Change in Young Adults Surprise

For this map are interested in the change of the share of young adults, that is 25 to 39 year olds, among the total population in each area between 2011 and 2016. Overall in Canada, the share of young adults increased from 12.9% in 2011 to 13.1% in 2016, or by 3.76% for each net new person in Canada between these years. We use this to model the expected share of young adults in 2016 in each area as the respective share in 2011 plus 3.76% for each net new person in that area. Thus, for Canada overall, our model matches the 2016 numbers by design. For all other areas we colour each region by the Bayesian surprise of how the share of young adults in 2016 differ from our model. Intuitively we colour each region by how consistent our model is with the observed data. The white areas show where our model is consistent with the data. The greener the area the less consistent the model is with the observations and the observations show more young adults units than our model predicts, the pink areas fewer young adults than the model predicts. For more details on how Bayesian Surprise Maps work please refer to our blog post or this excellent paper that this approach is based on. Among the provinces, Quebec saw a surprisingly large decrease in share of young adults, followed by New Brunswick with a more modest decrease. Saskatchewan and Ontario saw surprisingly large increases, closely followed by Manitoba and British Columbia. Among the larger Census Metropolitan Areas in Canada, Toronto and Vancouver displayed the most surprising increase in the share of young adults, closely followed by Edmonton. Montreal registered a frighteningly surprising decrease in young adults, followed Québec city, Ottawa and Calgary. The cities of Toronto and Vancouver are the drivers of the surprising increases in their census metropolitan areas, and drilling in further can help further pinpoint areas that saw surprising increases. Our surprise model is somewhat conservative in trying to distinguish changes from statistical noise, so when zooming into the census tract and further to the dissemination area level many areas will fade into statistical insignificant white with only some standing out to point toward surprising increases and decreases as model confidence diminishes with smaller populations in each area.

Author: CensusMapper Team

Dataset: CA11F, CA16