I need some ideas on gathering my samples for a survey

[Saint_Cad]

Everything above this is unimportant for what I need. It is not helping when it is hijacked with methodology. Trust me people - I know what I’m doing

@md-2000

Can you tell by sight who is Turkish? Are they considered “middle eastern”? What about Uzbekis? Are Nubians black or middle eastern? ("what’s a “Nubian”?) Also, I don’t see a distiction between south Asian and East Asian, two distinct ethnic classes.

It will be self identified. So maybe Moors will identify as Middle Eastern and not African.
BUT IT DOESN’T MATTER FOR WHAT I AM DOING SO PLEASE DO NOT RESPOND WITH ALL THE FAILURES OF SELF-IDENTIFYING VARIABLES. I KNOW ALL THAT.

@md-2000

The debate about “equity” is not irrelevant. Many people have heard the word, like me, and are aware of its dual applications. But I had to read one of the links to understand exactly what the difference was between equity and equality. So the problem is, even if you find enough “test subjects” is their answer meaningful? Or are they describing barriers to equality?

For what I need, yes. Again, let’s not worry about the definition, I have that covered.
I only need the question answered how I get 50 people from each group.
That’s it.
I have done scientific studies before with statistical work. I am not completely ignorant on how to do a professional study as you all think.

I just need ideas on the logistics of collecting the sample. That’s all. So please all, STOP WITH THE CRITIQUE OF THE METHODOLOGY!!! Just answer how do I collect 50 samples from each group.

@md-2000

Finally, some people may be offended if you try to ask their ethnic group. Are you instead going to guess their group? Or correct it based on visual cues if they are difficult? What about mixed? Keep in mind, Tiger Woods for example is the most successful Asian-American golfer. Plus, some people may be offended that you seem to be focusing on specific ethnic groups, so you can’t be obvious standing on a corner and asking only, say, apparently black people.

Already answered. If a person does not wish to participate, for example a Pacific Islander because their group is not represented or a person feels they cannot self identify, FOR WHAT I NEED, that’s OK.
“Oh, Saint Cad, the lot of us need to explain to you in excruciating detail how that is bad practice for professional research.” Don’t. Just don’t

@md-2000

Note I’m assuming you live near an area where there are in fact an assortment of ethnic types so you can make your quota.

I’m not which means online or having others farm out to others.

[Saint_Cad]

OMG I owe you a beer. Never even thought of Reddit. Even if it doesn’t give me the information I need it’s a great idea.

[Chronos]

@Saint_Cad

So please stop with the advice on methodology or definitions or whatnot. I know you are all trying to help but if this were a publishable study I know how to do all that. You are not helping so please … I’m begging can you all just stick to the question I asked and stop STOP STOP with the methodology.

Wait, what are you asking for, if not for methodology?

[TroutMan]

One suggestion: if the people responding to your survey don’t give the right answer or don’t understand the question, maybe don’t yell at them.

[Eyebrows_0f_Doom]

@puzzlegal

Diversity, equity, and inclusion is a catch phrase now. I understood what he meant by equity.

I thought this was a question about barriers to making money. Equity=how much money/assets you have.

[Saint_Cad]

@Chronos

Wait, what are you asking for, if not for methodology?

@Saint_Cad

How can I find 50 people from each group to answer whatever question I ask?

I know technically that is part of the methodology, but look at the responses I’m getting. They go way beyond how to find people for a survey. See below

@Orville_mogul

If there are other explanatory variables, for example income, gender, or education, that will complicate things.

Not important

@Orville_mogul

It should ideally be administered under a constant set of conditions to avoid other biases. Given the relatively large number of people you will need to reach, I would suggest a short written survey.

Not important for what I need. Doesn’t answer how I find my sample

@Orville_mogul

Finally, you will also need to think about the format of your question and how you intend to analyse the results. Multiple choice options will be a lot easier to analyse than open-ended responses, but will give less insight.

Not important for what I need. Doesn’t answer how I find my sample

@Orville_mogul

Which statistics techniques/skills is your teacher expecting you to demonstrate? I would imagine a stats class being interested in tests of significance etc.

Not what I need

@Orville_mogul

For that your question really should ideally produce a quantitative response for which a mean and standard deviation can be calculated. An open-ended question is OK for social research, but doesn’t lend itself well to usual statistical analysis

I need qualitative data, but again not needed to know how to find the people for my survey

@Orville_mogul

Multiple choice questions could be OK, but the responses should fall along a continuous spectrum, which would be hard in your case.

No, it doesn’t for what I need. But regardless doesn’t answer how to find the people for my survey

@Orville_mogul

Have you vetted your research question with your teacher?

I am the teacher generating data for my students.

@md-2000

First - is the statistic meaningful if you specificaly target certain groups by certain factors?

Doesn’t matter in finding my 50x5 people for my sample.

@md-2000

The debate about “equity” is not irrelevant.

For my FQ yes it is since the definition of equity has nothing to do with finding my sample.

[Saint_Cad]

@echoreply

Also, as your collection scheme gets more complex, the chances of having it come to the attention of someone at your school/district who is upset by it increases.

It actually aligns to district instructional ideals.

@echoreply

I know in higher-ed doing something like this may have to go through IRB and human subject approval, though it could probably be exempted if no personally identifiable information is requested, but in that case getting the exemption is the important step.

I teach that to my students when I taught Stats in college. For what I am doing it is not needed since results are completely anonymous and there is no treatment, simply responses to a survey so I do not need any clearance.

[StarvingButStrong]

As for finding your subjects:

My suggestion is to find places where people generally end up waiting in line. If you’re looking at even a ten minute wait to give your coffee order, you’re more likely to be bored enough to answer a couple un-important-to-you questions. So, how about people waiting to buy tickets for movies or shows or exhibits or anything like that?

I’d also suggest hitting a couple of different locations, maybe in or close to cities to be more likely to get a varied ethnic group.

[md-2000]

@Saint_Cad

@md-2000

The debate about “equity” is not irrelevant.

For my FQ yes it is since the definition of equity has nothing to do with finding my sample.

Sorry to go off on a tangent, but maybe if the lesson is statistics and sample selection, not answering social issues, perhaps pick something which involves less of a socially “touchy” issue than race (and consequent financial/career success).

So, if you don’t mind me asking - what is the goal of the exercise? To see how difficult it is to get answers from random people, or to critique the meaningfulness/randomness of the sample achieved, or standard deviation calculations?

[Yllaria]

@Saint_Cad

I am the teacher generating data for my students.

Are you going to clean up the data? That is, will you summarize and categorize the answers, so that they can be added up? Or will that be left to the students? The question is an essay prompt, so the answers are going to be wide and varying. If you don’t limit the field for the response, you’ll get answers pages long.

Will you throw out answers that obviously don’t apply to “social equity”? Or are such answers considered a separate category? Would they be the same category as random threats or abuse?

[kenobi_65]

@Saint_Cad

How can I find 50 people from each group to answer whatever question I ask?

Speaking as a professional market researcher: if you want something approaching a representative sample for each of your groups (rather than “people who you and your friends know”), and if you want to make sure that, within each group, you get a good spread of ages, gender, income level, and geography, you’re going to need to buy sample from a reputable market research company.

You’re looking at a total sample size of 250 (50 each for five groups), and I’d count on that costing somewhere in the mid 4 digits. (That assumes you just need them to provide respondents, and that you are hosting your survey yourself; if you need a research company to program your survey and host it on their website, it’ll be even more.)

[Saint_Cad]

@md-2000

Sorry to go off on a tangent, but maybe if the lesson is statistics and sample selection, not answering social issues,

The lesson is not on sample selection otherwise I would have my lesson on that. I’m just collecting data.

@md-2000

So, if you don’t mind me asking - what is the goal of the exercise? To see how difficult it is to get answers from random people,

No otherwise I would have the students collecting the data

@md-2000

r to critique the meaningfulness/randomness of the sample achieved

No

@md-2000

standard deviation calculations?

With qualitative data?

Don’t worry about what I’m going to with the data because if I do the thread what yet again devolve into people giving me advice I don’t need , so how about we worry about my actual question and not all this tangential stuff I already know

[Saint_Cad]

@Yllaria

Are you going to clean up the data? That is, will you summarize and categorize the answers, so that they can be added up?

Yes they will get the frequency tables of the responses.

@Yllaria

Will you throw out answers that obviously don’t apply to “social equity”?

Yep.

@kenobi_65

Speaking as a professional market researcher: if you want something approaching a representative sample for each of your groups

I don’t. This is not for a professional article. I just need some authentic data for students to work with and I hate contrived data.

[DPRK]

@Saint_Cad

Ok but let’s get back to the OP because the highjack on “what is equity” is not the question I asked.
How can I find 50 people from each group to answer whatever question I ask?
If it makes it easier to stop hijacking the thread. assume I need to ask 50 members from each group: Anglo, African, Asian, Latinx. Middle Eastern. “What’s your favorite ice cream flavor.”

First you need a random sample:

Sampling (statistics)

What is the population of interest, and how are you going to pick them and account for mode effects?

Survey data collection

You also need to prepare a proper survey. Don’t forget to hire some translators!!

It sounds like one of the questions will be, “Please select one or more of the following…” with options like Anglo, African, Latinx… Some people will check off 3 of them, others zero, you need to be prepared for all of that.

Then you have your free-form answers to “What is the biggest barrier…” to quantify

Survey methodology

And the actual quantitative analysis, dealing with latent variables, etc

There is a whole lot to do properly, and potential to screw up, at each stage.

[DPRK]

P.S. if you just want a simple, not necessarily rigorous, method of selecting people, how about just starting with a list of every resident of the city and choosing a random subset? Or a random subset of everybody over the age of 16 (or whatever) to exclusen really small children?

[Saint_Cad]

None of this has anything to do with what I asked for. Except maybe the random sampling but it doesn’t need to be random, cluster or convenience for example is fine.

I feel you all think I’m ignorant on how the do research. I’m not. I wrote a thesis for my master’s, numerous research papers for my doctoral program and I taught how to do research when I taught college-level statistics and the dean felt that I did a great job at it. So please stop with all of the stuff that assumes I do not know how to research. Quit assuming I’m trying to do research at a professional level. I know what I’m doing. The question to answer is
How do I find 50 Middle-Eastern people to answer a question?
Repeat for the other four ethnic/cultural groups.

Can we just limit answers to that question please?

[DPRK]

I apologize if my response was not clear or not helpful. Again, what I was suggesting is roughly along the lines of: you get the Chicago telephone directory. Now you have a list of names and how to get in touch with them. Randomly choose [a simple calculation is required to estimate how many you will need, taking into account that only 1/3 or 1/10 (or whatever) will respond, and there are some but not that many Tajiks in Chicago so if you really want a chance of netting 50 of them…] Of those who respond, one of the questions will be a multiple-choice question regarding ethnicity. With broad categories like “Middle-Eastern”, “European”, “African”, “Asian”, … you should not have a problem getting up to 50.

[Saint_Cad]

That seems like an inefficient way to do it.

[DPRK]

I admit I don’t [however, see below] know the most efficient way of doing it: online social media may be much more efficient than ringing strangers up. Especially if we don’t care about the most clean, rigorously randomly random sample. Also in that case if you need “50 Middle-Eastern people” there are dirty tricks like predicting likely ethnicity by first name.

It almost goes without saying that the most efficient would be to pay for some of the corporate AI/algorithmic/targeted advertising/spammer data sets which have scarily detailed data about everyone, but I can’t condone that.

[puzzlegal]

Go to ethnic fast food restaurants and hope that a decent fraction of the client’s are of that ethnicity. Ask people waiting in line if they are willing to answer a brief survey. Accept that you will oversample anglos, and just throw out a random sample of excess data.