Survey Sample

Hey guys,

Clearly it is too difficult to approach a nation-wide immigrant sample on immigrants, so our full sample will likely reside in BC. However, how does everyone feel about keeping our sample study within the lower mainland? For sample purposes, this seems to be the most realistic approach we can take as even with a province-wide sample it'd be essential we include samples from Vancouver Island, Okanagan, Interior etc. and I'm unsure on how we could go about and finding those samples. Thoughts?

Hey Cyrille,

I think sticking to lower mainland is definitely the smart thing to do. I think taking a random sample from lower mainland, which has a rich immigrant population, would definitely provide us with some really good results. I think maybe we all want to state which region of the lower mainland we reside in (i.e. UBC, Vancouver, W. Vancouver, Burnaby, Richmond, etc.) and we can all collect 6 respondents (I think that's what the prof said, if I'm not mistaken?) from each region.

Some people previously brought up the suggestion of sticking to UBC students. I think the major flaw here would be the age and occupation range - i.e. younger population who may not have had full careers, or even those who may have moved here too young to properly recount their first few years of settlement experiences. If we were to especially go along the route of asking about previous income status, current income status, etc. students wouldnt be able to really provide a useful response as they would have in majority relied upon their parents.

I'm not exactly sure how data-collection works- as I understand it we're expected to collect data on our own. This already violates a significant principle in surveying- random sampling, and this is just an entry level stats course and for most of us our first conducted survey. I don't see a problem with recognizing geographical limitations such as focusing ourselves on the Lower Mainland.

@Min, I don't think sticking to UBC students is a bad thing. If you wanted to study specifically how say, international student or new immigrant student's wealth correlates with their family's wealth etc. then using UBC students would be a very good way to accomplish such a thing. We could also expand it to include SFU students, Langara etc. However, no suggestions have yet been made to focus on students, so this is largely irrelevant.. but it all depends on exactly what we're trying to look at!

I could be wrong, but I was under the impression that Dr. Owen is putting all of the questions (as in ours + everything from all the other topics) and we are supposed to all administer the complete survey. That's why I think it's a little odd that so many of the research questions we are coming up with are specific to just immigrants. Again, I could be totally wrong, but I thought that our sample would be the same as everyone's, which means that there would certainly be locals. So, are we then expecting non immigrants to just skip our part? Sorry if someone has already addressed this, the thread is getting slightly hard to follow, and everyone's responses are quite lengthy so maybe I missed this.

I'm not too sure whether we are administrating the complete survey (i.e. surveys are compiled of the whole class's questions) or we are simply handing out just our survey with 5 questions, but either way our issue is immigration settlement, thus looking at immigrant's experiences and the data we collect them would be our focus. If the first arrangement is the case, then those who aren't considered immigrants wouldn't answer our survey questions, since their answers would be of no use to us (that is, if we are sticking to the two surveys proposed thus far). If the latter is what the professor wants us to do, then the immigrants of lower mainland would be our individual units of analysis.

Midas - My point of not wanting to stuck to solely UBC students was based on the idea that we might be going along with the research proposal of economic status before arriving in B.C. affecting opportunities and settlement progress thereafter. My reasons for were obviously because students wouldn't necessarily be able to know their family's financial status beforehand, nor perhaps be able to access such information from parents/guardians. If we were to go along the route of the ESL research topic, sticking to university students, as you've mentioned UBC, SFU, Langara, etc. would definitely not be a bad idea.

Since we are doing immigrant settlement, limiting the survey to just UBC students would limit the scope of our investigation. I don't think using UBC students as our whole sample would be productive just because student immigrants are probably less exposed to the realities of immigration than say their parents, or those who come to find employment and make a new life. School itself is a social institution that helps to promote integration and cultural exchange, as well as a support system, so students already have a leg up from those immigrants who lack these opportunities.

What do you guys think of targeting specific groups. Like for example in the lower mainland, there are specific areas that tend to attract one type of immigrant (e.g. Richmond=Chinese, Surrey=Indian, Commercial Drive=Italian etc.)

I think that communities provide social structure/support to immigrants so even if you come not knowing the language in vancouver, you're bound to find somewhere to fit in. Do we want to control this factor or limit it?

I completely agree with you Jessica, targeting a student population would provide us with little valuable knowledge on the realities of immigration settlement. Since international students come to UBC to study, they don't usually work more than p/t and their education is often paid for by their families, loans etc. Their views immigration settlement would be much different from others who have settled in BC in search of work/better opportunities for their family rather than educational purposes. If we can somehow branch outside of UBC and look at immigrant populations throughout the lower mainland that would be extremely significant towards our proposed research.

Since it seems we are sticking to Lower Mainland, instead of targeting specific communities depending on their ethnic background, how about targeting certain communities depending on income? This would split regions of lower mainland up, i.e. vancouver west, vancouver east, richmond, burnaby, downtown vancouver, etc. then these areas perhaps into areas that we know are better off and areas that are poorer? i.e. residents of yaletown vs. residents of chinatown/gastown. This may have more to do with the income aspect, rather than ethnic aspect of immigrant settlement.

I know this is probably no helpful since it's all speculation but I think Prof. Owen wants us to take on all 50 of the survey questions, and not just the 5 we will come up with. I agree that it's a little bit... "fruitless" that many of our questions seem to specifically target new immigrants. I suspect Prof. Owen wants us to find a sample of convenience, as we obviously don't have the resources to travel up and down the province. I suppose we will find out about how to correctly select the sample and carry out the survey this week or after the reading break.

I like the idea of artificially separating Vancouver into various regions, so long as we specifically notate the process and why we're doing it. It's inevitably going to end up being subjective and normative anyway given our time, spatial and resource constraints- however this process will significantly lighten our survey load while allowing some increased diversity in response.