Who's Voting for Who? Ontario Political Parties and Voter Segments

To see how our Ontario demographic segments relate to votes for different parties in Ontario's elections, we decided to look for some basic correlations (wiki) in the voting data.

Here are the results. You might consider this a test, to see if the segments make intuitive socio-political sense.

Take a look at our Ontario Geo-Segments Page for demographic information about neighborhoods, and ridings as well as a searchable map of Ontario!

The signs indicate a significant positive (+) or negative (-) correlation between an Ontario political party and each segment type (cluster).  

Note that this table doesn't show change between one year and the next, but instead whether a type can be said, with relative certainty, to support or to be hostile to a party, or indifferent to it altogether, for one election.

Turnout is included too, to measure whether a group tended to go to the polls (+), stay home (-) or follow the general tendency (blank).

Significant Correlations : Ontario Political Parties and Clusters 

 

Type	Cluster Number (deprecated)	Liberals 2011	NDP 2011	PC 2011	Green 2011	Turnout 2011
A01	511
A02	521		-
A03	522
A04	523
A05	531	-		+		-
A06	541					+
A07	551
A08	561	+	-			+
A09	562
A10	563		-		+
A11	564	+	-		-
A12	565
A13	566
A14	567	+	-
B15	311
B16	312	-		+		+
B17	313	-	-	+	+	+
B18	314		+			+
B19	315	-	-	+	+
B20	321	-
B21	331	-	+			-
B22	341	-		+	-	+
B23	351			+		+
B24	352		+	-	-
Type	Cluster Number (deprecated)	Liberals 2011	NDP 2011	PC 2011	Green 2011	Turnout 2011
C25	211
C26	221		+	-
C27	222	-	+	-
C28	223	+		-	+
C29	231	-	+	-	+	-
C30	241		+	-		-
C31	242	+		-
C32	251
C33	261			-
C34	262
C35	263
C36	264		+		-
C37	265
D38	111					-
D39	121					-
D40	122	-				-
D41	123	+		-	-	-
D42	124	+			-	-
D43	131					-
D44	132
D45	133		-			-
D46	141				+	-
D47	142	+	-			-
D48	143	+	-	-	-
E49	411	+				+
E50	421		+	-	-	-
E51	431	+		-	-	-
E52	441		+	-
E53	451	+			-
E54	461	+		-
Type	Cluster Number (deprecated)	Liberals 2011	NDP 2011	PC 2011	Green 2011	Turnout 2011

Here are 2007's scores:

Type	Cluster Number (deprecated)	Liberals 2007	NDP 2007	PC 2007	Green 2007	Turnout 2007
A01	511
A02	521		-		+	+
A03	522			+
A04	523					+
A05	531	-	-	+	+	-
A06	541					+
A07	551
A08	561	+	-			+
A09	562
A10	563	-			+
A11	564		-	-	+	+
A12	565	+
A13	566
A14	567	+	-
B15	311
B16	312	-			+	+
B17	313	-	-	+	+	+
B18	314		+
B19	315	-	-		+
B20	321
B21	331		+		-
B22	341	-		-	+	+
B23	351		-			+
B24	352		+	-	-
Type	Cluster Number (deprecated)	Liberals 2007	NDP 2007	PC 2007	Green 2007	Turnout 2007
C25	211
C26	221	-	+		-
C27	222	-	+		-
C28	223			+	-
C29	231	+	+		-	-
C30	241		+		-	-
C31	242
C32	251
C33	261	+			-
C34	262
C35	263	+
C36	264	+	+	-	-
C37	265
D38	111	+			-
D39	121	+		-	-	-
D40	122	-		+	+	-
D41	123	+			-	-
D42	124	+	-	-		-
D43	131					-
D44	132
D45	133		-			-
D46	141	+	-			-
D47	142	+	-
D48	143	+	-	-	-	+
E49	411					+
E50	421		+	-	-	-
E51	431	+			-	+
E52	441	-	+		-
E53	451	+		-
E54	461
Type	Cluster Number (deprecated)	Liberals 2007	NDP 2007	PC 2007	Green 2007	Turnout 2007

Boring Methodology Stuff

Remember that correlation does not imply causation! And that correlation does not by itself indicate slope, or the "intensity" of the support of a particular segment for an Ontario political party, merely that there is quite likely a relationship. We have established significance here as p > 0.05.

Analyses were conducted at the riding level, and not the polling division or dissemination area level, possibly exposing us to some kind of environmental fallacy.


We should point out that the party political information was researched after we named the groups. We didn't just give the good names to the groups whose voting record we liked best.