« Older Entries

Single Mothers of Children with Autism Spectrum Disorders (ASD): The Relationship among ASD Severity, Social Support, and Maternal Depression

Research has found that depression and anxiety are present in 26.7% and 33.7% of parents with ASD children, respectively (Machado Junior, Celestino, Serra, Caron, Ponde, 2014). Social support has also been identified as a critical factor that reduces the negative psychological effects of raising a child with ASD (Bishop et al., 2007; Bromley et al., 2004).

Therefore this study explored the relationship between depression and social supports for single mothers who are caregivers of a child with Autism Spectrum Disorder (ASD).  Using a cross-sectional research design, we recruited a sample of 42 mothers with low (n = 22) and high (n = 21) ASD-symptom children from three CLSCs in Montreal, Quebec.

The study surveyed respondents using the Patient Health Questionnaire-9 to assess the mother’s level of depression and the Interpersonal Support Evaluation List-Short Form to assess the level of social supports. As hypothesized, our findings showed that mothers of children with low autism reported lower depression scores than mothers of children with high autism who reported moderate to moderately-severe depression. Mothers of children with low autism had significantly higher social support scores than mothers of children with high autism. A moderate negative correlation between depression and social supports was found for mothers of low autism children.

Due to the study’s limitations, the results are not generalizable to the larger population; however, findings suggest the importance of social support during difficult times with this population of mothers.


Machado Junior, S. B., Celestino, M. I. O., Serra, J. P. C., Caron, J., & Pondé, M. P. (2014). Risk and protective factors for symptoms of anxiety and depression in parents of children with autism spectrum disorder. Developmental Neurorehabilitation, (0), 1-8.


Resiliency among Aboriginal children in foster care in Quebec

Our cross-sectional descriptive study sought to determine the impact of cultural supports on the resiliency of Aboriginal children in foster care in Quebec. We hypothesized that Aboriginal youth’s increased access to culturally-appropriate settings while in foster care would be associated with higher scores on a measure of resiliency.

46 Aboriginal youth in out-of-home care were surveyed using the Child and Youth Resilience Measure-28 (CYRM-28). Findings from a regression analysis comparing scores of groups categorized by placement type indicate that children in an Aboriginal foster placement in the community with access to cultural services (type 1) have a resiliency score 38 points greater than the score for children with limited access to their community or cultural services (type 3). Children in a non-Aboriginal foster placement with some access to their community or cultural services (type 2) have a resiliency score 18 points greater than the score for children with limited access to their community or cultural services (type 3). These results support our hypothesis.

The study has some limitations. Results are not generalizable due to the small sample size (N=46). Selection bias may be a factor; this is highlighted by the over-representation of transgender and female respondents as compared with estimates of these subsamples within the Canadian population (Gates, 2011; Statistics Canada, 2010). Another limitation is the cross-sectional nature of the design; a longitudinal study would allow for a measurement of changes in the youth’s resilience levels over time. Our findings highlight the need for longitudinal research with Aboriginal youth in foster care to examine the outcomes of culturally continuous placement settings as well as the potential long-term benefits of cultural supports for these youth.


Gates, G.J. (2011). How mnay people are Lesbian, Gay, Bisexual, and Transgender? UCLA: The Williams Institute. Retrieved from: https://escholarship.org/uc/item/09h684x2

Statistics Canada, 2010. Population Projections for Canada, Provinces and Territories: 2009 to 2036. Catalogue no.91-520-X. Ottawa, Ontario.

What social workers need to know about post-partum depression

A group of social workers conducted a study of variables suspected to influence postpartum depression. The purpose of this blog is to share findings and improve treatment design.

Giving birth is supposed to be a joyful experience but for some it is accompanied by difficult emotions and a sense of ineptitude. Experiencing postpartum depression (PPD) is more than just having the “baby blues”. PPD is a debilitating mental disorder that affects approximately 17.15% of Canadian women and 16.8% of Quebec women following childbirth (Lanes, Kuk, & Tamim, 2011). The prevalence of minor/major PPD is estimated at 8.46% and the prevalence of major PPD at 8.69% (Lanes et al., 2011). Given the significance of these numbers, the Lakeshore hospital decided to examine the prevalence of PPD in its community.

This research used a cross-sectional design to explore perceived self-efficacy and postpartum depression in two groups of women, one with low and the other with high levels of social support. Symptoms of depression were found in 59% of women indicating that this population runs a much higher risk of experiencing PPD. A higher prevalence of depressive symptoms was associated with a lower level of social support and a lower sense of self-efficacy, indicating that the presence of these variables may influence PPDS.

These findings inform practice by supporting the need to improve psychosocial functioning, the ability to mobilize support and enhance the parenting capabilities in this population. Suggested treatment includes individual or group interpersonal psychotherapy (O’Hara et al., 2000; Mulcahy et al., 2010).


Relationship between severity of behavioural problems and depression in caregivers of children with Autism Spectrum Disorder (ASD)

Our team of researchers asked how are severity of behavioural problems related to depression in caregivers of children with Autism Spectrum Disorder (ASD)? The team hypothesised that more severe behavioural problems would be associated to higher levels of caregiver depression.

A survey was created in order to gain data on age, gender, family composition (single or two parent homes), perceived severity of the child’s behavioural problems and self-reported symptoms of depression.  Measures included a Likert-scale questionnaire adapted from the Child’s Behavior Checklist to measure severity of child behavioral problems and the Beck’s Depression Inventory to measure caregiver depression.

A total of 42 caregivers of autistic children participated in this study (n=42). Using the Pearson correlation coefficient, researchers found a very strong positive correlation between the two variables, r= 0.8, n= 42, p= >0.05. A one-way repeated measures ANOVA was conducted to explore the impact of the family composition in addition to the severity of behavior, on the parent’s depression level. The sample was divided into four groups: single-parent with high behavioral severity (single-H), two-parent with high behavioral severity (dual-H), single-parent with low behavioral severity (single-L) and two-parent with low behavioral severity (dual-L). The means of depression severity of these four groups, were compared using the Anova. Results show that family composition does not have a significant impact on parental depression. However due to the limitation regarding small sample sizes, these results remain inconclusive. The team of researchers therefore suggest further research be conducted with larger sample sizes.

Fostered Aboriginal Youth and Ethnic Identity


Aboriginal youth are over-represented in the Canadian foster care system (Trocmé, Knoke & Blackstock, 2004; Fluke, Chabot, Fallon, MacLaurin & Blackstock, 2010) whilst aboriginal caregivers are under-represented (Brown, Ivanova, Mehta, Skrodzki & Rodgers, 2015). Our research project explored the impacts of being placed in kinship versus non-kinship homes on the ethnic identity of aboriginal foster children placed in early-childhood. This study hypothesized that being placed within a kinship home would be correlated to youth having stronger ethnic identity. Foster children, youth, and aboriginal youths’ positive attachment to their ethnic identities has been linked to increased physical and mental well-being (Gfellner & Armstrong, 2012; Moss, 2009; Corenblum, 2014; Jones & Galliher, 2007). Consequently, facilitating the organic development of ethnic identity is an important policy-goal for youth protection agencies.


Using the Multi-Group Ethnic Identity Measure (MEIM) (Phinney, 1992), our study tested 43 aboriginal adolescents who had been placed in foster homes during their early-childhood. Our results suggest that being placed with a kinship foster-parent is correlated with: aboriginal caregiver ethnicity, fewer moves between foster homes, and longer-lasting placements. Kinship-care and its correlates were all associated with higher ethnic identity scores, particularly for the MEIM sub-category of ‘exploration.’ Indeed, fewer moves remained significant in a multi-variate model. These results underline the effectiveness of kinship foster placement. In addition, they suggest that further exploration of the factors that facilitate stability and longevity in kinship-care foster homes will be important in developing foster placement policies and best practices. This is especially true for policies that intend to contribute to aboriginal youths’ sense of ethnic identity and thus wellbeing.


Brown, J. D., Sintzel, J., George, N., & St Arnault, D. (2010). Benefits of transcultural fostering. Child & Family Social Work, 15(3), 276-285.


Corenblum, B. (January 01, 2014). Development of racial-ethnic identity among First Nation children. Journal of Youth and Adolescence, 43, 3, 356-74.


Fluke, J., Chabot, M., Fallon, B., MacLaurin, B. & Blackstock, C. (2010). Placement decisions and disparities among aboriginal groups: an application of the decision

making ecology through multi-level analysis.Child Abuse & Neglect, 34, 57-69. doi: 10.1016/j.chiabu.2009.08.009.


Gfellner, B. M., & Armstrong, H. D. (2012). Ego development, ego strengths, and ethnic identity among First Nation adolescents. Journal of Research on Adolescence22(2), 225-234.


Jones, M. D., & Galliher, R. V. (2007). Ethnic identity and psychosocial functioning in Navajo adolescents.Journal of Research on Adolescence, 17(4), 683–696. http://doi.org/10.1111/j.1532-7795.2007.00541.x


Moss, M. (August 01, 2009). Broken circles to a different identity: an exploration of identity for children in out-of-home care in Queensland, Australia. Child & Family Social Work, 14(3), 311-321.


Phinney, J. S. (1992). The Multigroup Ethnic Identity Measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7(2), 156–176. http://doi.org/10.1177/074355489272003


Trocmé, N, Knoke, D & Blackstock, C. (2004). Pathways to the Overrepresentation of  Aboriginal Children in Canada’s Child Welfare System. Social Service Review, 78(4): 577-600.




Prevalence of PTSD Symptoms in Youth Living in Residential Care: Results and Implications

Our study examined the relationship between PTSD, social support, life adversity, and impairment caused by trauma. We utilized non-probabilistic, convenience sampling to screen males aged 13 – 15 years (N = 40) who were placed in a residential unit within the last 12 months for the aforementioned constructs.

Our findings suggest that 80% of participants met the criteria for PTSD symptoms. Of those classified as having experienced a high level of adversity, 75% reported having no social support. Our findings are congruent with other studies which showed high prevalence of PTSD symptoms amongst adolescents in child protection services (Greeson et al., 2011; Salazar et al., 2013).

Significant differences were detected in PTSD scores between those who indicated support and those who had none. Implications of this study highlight the lack of support for youth with trauma and the importance of support systems that may act in part as a protective factor in youth who experience PTSD symptoms (Fincham et al., 2009). It is suggested that those employed by child protection agencies would do well to make available peer support groups and other forms of social support to act as a buffer against the experiencing of these symptoms.

In terms of limitations, the small size, non-probabilistic sample reduced the power, and due to the nature of the questionnaire used, a formal PTSD diagnostic could not be made. Specific trauma or adverse childhood experiences should be explored in future research to better target the most appropriate interventions and treatments for this population of youth.


Fincham, D.S., Korthals Altes, L., Stein, D. J., & Seedat, S. (2009). Posttraumatic stress disorder symptoms in adolescents: risk factors versus resilience moderation. Comprehensive Psychiatry, 50, 193-199

Greeson, J. K., Briggs, E. C., Kisiel, C. L., Layne, C. M., Ake, G. S. ., Ko, S. J., Gerrity, E. T., Fairbank, J. A. (2011). Complex Trauma and Mental Health in Children and Adolescents Placed in Foster Care: Findings from the National Child Traumatic Stress Network. Child Welfare, 90 (6), p. 91-108

Salazar, A. M., Keller, T. E., Gowen, L. K., & Courtney, M. E. (2013). Trauma Exposure and PTSD among older Adolescents in Foster Care. Social Psychiatry and Psychiatric Epidemiology: the International Journal for Research in Social and Genetic Epidemiology and Mental Health Services, 48 (4), 545-551

The challenge of measuring social support

In studying the relationship between family support and post-traumatic stress (PTS) symptoms in adolescents in group homes, we predicted that having a supportive family would be correlated with less severe PTS symptoms. We found a reliable scale, the Trauma Symptom Checklist (Wolpaw et al., 2005), for PTS symptoms, but had a difficult time deciding how to measure social support. There is evidence to suggest that a high level of perceived social support is associated with fewer trauma-related symptoms in adolescents (e.g. Bal et al., 2006), but we wanted to see whether or not family-specific support would mitigate PTS symptoms in traumatized teens. There are some validated measures to gauge social support generally (e.g. Sarason et al., 1983). However, they did not fulfill our requirements of differentiating between family members, an aspect we wanted to explore, as some researchers highlight the prevalence of sibling support (Milevsky & Levitt, 2005).

Since we did not find a measure to account for all the variables we considered important, we designed our own, which included a Likert-scale ordinal question about the quality of the support, a nominal question about the specific family member, and an ordinal question about frequency of visitation. The downside of this is that it has not been empirically tested and may have problems with validity and reliability. It has not undergone previous tests to check its interrater reliability or internal consistency. It does, however, have face validity, which Rubin and Babbie (2008) describe as appearing to measure what we intend to measure.


Bal, S., Crombez, G., Van Oost, P., & Debourdeaudhuij, I. (2006).The role of social support in well-being and coping with self-reported stressful events in adolescents. Child Abuse & Neglect,  27(12): 1377-1395.

Milevsky, A. & Levitt, M. J. (2005). Sibling support in early adolescence: Buffering and compensation across relationships. European Journal of Developmental Psychology, 2(3): 299-320.

Rubin, A. & Babbie, E. R. (2008). Research Methods for Social Work (6th ed.). Belmont, CA: Thompson Brooks/Cole.

Sarason, I.G., Levine, H.M., Basham, R.B., et al. (1983). Assessing social support: The Social Support Questionnaire. Journal of Personality and Social Psychology, 44, 127- 139.

Wolpaw, J.M., Ford, J.D., Newman, E., Davis, J.L., & Briere, J.(2005). Trauma symptom checklist for children. In Grazo, T, Vincent, G & Seagrave, D (Eds), Mental health screening and assessment in juvenile justice (152-165). New York: Guilford Press.

Emotional Caregiver Burden of Carers of Persons with Dementia: A Cross Sectional Study

Our research team conducted a cross-sectional, descriptive, correlational study on caregiver burden for patients with dementia. We hypothesized that females would report higher subjective burden than males, however our findings conclude no significant difference. The mean burden score for males was 50.5 while the mean burden score for females was 49.22. A larger sample size would likely yield different results. We hypothesized that hours spent caring for the patient with dementia (high versus low) would not affect burden levels. Examining our analysis of variance of burden scores by hours of care, we found a significant difference in the high hours of care group versus low hours of care group. This is contrary to our hypothesis, in that we assumed that emotional burden was suspended from differing hours of care provided. Finally, our analysis between household income compared to burden scores R=0.1 yields no association. While this tentatively supports our hypothesis, which assumed there would be no association, due to our small sample size and low correlation, these findings cannot be generalized, as they may have been caused by chance. As discussed in Rubin & Babbie (2008; chapter 10), our cross-sectional does not generate external validity. Because our sample is small, and potentially tainted, our findings cannot be generalized to a wider population of caregivers for persons with dementia. We also encountered problems of internal validity; since our study is cross-sectional in design, there was no way of determining whether our two groups were comparable prior to conducting our study.

Measuring Change in Postpartum Depression Risk among Women who Received Interpersonal Psychotherapy

Our study evaluated the efficacy of interpersonal psychotherapy on women at-risk of postpartum depression (PPD). The Edinburgh Postnatal Depression Scale (EPDS) was administered both to screen for eligibility and at post-intervention. Interpersonal psychotherapy significantly reduced PPD risk (p = .05).

The EPDS was developed by Cox, Holden and Sagovsky (1987) who modified 21 items from the Irritability, Depression and Anxiety Scale and the Hospital Anxiety and Depression Scale and constructed the remaining items. The tool was then administered to women know to have depression and non-depressed women in order to measure criterion validity (concurrent) and split half-reliability (Cox, Holden & Sagovsky, 1987). The authors concluded that the tool is valid and reliable, with a score of 13 and above indicating PPD risk.

The authors caution the EPDS is designed to identify depressive symptoms within the last 7 days and is not a diagnostic tool. A clinical assessment is required to confirm a PPD diagnosis as the DSM stipulates that symptoms must be present for 2 weeks (Cox, Holden & Sagovsky, 1987). This caveat is relevant to the validity of our findings as it is possible that administering the EPDS once at post-test may have measured sub-clinical depression risk rather than PPD risk. Matthey and Ross-Hamid (2012) recommend that studies using the EPDS repeat its administration for increased validity, however this suggestion was not adopted in our study. It is possible that this limitation decreased the power of our study and may explain why our findings did not reach greater significance.


Cox J.L., Holden J.M., & Sagovsky R. (1987). Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. British Journal of Psychiatry, 150, 782–786.

Matthey, S. & Ross-Hamid, C. (2012). Repeat testing on the Edinburgh Depression Scale and the HADS-A in pregnancy: Differentiating between transient and enduring distress. Journal of affective disorders, 141(2), 213-221.


Multilevel statistical model: What does it mean?

Is individual independent of context, the place or the time he lives in? Social or behavioral scientists would say, obviously NOT. Individual behavior tends to be shaped by his/her personal traits as well as by the factors of environment s/he lives in. Most of the data we collect in social and behavioral science research are hierarchical or clustered (Goldstein, 1999) in geographical areas or in periods of time. Individual characteristics vary within context as well as across contexts. There is a notion that individual behavior has to be examined in the context in which it takes place (Jones and Moon 1987). Hence, social and behavioral science researchers are in a dilemma. If they perform individual analysis as we do most of the time, they will miss the context, and if they undertake an aggregate analysis, they will miss individual level variation, and will approach to potential ‘ecological fallacy’, the invalid transfer of aggregate results to individuals (Subramanian and Jones, 2012). Hence, we are in need of a tool which will consider individual variation, contextual variation and the variation in individual-contextual interaction at the same time. Multilevel modeling (MLM) is such an approach which is able to conduct analysis simultaneously at the individual as well as contextual level. Recently, I participated in a training session on multilevel modeling organized by Quebec Inter-university Center for Social Statistics (QICSS). The session was conducted by Dr. Subramanian, a Harvard professor of Public Health and Geography. He has been teaching and working on MLM for more than twenty years. He is amazing in explaining MLM lucidly, and making it apprehensible to the beginners like me. I have developed a basic understanding of the concept and application of MLM, and I would like to share some of the basics of MLM.

Multilevel refers to the level of analysis where data tend to be nested in different structures. The tendency of data of being nested in structures is not ignorable. Let us suppose, we are interested to conduct a study on the performance of 8th grade students in the city of Montreal, and we want to find the association between students’ school performance and parents’ level of education. We, generally, follow multi-stage stratified sampling technique to select desired sample. Let’s say, there are 20 school districts in Montreal. We want to select five schools from each school district and 20 students from each school following simple random sampling method. Now, we get a three level data structure where students are level-1 units, schools fall in level-2, and the school districts are level-3 units. What sort of impression does this dataset give us? We might find that students’ performance varies on parents’ level of education. Isn’t it logical to think that the performance of the students also varies due to the variation in quality of schools, and the quality of schools also varies across the school districts? In other words, the students of a particular school might tend to be alike in performance and vary from other schools, and the schools in a particular district might be alike, and vary in quality from the schools in other districts. In an OLS regression model, we would regress students’ performance on parents’ level of education, and might find a pattern and a large volume of residuals or standard errors. We would dump the residuals considering wastage or noises as we don’t have anything to do with this garbage. Isn’t this volume of residuals what we term as garbage, produced due to our lack of mechanism to consider the variation of students’ performance for the variation in quality of schools, and for the variations across districts? Multilevel modeling provides us the tool to consider all these variations redistributing them across levels.

Let us assume that 8th grade students’ performance is measured in a single test across schools, which measures score ranging from 0-100, and parents’ level of education is measured in year of schooling. In a single level analysis or in an OLS regression we can use the following model to predict students’ performance:


In this model,β0, the intercept, gives us the average score of a student , and β1 , the slope, gives us the average change in score for an unit change in parents’ year of schooling. The intercept and the slope represent the fixed part of the regression model and provide us estimates of the average score and year of schooling relationship. The residuals, ε0  represent the random part of the model, and provide us the individual differences from the fixed regression line, the mean of which is considered 0 and the variance σ^ε0. If we extend this analysis into a multilevel with a two- level model, where students are level-1 units and schools are level-2 units, we will use the following model:


This is a combined model derived by substituting the macro model into micro model where both the intercept and the slop are variable across schools.

The micro model frames as:


and the macro model frames as


The fixed part of this model is β0+β1 , and the random part is eq6  .At level-2, we now, get two additional residuals, u0j and u1j in the random part of the model, which explain the variability of both the intercept,β0, and the slope, β1, respectively. This feature of analyzing variance in different levels makes MLM distinguishable from the standard linear regression models. The means of these new residuals are again 0, and the variances are  σ^u0 and σ^u1 the covariance is σu0u1  . We also have the level-1 residuals,, with a mean 0, and variance, σ^ε0  .

In this model, ‘i’ denotes the number of level-1 units, the 8th grade students, and ‘j’ denotes the number of level-2 units, the schools. Now, how do all these new parameters work in this model? Here, , the intercept β0, gives us the average score of a student, and u0j gives us the differentials of  across schools. The slope, β1 , gives us the average change in score for an unit change in year of schooling, and u1j gives us the differentials of  across schools. The level-1 residuals, ε0, provide us the individual differences of students.

MLM provides us the scope to partition variation according to the different levels, and provide us a new statistic known as intra-class correlation, or intra-unit correlation, or variance partitioning coefficient denoted by Greek letter rho,ρ, which takes values between 0 and 1. If ρ approaches 1, it implies an ecologic model indicating that the students within a school are highly similar to each other in terms of their score. If  ρ approaches to 0, it implies independence between students within a school indicating that the source of variation in the scores is mainly at the student level. Now, we get an impression whether we should go for a multilevel modeling or remain with standard linear regression models.

I am looking forward to using MLM to analyze Program for International Student Assessment (PISA) dataset of 15-year-old students’ performance in financial literacy in the 18 countries and economies in 2012. This secondary dataset is available to make comparative analysis of financial literacy globally. It is, now, time to make our observational data more meaningful.


Goldstein, H. (1999). Multilevel statistical models. Institute of Education, Multilevel models project. London

Jones, K., and Moon, G. (1987). Health, disease and society. London. Routledge.

Subramanian, S. V., and Jones, K. (2012). Multilevel statistical models: concept and applications. Harvard University. Massachusetts

« Older Entries
Blog authors are solely responsible for the content of the blogs listed in the directory. Neither the content of these blogs, nor the links to other web sites, are screened, approved, reviewed or endorsed by McGill University. The text and other material on these blogs are the opinion of the specific author and are not statements of advice, opinion, or information of McGill.