0.12, 0.00] -0.04 0.03 0.262 [-0.11, 0.03] Harassment x Living comfortable (reference) — — — — — — — — — — — — Note: Interaction effects are interpreted at conventional and marginal levels of statistical significance. Specifically, results are considered: p < .001: highly significant*** ; p < .01: significant ** ; p < .05: moderately significant* ; p ≤ .10: marginally significant (trend-level)°. Findings with p ≤ .10 are discussed when theoretically relevant or consistent across models. All models were adjusted for potential confounders, including age, generation, gender, legal status, subregion of residence, Francophone origin, highest educational attainment, language proficiency, household income, and religion. Each two-way interaction was tested in a separate model. This table shows a comprehensive overview. Full table with main effect for microaggression, interaction effects by interaction term in appendix (A3 table 3.4) Across all outcomes of anxiety, sleep quality, and life satisfaction; exposure to microaggressions or harassment was significantly associated with poorer mental health. However, interaction effects varied by moderator and outcome. Microaggression Gender significantly moderated the effect of microaggressions on anxiety, with a weaker association among women (b=–0.04, p=.007). No such interaction was observed for sleep (b=–0.01, p=.583) or life satisfaction (b=–0.01, p=.399). Legal status moderated anxiety outcomes: participants with no or short-term permits showed a weaker association between microaggressions and anxiety (b=–0.05, p=.023). No interaction effects were found for sleep (b=–0.01, p=.633) or life satisfaction (b=–0.03, p=.150). Language proficiency showed no significant interaction effects across anxiety (b=–0.03, p=.221), sleep (b=0.03, p=.306), or life satisfaction (b=0.01, p=.585), though it was strongly associated with worse mental health in main effects. For instance, individuals with little to no proficiency reported significantly poorer sleep (b=–1.00, p=.002) and greater anxiety (b=0.63, p=.015). Household income did not significantly moderate the effect of microaggressions: interaction terms were marginally significant for anxiety (b=–0.04, p=.067), and non-significant for sleep (b=0.03, p=.322) and life satisfaction (b=0.001, p=.948). However, income level was significantly associated with all outcomes in main effects. Participants with income difficulties reported higher anxiety (b=0.83, p<.001), poorer sleep quality (b=–1.08, p<.001), and lower life satisfaction (b=–0.73, p=.001) compared to those living comfortably. Harassment For anxiety, harassment exposure predicted significantly higher symptoms (b=0.10, p<.001). A significant interaction was observed for legal status: individuals with no or short-term work permits showed a weaker association between harassment and anxiety (b=–0.10, p=.001). Moderated effects for women were seen for harassment on sleep quality (b=–0.06, p=.054°) and life satisfaction (b=–0.04, p=.070°). No significant moderating effects were found for language proficiency. For household income, a marginal effect was observed for difficult or very difficult income (b=–0.06, p=.094). Still, female gender (b=0.32, p=.017) and low income (b=0.71, p<.001) were independently associated with higher anxiety in main effects. For sleep quality, harassment was associated with poorer outcomes (b=–0.06, p=.006), with moderation emerging across multiple domains. A significant interaction was found for moderate language proficiency (b=–0.12, p=.036), suggesting those with moderate skills experienced sharper declines in sleep quality when harassed. A marginal interaction with gender (b=–0.06, p=.054) indicated a potential heightened vulnerability to poorer sleep quality among women. Household income significantly moderated the effect of harassment, with significant interactions for both difficult or very difficult income (b=0.03, p=.050) and sufficient income (b=0.03, p=.050). While APPENDIX A1: Directed Acyclic Graph (DAG) A2: Table 1.3: Direct effects of discriminatory effects on mental health Table 1.3: Direct effects of discriminatory effects on mental health (total sample N= 641 (unweighted) ; 278 did not report recent experiences of discrimination) Effect N (Yes) % Yes N (No) % No Missin g This has caused you a lot of stress 182 28.2% 462 71.6% 278 You think about it often 178 27.6% 467 72.4% 278 It keeps you awake 47 7.3% 598 92.7% 278 You've lost your self-confidence 60 9.3% 585 90.7% 278 You don't worry about it 356 55.2% 289 44.8% 278 You avoid this type of situation (friendships, relationships, certain places) 189 29.3% 456 70.7% 278 You try to prepare yourself mentally for this type of situation 283 43.9% 362 56.1% 278 Table 2.1 Mental health Well being scores on socio-demographics / covariates Life satisfaction average (Max score = 10) Quality of sleep (Max score = 10) Anxiety (max score = 6) Wellbeing WHO5 (max score = 100) Wellbeing WHO-5 Mean ; F ; p-value Mean ; F ; p-value Mean ; F ; p-value Mean ; F ; p-value Lower wellbeing (0-50) % Higher wellbeing (50<) % Chi² and p-value unweighted frequencies Sexe F = 8,576 ; p=,003** F=6,184 ; p=,013* f = 10,803 p=,001** F = 20,250 ; p= <,001*** 45.3% 54.7% X² = 13.159 ; p=<.001*** N= 923 Hommes 7.72 7,12 1,06 58,5067 39,10% 60,90% N= 463 Femmes 7.43 6,8 1,4 51,6202 51,10% 48,90% N= 460 Age F = 1,514 ; p=,221 F= 1,159 ; p=,314 f = ,372 ; p=,690 F = ,425 ; p= ,654 45.3% 54.7% X² = 3.359 ; p=,186 N=923 18-29 7,71 6,86 1,28 55,9397 40,70% 59,30% N=314 30-49 7,51 6,95 1,19 54,797 46,80% 53,20% N=416 50+ 7,54 7,14 1,29 53,8996 48,30% 51,70% N=193 Migration generation F = 9,801 ; p= <,001*** F= 1,951 ; p=,143 F= 1,753 ; p=,174 F = 6,349 ; p=,002** 45.3% 54.7% X² = 15.147 ; p=<.001*** N=923 First generation 7,48 6,93 1,26 53,5829 48,70% 51,30% N=636 1.5 gen 7,46 6,74 1,41 62,3059 29,00% 71,00% N=106 Second generation 8,07 7,2 1,05 56,1606 40,80% 59,20% N=181 Region of residence F = 5,593 ; p = ,004** F = 4,236 ; p= ,013* F =1,376 ; p =,253 F = ,111 ; p = ,895 45.3% 54.7% X² = 1.203 ; p=,548 N=923 Brussels 7,38 6,82 1,1215 54,9241 45,30% 54,70% N=276 Wallonia 7,48 6,78 1,3447 54,4962 47,70% 52,30% N=333 Flanders 7,77 7,18 1,2275 55,3533 43,40% 56,60% N=314 Legal Status F = 22,241 ; p=<,001*** F = 9,984 ; <,001*** F = 5,714 : p= ,003** F = 7,603 ; <,001*** 45.2% 54.8% X² = 14.691 ; p=<.001*** N=918 No residence permit or 1 year or less 6,97 6,46 1,5644 49,3807 56,70% 43,30% N=263 Residence permit of more than a year 7,6 6,9 1,2145 57,2464 38,50% 61,50% N=176 Belgian nationality 7,79 7,16 1,1298 56,4325 42,90% 57,10% N=479 Origine (francophone ou pas) F = 6,847 ; p=,009** F = 5,122 ; p=,024* F= ,177 ; p=,674 F = ,019 ; p= ,890 45.3% 54.7% X² = 0.435 ; p=,509 N=923 Francophone origin country 7,5 6,88 1,2483 55,0152 45,80% 54,20% N=751 Other African origin 7,82 7,23 1,1952 54,7549 43,20% 56,80% N=172 Highest educational attainment F=2,934 ; p=,033* F= 1,616 ; p=,184 F= 1,504 ; p=,212 F = ,382 ; p=,766 45.2% 54.8% X² = 2.560 ; p=,464 N=923 Primary or less 7,5 7,26 1,4136 56,4159 45,50% 54,50% N=129 Secondary 7,84 6,82 1,2831 53,7336 43,30% 56,70% N=208 Short Tertiary 7,48 6,98 1,2687 55,2485 42,70% 57,30% N=255 Long Tertiary 7,49 6,89 1,0903 54,9125 48,80% 51,20% N=331 Household standard of living F = 57,085 ; p=<,001*** F= 16,951 ; p=<,001*** F = 26,833 ; p<,001*** F = 5,361 ; p=<,001*** 45.3% 54.7% X² = 13.921 ; p=,003** N=914 Lives comfortably 8,13 7,64 0,9089 60,7415 33,50% 66,50% N=148 Revenue is sufficient 7,76 7,09 1,0481 55,2033 45,00% 55,00% N=385 Difficult to live with current revenue 7,26 6,62 1,5433 51,8841 51,00% 49,00% N=284 very difficult to live with current revenue 6,6 6,08 1,7571 52,1658 48,80% 51,20% N= 97 Proficiency in language of the region F = 4,253 ; p=,005** F = 1,932 ; p=,123 F = 6,484 ; p= <,001*** F = 2,067 ; p=,103 45.2% 54.8% X² = 1.812 ; p=,612 N=921 Very good 7,75 7,06 1,0112 55,3739 44,30% 55,70% N=494 Good 7,42 7,02 1,4779 53,9445 45,90% 54,10% N=178 Moderate 7,31 6,79 1,4909 59,4266 41,20% 58,80% N=88 Little to none 7,42 6,68 1,432 52,2986 49,10% 50,90% N=161 Religion F = 2,497 ; p=,029* F = 1,765 ; p= ,117 F = 1,970 ; p=,081 F = 1882 ; p=,095 45.4% 54.6% X² = 7.527 ; p=,184 N=922 Catholic 7,31 6,9 1,3011 52,1968 50,50% 49,50% N=301 Protestant 7,63 6,93 1,3848 58,0741 39,30% 60,70% N=133 Evangelical 7,7 7,08 0,935 55,1896 46,70% 53,30% N=133 Muslim 7,71 6,96 1,3829 55,1147 45,60% 54,40% N=187 No religion 7,63 6,85 1,0703 54,1045 43,60% 56,40% N=101 Other 7,76 7,52 1,1315 59,6984 36,60% 63,40% N=67 *p < .05. **p < .01. ***p < .001 (two-tailed test). Table 2.4: Post-hoc tukey comparisons of mental health outcomes by number of microaggression types experienced Table 2.4: Post-hoc tukey comparisons of mental health outcomes by number of microaggression types experienced (N=854) Dependent Variable Group Comparison Mean Difference P-value ANOVA F (p) Interpretation Anxiety (PHQ2) No vs 1–2 Types -0.478 0.002** 18.32 (p < .001) Higher anxiety for 1–2 types vs none No vs 3+ Types -0.834 <0.001*** 18.32 (p < .001) Highest anxiety in 3+ types group 1–2 vs 3+ Types -0.356 0.010** 18.32 (p < .001) Anxiety increases with exposure Life Satisfaction No vs 1–2 Types 0.422 0.004** 17.32 (p < .001) Lower satisfaction in 1–2 types No vs 3+ Types 0.769 <0.001*** 17.32 (p < .001) Significantly lower with 3+ types 1–2 vs 3+ Types 0.347 0.008** 17.32 (p < .001) Satisfaction decreases with exposure Quality of Sleep No vs 1–2 Types 0.472 0.014* 17.25 (p < .001) Better sleep in no exposure group No vs 3+ Types 0.974 <0.001*** 17.25 (p < .001) Worst sleep in 3+ types group 1–2 vs 3+ Types 0.502 0.002** 17.25 (p < .001) Sleep quality declines with more exposure WHO-5 Wellbeing No vs 1–2 Types -2.368 0.494 0.92 (p = .398) NS – no significant difference No vs 3+ Types -2.685 0.404 0.92 (p = .398) NS – no significant difference 1–2 vs 3+ Types -0.317 0.984 0.92 (p = .398) NS – no significant difference *p < .05. **p < .01. ***p < .001 (two-tailed test). Table 2.5: Post-hoc tukey comparisons of mental health outcomes by number of harassment types experienced Table 2.5: Post-hoc tukey comparisons of mental health outcomes by number of harassment types experienced (N=884) Dependent Variable Group Comparison Mean Difference P-value ANOVA F (p) Interpretation Anxiety (PHQ2) No vs 1 Type -0.698 <0.001*** 25.91 (p < .001) Significantly higher anxiety with 1 type No vs 2 Types -0.933 <0.001*** 25.91 (p < .001) Anxiety rises with increased types No vs 3+ Types -1.137 <0.001*** 25.91 (p < .001) Highest anxiety with 3+ harassment types Life Satisfaction No vs 1 Type 0.593 <0.001*** 20.35 (p < .001) Satisfaction declines with any harassment No vs 2 Types 0.748 <0.001*** 20.35 (p < .001) Lower satisfaction with 2 types No vs 3+ Types 1.035 <0.001*** 20.35 (p < .001) Lowest satisfaction at 3+ types Quality of Sleep No vs 1 Type 0.472 <0.001*** 23.41 (p < .001) Better sleep in no exposure No vs 2 Types 1.273 <0.001*** 23.41 (p < .001) Worst sleep with 2 types No vs 3+ Types 0.923 <0.001*** 23.41 (p < .001) Lower sleep quality in 3+ harassment types WHO-5 Wellbeing No vs 1 Type -1.073 0.952 23.41 (p < .001) NS – no significant difference No vs 2 Types 2.656 0.589 0.92 (p = .432) NS – no significant difference No vs 3+ Types -0.997 0.984 0.92 (p = .432) NS – no significant difference *p < .05. **p < .01. ***p < .001 (two-tailed test). Table 2.6: Non-parametric comparisons of mental health outcomes by number of harassment types experienced (Mann–Whitney U tests) Table 2.6: Non-parametric comparisons of mental health outcomes by number of harassment types experienced (Mann– Whitney U tests) Dependent Variable Comparison Group 1 (N) Group 2 (N) Z-score P-value Interpretation Anxiety (PHQ2) No vs 1 Type 460 229 -6.633 <.001* Higher anxiety with 1 type No vs 2 Types 460 199 -7.715 <.001* Higher anxiety with 2 types No vs 3+ Types 460 96 -6.460 <.001* Highest anxiety with 3+ types 1 vs 2 Types 229 199 -1.404 .160 No significant difference 1 vs 3+ Types 229 96 -1.709 .087 No significant difference 2 vs 3+ Types 199 96 -.625 .532 No significant difference Life Satisfaction No vs 1 Type 459 227 -5.246 <.001* Lower satisfaction with 1 type No vs 2 Types 459 198 -5.901 <.001* Lower satisfaction with 2 types No vs 3+ Types 459 96 -5.998 <.001* Lowest satisfaction with 3+ types 1 vs 2 Types 227 198 -.849 .396 No significant difference 1 vs 3+ Types 227 96 -1.998 .046 No significant difference 2 vs 3+ Types 198 96 -1.226 .220 No significant difference Quality of Sleep No vs 1 Type 460 229 -5.256 <.001* Poorer sleep with 1 type No vs 2 Types 460 199 -8.160 <.001* Poorer sleep with 2 types No vs 3+ Types 460 96 -4.711 <.001* Worst sleep with 3+ types 1 vs 2 Types 229 199 -2.417 .016 Not significant after correction 1 vs 3+ Types 229 96 -.380 .704 No significant difference 2 vs 3+ Types 199 96 -1.664 .096 No significant difference Wellbeing (WHO-5) No vs 1 Type 461 229 -.096 .924 No significant difference No vs 2 Types 461 199 -1.972 .049 Not significant after correction No vs 3+ Types 461 96 -.342 .733 No significant difference 1 vs 2 Types 229 199 -1.638 .101 No significant difference 1 vs 3+ Types 229 96 -.209 .835 No significant difference 2 vs 3+ Types 199 96 -1.799 .072 No significant difference Significance threshold Bonferroni-corrected to p < .009* due to six comparisons per DV (0.05/6 = 0.0083 ≈ 0.009). A3: Table 3.13.8: Full Multivariate regression models estimating effect among microaggression and harassment on secondary mental health outcome: Life satisfaction, Sleep quality, Anxiety (N=908) Table 3.1 Model: Multivariate regression analysis estimating effect of microaggression on WHO-5Wellbeing. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B SE Sig. (Constant) 64,74 6,06 <,001 SUM_microaggression 0,12 0,12 0,35 Gender_dummyfemale -7,47 1,56 <,001 age_personne -0,02 0,08 0,78 Legal_Status_Dummy1 -6,52 2,64 0,01 Legal_Status_Dummy2 1,12 2,43 0,65 Languageproficiencyregion_dummy1verygood -2,00 2,88 0,49 Languageproficiencyregion_dummy2good -1,11 2,87 0,70 Languageproficiencyregion_dummy3moderate 4,53 3,15 0,15 householdincome_dummy1 6,95 2,41 0,00 householdincome_dummy2 2,69 1,80 0,14 subregion_of_residenceBelgium_dummy1 1,77 2,03 0,38 subregion_of_residenceBelgium_dummy2 -2,25 2,30 0,33 Generation_dummy1 -1,31 2,99 0,66 Generation_dummy2 6,04 3,08 0,05 HighestEducation_4cat_dummy1 -0,37 2,72 0,89 HighestEducation_4cat_dummy2 -3,66 2,37 0,12 HighestEducation_4cat_dummy3 -0,73 2,10 0,73 religion_dummy1 -8,25 3,21 0,01 religion_dummy2 -2,97 3,48 0,39 religion_dummy3 -4,46 3,45 0,20 religion_dummy4 -6,13 3,30 0,06 religion_dummy5 -8,14 3,67 0,03 francophone_dummy1 1,67 2,12 0,43 Table 3.2 Model: Multivariate regression analysis estimating effect of harassment on WHO-5-Wellbeing. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B Std. Error Sig. (Constant) 64,89 6,02 <,001 SUM_harassment 0,19 0,19 0,32 Gender_dummyfemale -7,57 1,56 <,001 age_personne -0,03 0,08 0,76 Legal_Status_Dummy1 -6,42 2,64 0,02 Legal_Status_Dummy2 1,09 2,43 0,65 Languageproficiencyregion_dummy1verygood -1,96 2,88 0,50 Languageproficiencyregion_dummy2good -1,02 2,87 0,72 Languageproficiencyregion_dummy3moderate 4,68 3,15 0,14 householdincome_dummy1 6,89 2,41 0,00 householdincome_dummy2 2,68 1,79 0,14 subregion_of_residenceBelgium_dummy1 1,76 2,03 0,39 subregion_of_residenceBelgium_dummy2 -2,23 2,30 0,33 Generation_dummy1 -1,28 2,99 0,67 Generation_dummy2 6,11 3,07 0,05 HighestEducation_4cat_dummy1 -0,31 2,72 0,91 HighestEducation_4cat_dummy2 -3,57 2,37 0,13 HighestEducation_4cat_dummy3 -0,73 2,10 0,73 religion_dummy1 -8,08 3,20 0,01 religion_dummy2 -2,73 3,47 0,43 religion_dummy3 -4,31 3,45 0,21 religion_dummy4 -6,09 3,30 0,07 religion_dummy5 -8,12 3,67 0,03 francophone_dummy1 1,63 2,12 0,44 Table 3.3 Model: Multivariate regression analysis estimating effect of harassment on anxiety. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B Std. Error Sig. (Constant) 0,80 0,39 0,04 SUM_harassment 0,10 0,01 <,001 Gender_dummyfemale 0,32 0,10 0,00 age_personne 0,01 0,01 0,13 Legal_Status_Dummy1 0,62 0,17 <,001 Legal_Status_Dummy2 0,19 0,16 0,24 Languageproficiencyregion_dummy1verygood -0,39 0,19 0,04 Languageproficiencyregion_dummy2good 0,09 0,19 0,64 Languageproficiencyregion_dummy3moderate 0,13 0,20 0,53 householdincome_dummy1 -0,50 0,16 0,00 householdincome_dummy2 -0,37 0,12 0,00 subregion_of_residenceBelgium_dummy1 -0,18 0,13 0,18 subregion_of_residenceBelgium_dummy2 -0,12 0,15 0,41 Generation_dummy1 -0,39 0,19 0,04 Generation_dummy2 0,10 0,20 0,63 HighestEducation_4cat_dummy1 0,13 0,18 0,48 HighestEducation_4cat_dummy2 0,19 0,15 0,21 HighestEducation_4cat_dummy3 0,19 0,14 0,17 religion_dummy1 0,20 0,21 0,33 religion_dummy2 0,31 0,22 0,17 religion_dummy3 -0,15 0,22 0,49 religion_dummy4 0,26 0,21 0,22 religion_dummy5 -0,04 0,24 0,88 francophone_dummy1 0,02 0,14 0,90 Table 3.4 Model: Multivariate regression analysis estimating effect of microaggression on life anxiety. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B Std. Error Sig (Constant) 0,85 0,40 0,03 SUM_microaggression 0,04 0,01 <,001 Gender_dummyfemale 0,36 0,10 <,001 age_personne 0,01 0,01 0,12 Legal_Status_Dummy1 0,57 0,17 0,00 Legal_Status_Dummy2 0,19 0,16 0,24 Languageproficiencyregion_dummy1verygood -0,41 0,19 0,03 Languageproficiencyregion_dummy2good 0,03 0,19 0,86 Languageproficiencyregion_dummy3moderate 0,07 0,21 0,73 householdincome_dummy1 -0,49 0,16 0,00 householdincome_dummy2 -0,39 0,12 <,001 subregion_of_residenceBelgium_dummy1 -0,18 0,13 0,17 subregion_of_residenceBelgium_dummy2 -0,14 0,15 0,37 Generation_dummy1 -0,40 0,20 0,04 Generation_dummy2 0,09 0,20 0,67 HighestEducation_4cat_dummy1 0,10 0,18 0,58 HighestEducation_4cat_dummy2 0,15 0,16 0,32 HighestEducation_4cat_dummy3 0,19 0,14 0,17 religion_dummy1 0,14 0,21 0,52 religion_dummy2 0,22 0,23 0,34 religion_dummy3 -0,21 0,23 0,35 religion_dummy4 0,25 0,22 0,25 religion_dummy5 -0,03 0,24 0,90 francophone_dummy1 0,03 0,14 0,84 Table 3.5 Model: Multivariate regression analysis estimating effect of harassment on sleep quality. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B Std. Error Sig. (Constant) 6,96 0,48 <,001 SUM_harassment -0,09 0,02 <,001 Gender_dummyfemale -0,28 0,12 0,02 age_personne 0,01 0,01 0,24 Legal_Status_Dummy1 -0,42 0,21 0,05 Legal_Status_Dummy2 -0,25 0,19 0,20 Languageproficiencyregion_dummy1verygood 0,75 0,23 0,00 Languageproficiencyregion_dummy2good 0,55 0,23 0,02 Languageproficiencyregion_dummy3moderate 0,21 0,25 0,39 householdincome_dummy1 0,87 0,19 <,001 householdincome_dummy2 0,40 0,14 0,01 subregion_of_residenceBelgium_dummy1 -0,12 0,16 0,45 subregion_of_residenceBelgium_dummy2 0,32 0,18 0,08 Generation_dummy1 0,04 0,24 0,88 Generation_dummy2 -0,38 0,25 0,12 HighestEducation_4cat_dummy1 0,54 0,22 0,01 HighestEducation_4cat_dummy2 -0,14 0,19 0,46 HighestEducation_4cat_dummy3 -0,04 0,17 0,80 religion_dummy1 -0,56 0,26 0,03 religion_dummy2 -0,63 0,28 0,02 religion_dummy3 -0,30 0,27 0,28 religion_dummy4 -0,53 0,26 0,04 religion_dummy5 -0,60 0,29 0,04 francophone_dummy1 -0,25 0,17 0,13 Table 3.7 Model: Multivariate regression analysis estimating effect of microaggression on sleep quality. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B SE P (Constant) 7,02 0,48 <,001 SUM_microaggression -0,05 0,01 <,001 Gender_dummyfemale -0,33 0,12 0,01 age_personne 0,01 0,01 0,29 Legal_Status_Dummy1 -0,37 0,21 0,08 Legal_Status_Dummy2 -0,26 0,19 0,19 Languageproficiencyregion_dummy1verygood 0,77 0,23 <,001 Languageproficiencyregion_dummy2good 0,60 0,23 0,01 Languageproficiencyregion_dummy3moderate 0,28 0,25 0,27 householdincome_dummy1 0,85 0,19 <,001 householdincome_dummy2 0,40 0,14 0,01 subregion_of_residenceBelgium_dummy1 -0,13 0,16 0,43 subregion_of_residenceBelgium_dummy2 0,33 0,18 0,07 Generation_dummy1 0,05 0,24 0,84 Generation_dummy2 -0,35 0,25 0,15 HighestEducation_4cat_dummy1 0,57 0,22 0,01 HighestEducation_4cat_dummy2 -0,10 0,19 0,60 HighestEducation_4cat_dummy3 -0,05 0,17 0,78 religion_dummy1 -0,48 0,26 0,06 religion_dummy2 -0,52 0,28 0,06 religion_dummy3 -0,23 0,28 0,41 religion_dummy4 -0,51 0,26 0,05 religion_dummy5 -0,59 0,29 0,04 francophone_dummy1 -0,27 0,17 0,11 Table 3.7 Model: Multivariate regression analysis estimating effect of harassment on life satisfaction. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B SE P (Constant) 8,10 0,37 <,001 SUM_harassment -0,07 0,01 <,001 Gender_dummyfemale -0,27 0,10 0,01 age_personne 0,00 0,01 0,65 Legal_Status_Dummy1 -0,68 0,16 <,001 Legal_Status_Dummy2 -0,16 0,15 0,29 Languageproficiencyregion_dummy1verygood 0,39 0,18 0,03 Languageproficiencyregion_dummy2good -0,03 0,18 0,88 Languageproficiencyregion_dummy3moderate -0,18 0,19 0,36 householdincome_dummy1 0,73 0,15 <,001 householdincome_dummy2 0,38 0,11 <,001 subregion_of_residenceBelgium_dummy1 -0,18 0,12 0,14 subregion_of_residenceBelgium_dummy2 0,17 0,14 0,23 Generation_dummy1 0,02 0,18 0,90 Generation_dummy2 -0,46 0,19 0,02 HighestEducation_4cat_dummy1 -0,03 0,17 0,86 HighestEducation_4cat_dummy2 0,20 0,15 0,18 HighestEducation_4cat_dummy3 -0,15 0,13 0,25 religion_dummy1 -0,37 0,20 0,06 religion_dummy2 -0,23 0,21 0,27 religion_dummy3 0,01 0,21 0,97 religion_dummy4 0,02 0,20 0,91 religion_dummy5 -0,16 0,22 0,47 francophone_dummy1 -0,23 0,13 0,08 Table 3.8 Model: Multivariate regression analysis estimating effect of microaggression on life satisfaction. Adjusted for covariates; age, gender, generation, legal status, subregion of residence, Francophone origin, highest educational degree, language proficiency, household income, and religion. Unstandardized Coefficients B SE P (Constant) 8,07 0,37 <,001 SUM_microaggression -0,03 0,01 <,001 Gender_dummyfemale -0,30 0,10 0,00 age_personne 0,00 0,01 0,61 Legal_Status_Dummy1 -0,65 0,16 <,001 Legal_Status_Dummy2 -0,16 0,15 0,30 Languageproficiencyregion_dummy1verygood 0,40 0,18 0,02 Languageproficiencyregion_dummy2good 0,01 0,18 0,95 Languageproficiencyregion_dummy3moderate -0,13 0,19 0,49 householdincome_dummy1 0,72 0,15 <,001 householdincome_dummy2 0,39 0,11 <,001 subregion_of_residenceBelgium_dummy1 -0,18 0,13 0,15 subregion_of_residenceBelgium_dummy2 0,18 0,14 0,21 Generation_dummy1 0,03 0,18 0,88 Generation_dummy2 -0,45 0,19 0,02 HighestEducation_4cat_dummy1 -0,01 0,17 0,95 HighestEducation_4cat_dummy2 0,22 0,15 0,13 HighestEducation_4cat_dummy3 -0,15 0,13 0,24 religion_dummy1 -0,32 0,20 0,10 religion_dummy2 -0,16 0,21 0,45 religion_dummy3 0,05 0,21 0,80 religion_dummy4 0,04 0,20 0,86 religion_dummy5 -0,16 0,23 0,47 francophone_dummy1 -0,24 0,13 0,07 A4 : Table 3.3-3.4 Full multivariate regression adjusted models with main effect and moderation effect for interaction terms Table 3.3: Two-way interaction multivariate regression models of mental health outcomes (anxiety, sleep quality, and life satisfaction) by harassment exposure and moderators (gender, legal status, language proficiency per region, and household income). All models were adjusted for potential confounders, including age, generation, gender (if not a moderator), legal status, subregion of residence, Francophone origin, highest educational attainment, language proficiency, household income, and religion. Each two-way interaction was tested in a separate model. (N=908) Anxiety Sleep quality Life satisfaction With two-way interaction b SE p-value 95% CI b SE p-value 95% CI b SE p-value 95% CI Microaggression x gender (intercept) -0.16 0.34 0.643 [-0.82, 0.51] 8,61 0.41 <.001 [7.80, 9.42] 9,161 0.32 <.001 [8.54, 9.78] Microaggression (main effect) 0.07 0.01 <.001 [0.04, 0.09] -0.05 0.01 .001 [-0.07, -0.02] -0.03 0.01 .017 [-0.05, -0.01] Microaggression xfemale -0.04 0.02 0.007 [-0.08, -0.01] -0.01 0.02 .583 [-0.05, 0.03] -0.01 0.02 .399 [-0.04, 0.02] Microaggression x male (reference) — — — — — — — — — — — — female (Gender_dummy1) 0.69 0.16 <.001 [0.38, 1.00] -0.25 0.20 .210 [-0.63, 0.14] -0.20 0.15 .188 [-0.49, 0.10] Microaggression x legal status (intercept) -0.14 0.34 0.682 [-0.81, 0.53] 8.64 0.41 <.001 [7.82, 9.45] 9.15 0.32 <.001 [8.53, 9.77] Microaggression (main effect) 0.06 0.01 <.001 [0.04, 0.08] -0.05 0.01 <.001 [-0.08, -0.03] -0.03 0.01 <.001 [-0.05, -0.01] Microaggression x No work permit or work permit of 1 year or less -0.05 0.02 0.023 [-0.08, -0.01] -0.01 0.02 0.633 [-0.06, 0.04] -0.03 0.02 0.15 [-0.06, 0.01] Microaggression x Work permit of more than 1 year -0.02 0.02 0.446 [-0.06, 0.03] 0.02 0.03 0.495 [-0.04, 0.07] 0.00 0.02 0.972 [-0.04, 0.04] Microaggression x Belgian nationality or no need for a work permit (reference) — — — — — — — — — — — — No work permit or work permit of 1 year or less (legalstatus1) 0.94 0.24 <.001 [0.47, 1.40] -0.27 0.29 0.353 [-0.84, 0.30] -0.43 0.22 0.050 [-0.87, 0.01] Work permit of more than 1 year(legalstatus2) 0.31 0.23 0.175 [-0.13, 0.76] -0.39 0.28 0.162 [-0.93, 0.15] -0.16 0.21 0.454 [-0.58, 0.26] Microaggression x language proficiency per region (intercept) -0.01 0.34 0.971 [-0.68, 0.66] 8.69 0.42 <.001 [7.87, 9.50] 9.19 0.32 <.001 [8.56, 9.82] Microaggression (main effect) 0.05 0.01 <.001 [0.03, 0.07] -0.06 0.01 <.001 [-0.09, -0.04] -0.03 0.01 0.003 [-0.05, -0.01] Microaggression x little to none -0.03 0.02 0.221 [-0.07, 0.02] 0.03 0.03 0.306 [-0.03, 0.08] 0.01 0.02 0.585 [-0.03, 0.06] Microaggression x moderate -0.04 0.03 0.132 [-0.09, 0.01] 0.02 0.03 0.525 [-0.04, 0.09] -0.03 0.03 0.281 [-0.08, 0.02] Microaggression x good 0.03 0.02 0.179 [-0.01, 0.07] 0.01 0.03 0.623 [-0.04, 0.07] -0.01 0.02 0.765 [-0.05, 0.03] Microaggression x very good (reference) — — — — — — — — — — — — little to none (languageprofdummy1) 0.63 0.26 0.015 [0.12, 1.14] -1.00 0.32 0.002 [-1.62, -0.38] -0.48 0.24 0.050 [-0.96, 0.00] moderate (languageprofdummy2) 0.83 0.30 0.006 [0.24, 1.41] -0.67 0.36 0.068 [-1.38, 0.05] -0.30 0.28 0.284 [-0.85, 0.25] good (languageprofdummy3) 0.25 0.21 0.231 [-0.16, 0.66] -0.27 0.26 0.282 [-0.77, 0.23] -0.34 0.20 0.087 [-0.72, 0.05] Microaggression x household income (intercept) -0.24 0.36 0.505 [-0.94, 0.46] 8.73 0.43 <.001 [7.88, 9.57] 9.20 0.33 <.001 [8.55, 9.86] Microaggression (main effect) 0.07 0.02 <.001 [0.03, 0.11] -0.07 0.02 0.006 [-0.11, -0.02] -0.04 0.02 0.057 [-0.07, 0.00] Microaggression x Difficult or very difficult with current income -0.04 0.02 0.067 [-0.09, 0.00] 0.03 0.03 0.322 [-0.03, 0.08] 0.001 0.02 0.948 [-0.04, 0.04] Microaggression x Sufficient income -0.03 0.02 0.249 ,[-0.07, 0.02] 0.003 0.03 0.927 [-0.05, 0.06] 0.003 0.02 0.885 [-0.04, 0.05] Microaggression x Living comfortable (reference) — — — — — — — — — — — — Difficult or very difficult with current income (householdincome1) 0.83 0.24 <.001 [0.35, 1.30] -1.08 0.29 <.001 [-1.66, -0.50] -0.73 0.23 0.001 [-1.18, -0.28] Sufficient income (householdincome2) 0.30 0.22 0.177 [-0.14, 0.74] - 0.47 0.27 0.081 [-1.00, 0.06] -0.35 0.21 0.092 [-0.76, 0.06] Table 3.4: Two-way interaction multivariate regression models of mental health outcomes (anxiety, sleep quality, and life satisfaction) by harassment exposure and moderators (gender, legal status, language proficiency per region, and household income). All models were adjusted for potential confounders, including age, generation, gender, legal status, subregion of residence, Francophone origin, highest educational attainment, language proficiency, household income, and religion. Each two-way interaction was tested in a separate model. (N=908) Anxiety Sleep quality Life satisfaction With two-way interactions: b SE p-value 95% CI b SE p-value 95% CI b SE p-value 95% CI Harassment x gender (intercept) -0.09 0.33 0.777 [-0.74, 0.56] 8.54 0.41 <.001 [7.74, 9.33] 9.18 0.31 <.001 [8.57, 9.79] Harassment (main effect) 0.10 0.02 <.001 [0.06, 0.13] -0.06 0.02 0.006 [-0.10, -0.02] -0.05 0.02 0.005 [-0.08, -0.01] Harassment xfemale 0.00 0.03 0.994 [-0.05, 0.05] -0.06 0.03 0.054 [-0.12, 0.00] -0.04 0.02 0.070 [-0.09, 0.00] Harassment x male (reference) — — — — — — — — — — — — female (Gender_dummy1) 0.32 0.13 0.017 [0.06, 0.57] -0.08 0.16 0.609 [-0.40, 0.23] -0.12 0.12 0.332 [-0.36, 0.12] Harassment x legal status (intercept) -0.20 0.33 0.553 [-0.85, 0.45] 8.63 0.41 <.001 [7.82, 9.43] 9.22 0.31 <.001 [8.61, 9.84] Harassment (main effect) 0.12 0.02 <.001 [0.09, 0.15] -0.10 0.02 <.001 [-0.14, -0.06] -0.06 0.02 <.001 [-0.09, -0.03] Harassment x No work permit or work permit of 1 year or less -0.10 0.03 <0.001 [-0.16, -0.04] 0.02 0.04 0.599 [-0.05, 0.09] -0.04 0.03 0.144 [-0.10, 0.01] Harassment x Work permit of more than 1 year -0.05 0.04 0.158 [-0.13, 0.02] 0.04 0.05 0.432 [-0.06, 0.13] 0.03 0.04 0.435 [-0.04, 0.10] Harassment x Belgian nationality or no need for a work permit (reference) — — — — — — — — — — — — No work permit or work permit of 1 year or less (legalstatus1) 0.95 0.20 <.001 [0.56, 1.34] -0.50 0.25 0.044 [-0.98, -0.01] -0.55 0.19 0.003 [-0.92, -0.18] Work permit of more than 1 year(legalstatus2) 0.36 0.20 0.077 [-0.04, 0.75] -0.37 0.25 0.138 [-0.85, 0.12] -0.25 0.19 0.183 [-0.62, 0.12] Harassment x language proficiency per region (intercept) -0.07 0.33 0.830 [-0.72, 0.58] 8.60 0.41 <.001 [7.80, 9.39] 9.16 0.31 <.001 [8.55, 9.77] Harassment (main effect) 0.08 0.02 <.001 [0.05, 0.12] -0.08 0.02 <.001 [-0.12, -0.04] -0.04 0.02 0.019 [-0.07, -0.01] Harassment x little to none -0.01 0.03 0.850 [-0.07, 0.06] 0.02 0.04 0.626 [-0.06, 0.10] -0.06 0.03 0.083 [-0.12, 0.01] Harassment x moderate 0.06 0.05 0.187 [-0.03, 0.16] -0.12 0.06 0.036 [-0.24, -0.01] -0.14 0.05 0.002 [-0.22, -0.05] Harassment x good 0.05 0.03 0.111 [-0.01, 0.12] -0.01 0.04 0.711 [-0.10, 0.07] -0.05 0.03 0.108 [-0.11, 0.01] Harassment x very good (reference) — — — — — — — — — — — — little to none (languageprofdummy1) 0.40 0.22 0.067 [-0.03, 0.83] -0.80 0.27 0.003 [-1.33, -0.27] -0.18 0.20 0.379 [-0.58, 0.22] moderate (languageprofdummy2) 0.30 0.25 0.239 [-0.20, 0.79] -0.10 0.31 0.736 [-0.71, 0.50] -0.08 0.24 0.739 [-0.54, 0.39] good (languageprofdummy3) 0.31 0.17 0.068 [-0.02, 0.65] -0.14 0.21 0.503 [-0.56, 0.27] -0.24 0.16 0.135 [-0.56, 0.07] Harassment x household income (intercept) -0.21 0.34 0.540 [-0.87, 0.46] 8.54 0.41 <.001 [7.74, 9.33] 9.19 0.32 <.001 [8.56, 9.81] Harassment (main effect) 0.14 0.03 <.001 [0.08, 0.20] -0.06 0.02 0.006 [-0.10, -0.02] -0.07 0.03 0.014 [-0.12, -0.01] Harassment x Difficult or very difficult with current income (householdincome1) -0.06 0.04 0.094 [-0.13, 0.01] -0.06 0.03 0.050 [-0.12, 0.00] 0.03 0.03 0.381 [-0.04, 0.09] Harassment x Sufficient income (householdincome2) -0.05 0.04 0.190 [-0.12, 0.02] -0.06 0.03 0.050 [-0.12, 0.00] -0.04 0.03 0.262 [-0.11, 0.03] Harassment x Living comfortable (reference) — — — — — — — — — — — — Difficult or very difficult with current income (householdincome1) 0.71 0.20 0.190 [0.32, 1.11] -0.87 0.19 <.001 [-1.25, -0.50] -0.85 0.19 <.001 [-1.22, -0.48] Sufficient income (householdincome2) 0.29 0.19 0.120 [-0.08, 0.66] -0.46 0.18 0.009 [-0.81, -0.12] -0.24 0.18 0.18 [-0.58, 0.11] Patil, P. A., Porche, M. V., Shippen, N. A., Dallenbach, N. T., & Fortuna, L. R. (2017). Which girls, which boys? The intersectional risk for depression by race and ethnicity, and gender in the U.S. Clinical Psychology Review, 66, 51–68. https://doi.org/10.1016/j.cpr.2017.12.003 Pieterse, A., & Powell, S. (2016). A theoretical overview of the impact of racism on people of color. In American Psychological Association eBooks,11–30. https://doi.org/10.1037/14852-002 Polanco-Roman, L., Danies, A., & Anglin, D. M. (2016). Racial discrimination as race-based trauma, coping strategies, and dissociative symptoms among emerging adults. Psychological Trauma: Theory, Research, Practice, and Policy, 8(5), 609–617. https://doi.org/10.1037/tra0000125 Quinn, N. (2013). Participatory action research with asylum seekers and refugees experiencing stigma and discrimination: the experience from Scotland. Disability & Society, 29(1), 58–70. https://doi.org/10.1080/09687599.2013.769863 Rodriguez, V. E., Enriquez, L. E., Ro, A., & Ayón, C. (2023). Immigration-Related Discrimination and Mental Health among Latino Undocumented Students and U.S. Citizen Students with Undocumented Parents: A Mixed-Methods Investigation. Journal Of Health And Social Behavior, 64(4), 593–609. https://doi.org/10.1177/00221465231168912 Roberg, R., Camargo, T., & Marks, A. K. (2024). A Mixed-Methods Exploration of Legal Vulnerability, Trauma, and Psychological Wellbeing in Immigrant Caregivers and Youth. Trauma Care, 4(1), 60–74. https://doi.org/10.3390/traumacare4010006 Rüdel, J., & Joly, M. (2024). Perceived loneliness: Why are Syrian refugees more lonely than other newly arrived migrants in Germany? Comparative Migration Studies, 12(1). https://doi.org/10.1186/s40878-024-00398-9 Sam, D. L. (2024). 50+ years of psychological acculturation research: Progress and challenges. International Journal Of Intercultural Relations, 103, 102076. https://doi.org/10.1016/j.ijintrel.2024.102076 Sartorius, N. (2022). The Complexity of Comorbidity in Patients with Severe Mental Disorders. European Psychiatry, 65(S1), S16. https://doi.org/10.1192/j.eurpsy.2022.66 Savaş, Ö., Greenwood, R. M., Blankenship, B. T., Stewart, A. J., & Deaux, K. (2021). All immigrants are not alike: Intersectionality matters in views of immigrant groups. Journal Of Social And Political Psychology, 9(1), 86–104. https://doi.org/10.5964/jspp.5575 Sawyer, P. J., Major, B., Casad, B. J., Townsend, S. S. M., & Mendes, W. B. (2012). Discrimination and the Stress Response: Psychological and Physiological Consequences of Anticipating Prejudice in Interethnic Interactions. American Journal Of Public Health, 102(5), 1020–1026. https://doi.org/10.2105/ajph.2011.300620 Schaeffer, M., & Kas, J. (2023). The Integration Paradox: A Review and Meta-Analysis of the Complex Relationship Between Integration and Reports of Discrimination. International Migration Review, 58(3), 1384–1409. https://doi.org/10.1177/01979183231170809 Schaeffer, M., & Kas, J. (2024). The integration paradox: Does awareness of the extent of ethno‐racial discrimination increase reports of discrimination? Political Psychology, 46(3), 623-636 https://doi.org/10.1111/pops.13027 Scheim, A. I., & Bauer, G. R. (2019). The Intersectional Discrimination Index: Development and validation of measures of self-reported enacted and anticipated discrimination for intercategorical analysis. Social Science & Medicine, 226, 225–235. https://doi.org/10.1016/j.socscimed.2018.12.016 Schouler-Ocak, M., & Moran, J. (2022). Multiple Discrimination and Its Consequences for the Mental Health of Ethnic Minorities. European Psychiatry, 65(S1), S16. https://doi.org/10.1192/j.eurpsy.2022.65 Schouler-Ocak, M., & Moran, J. K. (2022b). Racial discrimination and its impact on mental health. International Review Of Psychiatry, 35(3–4), 268–276. https://doi.org/10.1080/09540261.2022.2155033 Simpson, J. & CRIAW/ICREF. (2009). A toolkit for applying intersectionality (1st edition). https://also-chicago.org/wp-content/uploads/2017/03/Everyone_Belongs-A-toolkit-forapplying-intersectionality.pdf Sue, D. W., Capodilupo, C. M., Torino, G. C., Bucceri, J. M., Holder, A., Nadal, K. L., & Esquilin, M. (2007). Racial microaggressions in everyday life: Implications for clinical practice. American Psychologist, 62(4), 271–286. https://doi.org/10.1037/0003066X.62.4.271 Sullivan, J. M., Harman, M., & Sullivan, S. (2021). Gender differences in African Americans' reactions to and coping with discrimination: Results from The National Study of American Life. Journal of Community Psychology, 49(7), 2424–2440. https://doi.org/10.1002/jcop.22665 Szaflarski, M. and Bauldry, S. (2019), "The Effects of Perceived Discrimination on Immigrant and Refugee Physical and Mental Health", Immigration and Health (Advances in Medical Sociology, 19, 173-204. https://doi.org/10.1108/S1057-629020190000019009 Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social categorization and intergroup behaviour. European Journal Of Social Psychology, 1(2), 149–178. https://doi.org/10.1002/ejsp.2420010202 Tao, X., & Fisher, C. B. (2021). Exposure to Social Media Racial Discrimination and Mental Health among Adolescents of Color. Journal Of Youth And Adolescence, 51(1), 30–44. https://doi.org/10.1007/s10964-021-01514-z Tellez Lieberman, J., Valdez, C.R., Pintor, J.K., et al. (2023). “It felt like hitting rock bottom”: A qualitative exploration of the mental health impacts of immigration enforcement and discrimination on US-citizen, Mexican children. Latino Studies, 21, 323–347. https://doi.org/10.1057/s41276-023-00415-5. Tip, L. K., Brown, R., Morrice, L., Collyer, M., & Easterbrook, M. J. (2018). Improving Refugee Well-Being With Better Language Skills and More Intergroup Contact. Social Psychological And Personality Science, 10(2), 144–151. https://doi.org/10.1177/1948550617752062 Tobin, C. S. T., & Moody, M. D. (2021). Does Early Life Racial Discrimination Explain a Mental Health Paradox among Black Adults? Journal Of Aging And Health, 33(5–6), 396–408. https://doi.org/10.1177/0898264320988187 Turan, J. M., Elafros, M. A., Logie, C. H., Banik, S., Turan, B., Crockett, K. B., Pescosolido, B., & Murray, S. M. (2019). Challenges and opportunities in examining and addressing intersectional stigma and health. BMC Medicine, 17(1). https://doi.org/10.1186/s12916018-1246-9 United Nations. (1965). International Convention on the Elimination of All Forms of Racial Discrimination. Office of the High Commissioner for Human Rights. https://www.ohchr.org/en/instruments-mechanisms/instruments/internationalconvention-elimination-all-forms-racial UN General Assembly, Universal Declaration of Human Rights, 217 A (III), 10 December 1948, https://www.refworld.org/legal/resolution/unga/1948/en/11563 Vargas, S. M., Huey, S. J., & Miranda, J. (2020). A critical review of current evidence on multiple types of discrimination and mental health. American Journal Of Orthopsychiatry, 90(3), 374–390. https://doi.org/10.1037/ort0000441 Vaid, E., & Wadsworth, M. E. (2023). Perceived Racial Discrimination, Poverty-Related Stress, Civic Efficacy, and Psychological Problems in Low Socioeconomic Preadolescents. The Journal Of Early Adolescence, 44(7), 909–933. https://doi.org/10.1177/02724316231212561 Weeks, M. R., & Sullivan, A. L. (2019). Discrimination Matters: Relations of Perceived Discrimination to Student Mental Health. School Mental Health, 11(3), 425–437. https://doi.org/10.1007/s12310-019-09309-1 Williams, D. R., Lawrence, J. A., & Davis, B. A. (2019). Racism and Health: Evidence and Needed Research. Annual Review Of Public Health, 40(1), 105–125. https://doi.org/10.1146/annurev-publhealth-040218-043750 World Health Organization. (2024). The World Health Organization-Five Well-Being Index (WHO-5). WHO-UCN-MSD-MHE-2024.01. World Health Organization. https://www.who.int/publications/m/item/WHO-UCN-MSD-MHE-2024.01