Note: This was a bilingual chat session, which means that the participants were able to submit their questions in English or French. Statistics Canada respects the Official Languages Act and is committed to ensuring that information products of equal quality are available in both English and French. For that reason, all the questions and answers have been translated in the other official language.
Moderator at 12:44:49
Yes, our experts are currently answering all of your questions. They will appear in the live chat feed shortly. Thank you for your patience!
Adam Cotter at 12:49:17
Hi, and thank you for your question. The analytical data file is available in our Research Data Centres, and we are releasing a public use microdata file in the summer. As for your question about sample size, it would depend on the city and area that you're interested in - in the report, we present information at the census metropolitan area level, but there are also lower levels of geography available on the file, such as census subdivision or census tract, which would be closer to neighbourhood level.
Julie Sauvé at 12:52:23
Thank you for your question.
More detailed data could be available for the largest CMAs. However, for less common social problems such as people being attacked because of their skin colour, the data may not be releasable.
Adam Cotter at 12:52:46
Hi, and thanks for your question. For the questions about disorder, we ask respondents whether the different types of neighbourhood disorder are big, moderate, or small problems, or not problems at all. So for those responses, they are tied to whatever someone perceives their neighbourhood to be.
When we're looking at information by certain types of neighbourhood characteristics (for example, Table 5), this information about the neighbourhood is based on the census tract (and on the census subdivision for areas not broken down into census tracts).
Adam Cotter at 12:56:37
Hi, thanks for your question. The survey does not collect information at the neighbourhood level specifically, although it is possible to examine the data at the census subdivision or census tract level, with results being subject to sample size of the area of interest. These questions have also been included in the General Social Survey on Victimization in 2004 and 2009.
Overall, though it's not specific to neighbourhood, perceptions of disorder were relatively stable across the provinces when we compared results from 2004 to 2014.
Tamara Knighton at 12:57:11
Thank you for your question. Data quality becomes an issue when presenting results at a level of detail below the census metropolitan area (CMA). Consequently, there are currently no plans to break down the data at a lower level of geography.
Tamara Knighton at 13:07:27
A previous report examined this relationship and found that people who reported the presence of social disorder in their neighbourhood recorded a rate of violent victimization almost three times higher than people who did not perceive social disorder. Please see the following link for the report: Juristat – Criminal victimization in Canada, 2014.
Adam Cotter at 13:09:43
Hi, and thanks for participating. The data in the report is based on individual perceptions, which are not necessarily reflective of objective conditions. While it would certainly be interesting to explore perceptions by more specific neighbourhood characteristics, we were unable to do that using the GSS on Victimization. One way we tried to take into account neighbourhood characteristics was by looking at the demographic characteristics of the neighbourhood. Although these characteristics are not measures of disorder, they do provide some contextual information about the neighbourhood and how they are related to perceptions of disorder.
The focus on perceptions also takes into account the "threshold effect" that is sometimes referred to in the academic literature. Basically, the same objective conditions may be perceived as a problem by one individual but not by another. Not only are perceptions of disorder based on the number, frequency, or seriousness of the issue, but they also depend on personality, experience, location, and time of day, or many other factors.
Julie Sauvé at 13:14:40
Thank you for this question.
We did not analyze this in the article that was just published. The differences among various countries could be related to various factors that are not measured by the survey, such as a different culture or history. However, the questions in the Canadian GSS are based on questions in the England and Wales crime survey. It should be noted, though, that there are slight differences in the wording of the questions.
Adam Cotter at 13:16:18
The model that is published in Table 7 was estimated simultaneously in order to determine which variables remained associated with perceptions of neighbourhood disorder, other factors being equal. First, a model was developed only for neighbourhood characteristics. Second, a model was developed, which included only the demographic characteristics. The final model combined these variables and tests each variable for significance with all other variables in the model held constant at their average value.
Tamara Knighton at 13:16:26
Thank you for your question and I'm pleased that you enjoyed the report. A similar report was prepared using 2004 data and can be found at the following link: Life in metropolitan areas.
Moderator at 13:25:49
You're welcome. Thanks for participating!
Adam Cotter at 13:29:46
That's an interesting question. We know that neighbourhood composition is one factor, among many, that is associated with perceptions of neighbourhood disorder. Since we only ask respondents whether or not they perceive neighbourhood disorder and not why they perceive it, our data don't provide the answer to "why".
Moderator at 13:31:37
Yes, the full transcript will be made available on our website shortly.
Julie Sauvé at 13:32:43
Thank you for the question.
In the survey, when a person responds that one of the social phenomena presented constitutes a big problem or a moderate problem, we consider that the person perceives disorder in the neighbourhood. On the other hand, when a person responds that the phenomenon in question is a small problem or not a problem at all, we consider that he or she does not perceive disorder in the neighbourhood. It is useful to take the "threshold effect" into consideration. In other words, the measurement of disorder in the neighbourhood is based on perceptions and not necessarily on objective conditions. The threshold effect refers to the notion whereby the level at which a behaviour or situation becomes problematic differs for each person, based on a number of factors.
The threshold effect is therefore a reminder that perceptions of disorder are not only based on the number, frequency or seriousness of the situation, but are also related to the person's personality, experiences and location as well as the time of day.