About
8
Publications
1,141
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
18
Citations
Introduction
Publications
Publications (8)
Across Canada, hate crimes, especially those motivated by race, ethnicity, or religion, are still prevalent. For example, in 2019, 46% of police-reported hate crimes were motivated by race or ethnicity, and 32% were motivated by religion (Moreau, 2021). In Canada, Muslims are the second most targeted religious group in terms of hate crimes. However...
Governments around the world are pressed to invest in postsecondary education. However, little research exists to document whether large-scale educational reforms aimed at increasing rates of postsecondary attendance benefit youth’s developmental outcomes. This study tested whether an educational reform occurring in Québec, Canada, in the 1960s inc...
Cannabis has been documented as a preferred substance for sexualized drug use (SDU) in several different populations, including men who have sex with men (MSM; Edmundson et al., 2018; Hurley & Prestage, 2009; Parent et al., 2021; Schmidt et al., 2016). Populations that engage in cannabis use experience both positive and negative sexual outcomes, wh...
Recent studies suggest that educational attainment may be protective against the development of adult criminality, particularly for children who are aggressive. Children with strong academic skills would also be less likely to commit criminal offenses, which might be due to the relation between achievement and education. Additionally, other risks (...
Children from poor neighbourhoods showing early aggressive behaviour are at risk for criminal offending. The role of education as a mediator, neighbourhood disadvantage and aggression as moderators for criminal offending were examined in a lower-income, community sample (n = 3,521; 48% males), across a 40-year period from childhood to mid-adulthood...
Questions
Question (1)
I ran an LCA and the BIC indicates a 4 class model fit is better than the 3. The literature would support a 3 or 4 class solution. The problem is that one of the classes has only 5 cases. I was looking for references that would support marking those cases as missing and running with the only the three classes when I run a regression predicting these classes. I would use the 3 class LCA but I still only get 23 cases in one class when I force the 3 class fit. I can't find a specific reference that says I can use only certain classes (in this case 3/4) from the 4 class solution.
Thanks!