Another 'futile quest'? A simulation study of Yang and Land's Hierarchical Age-Period-Cohort model

Educational expansion and the education gradient in health: A hierarchical age-period-cohort analysis

“…This model was developed by Yang and Land (2006, 2008 as a solution to the 'identification problem', induced by the exact linear relationship of the age, period, and cohort categories (Period = Age + Cohort) (Bell & Jones, 2014a). The HAPC model acknowledges individuals as simultaneously nested in periods and cohorts.…”

“…Clarke et al, 2009;Reiter et al, 2009;Stegmueller, 2014;Tawfik et al, 2012), Bell and Jones (2013a, 2013b, 2014a questioned the validity of this approach. They asserted that it is impossible to assess the net effects of time-related changes accurately, without making the assumption that at least one of the age-period-cohort effects is equal to zero.…”

“…In more recent cohorts, the low educated can compensate for the accumulation of education-related disadvantages through increased access to resources, facilitated by the higher educated. Hierarchical age-periodcohort models lend themselves to unraveling these complex intersection processes (Bell & Jones, 2014a).…”

Life-course and cohort trajectories of mental health in the UK, 1991–2008 – A multilevel age–period–cohort analysis

“…Whilst this has been part of the literature for decades (Glenn, 1977;Mason et al, 1973;Ryder, 1965), serious misunderstandings remain across the social sciences (Bell & Jones, 2014b). This section clarifies these misunderstandings before considering their relevance to understanding changing mental health.…”

“…Third, birth year (our measure of cohort) is exactly collinear to the individual mean of age. Thus, the inclusion of cohorts in the model means that age effects are already the corrected within-individual effects (Bell & Jones, 2014b). …”