Wednesday, April 8, 2009

A brief aside...or Stop (if you will)

So it's going to be a while before I post Archaeodigms part 3 as I'm quite busy now and I've got a ton of other work to get done before writing that up. I can however update you some of the work that I'm currently getting done (woooooo...?)!

For those of you who read Steve Chrisomalis' blog Glossographia (and for those of you who don't...you should!) you may remember him discussing the Stop Toutes Directions project. I'm not really going to explain the project here because I just don't have the time, but I encourage you to check the website out if you are interested. In a nutshell it's a project analyzing various aspects of the distribution of Stop Signs in Montreal, Canada. A few of Steve's former students (myself included) are still working with him to further analyze our large Stop Sign database in hopes of putting together an edited volume for publication. My contribution to the project is producing a GIS of spatial linguistic variation of stop signs vs. spatial linguistic variation of language spoken in the home using census tract data. I've done the GIS as a project for my socio-economic application of GIS class, which I presented today, but I wanted to share the preliminary GIS maps I produced for the project and give a little bit of preliminary interpretation for you all. The following maps are of census tracts we surveyed in the central Montreal area.

Ok, so. The colour gradient of the CTs on this map shows the proportional breakdown (in a percentage) of Stop Signs in central Montreal, the more red the CT polygon is, the higher the proportion of signs in that CT that are 'Stop' signs, the more blue the CT polygon is, the smaller the proportion of 'Stop' signs in that CT. The little pie charts indicate the proportion of language spoken in the home for each CT (taken from census tract data), with red indicating anglophone homes, blue indicating francophone homes, and green indicating bilingual homes. (Click the image for a larger view)

This next map is essentially the last map inverted, with more blue CTs having a higher proportion of 'Arret' signs and more red CT's having a lower proportion of 'Arret' signs. The pie charts represent the same thing as the last map. (Click image for larger view)

This last map is the coolest (I think) in that it tells us the most. In this map the gradient goes from yellow/tan to blue, with yellow/tan CTs representing a low proportion of 'Arret/Stop' signs, and the most blue CTs representing the greatest proportion of 'Arret/Stop' signs. Again, the pie charts still represent the proportions of language spoken at home CT data. (Click image for larger view)

So what does this mean? Well if you look at the maps you can see that there is some distinct clustering and patterns showing up in some areas of Montreal and not in others. In north-western and downtown Montreal there seems to be at least a correlation between a dominance of English spoken in the home and a dominance of 'Stop' signs. In eastern Montreal there also seems to be a correlation between a dominance of French spoken in the home and a dominance of 'Arret' signs. The most interesting pattern is highlighted in the last map. If you take a look at it all of the CTs which have the highest proportion of 'Arret/Stop' signs are correlated with predominantly anglophone areas. Why is this interesting? Because of language laws in quebec which technically make clearly bilingual signs illegal (as both 'Stop' and 'Arret' signs are technically french and thus legal, whereas 'Arret/Stop' signs, which are logically bilingual, are illegal). Now I don't have the time to get into the peculiarities of Quebec language politics, but I think this is enough info to at least give a bit of a primer on the situation. Though I've only shown some correlations here (and we all now that correlation doesn't necessarily mean a causal link), it seems clear that it's predominantly anglophone CTs where the law is being broken most often. This could be because anglophones are a minority in Montreal (and Quebec), and are thus unlikely to report language law infractions, and thus assist in enforcing a provincial language law that puts them at a disadvantage, whereas in predominantly francophone CTs there is perhaps a higher chance that illegal signs are reported as the language laws are intended to preserve the French language and thus benefit Francophones (I think in a cultural heritage sense?). Anyway, this is just a working hypothesis and a preliminary analysis, and some more statistics need to be run, but I thought it was pretty interesting, and I hope you do to!

This post turned out to be a lot longer than I expected, which means I need to get back to work! Until next time (which will hopefully be Archaeodigms part 3)!

1 comment:

glossographia said...

Oh, wow, you weren't kidding! Those really look fantastic and professional! I'm sending you an email right now with a couple of questions/suggestions. Thanks for this!

'Stopmaster' Steve