Re: Annual Survey of American Jewish Opinion (was: Re: Ruling class Jews

["David Quarter" <davidquarter@xxxxxxxxxxxx>]

From:                   loupaulsen@xxxxxxxxxxx
To:                     marxism@xxxxxxxxxxxxxxx
Subject:                Annual Survey of American Jewish Opinion
(was: Re: Ruling class Jews
Date sent:              Mon, 27 Oct 2003 20:58:11 +0000
Send reply to:          marxism@xxxxxxxxxxxxxxx

> David Quarter rejoined:
>
> >
> >   Ooh...I see. Cause a group might prefer a particular research
> > finding or another, that somehow will translate into data being
> > manipulated to get this result?
>
> Lou wrote:


 You don't have to manipulate the data if you control the data
collection procedure - see below!


  David's replies (on hopefully for the last time): There are ways, for
example, of asking questions to bring about a desired effect. I don't
think it's a straight forward as you claim...


 It's like saying that since a Marxist would prefer that the working
> > class see the ruling class as their main enemy, then somehow
any
> > Marxist survey conducted on attitudes towards the ruling class
> > amongst workers is not to be taken seriously.
>
> > Second, when an 'independent' academic study is conducted
to
> > survey Jewish attitudes towards  Israel are their not similar
stakes,
> > institutional pressures?
>
 Lou wrote:

 But there are also academic-cultural pressures to fully disclose
sources of bias in your sample selection and so on.

 David': Yes. But that does not mean  it always works that way.
And there are also ways of hiding biases.



>
> > Are you for a second suggesting that academics are not
burdened
> > by the worry of finding the "right result", have no pressure put
on
> > them from persons or groups funding their research to find a
> > particular finding or that academics have no pre-existing views
> > which may influence how they conduct their research?
>
Lou:

 I've been in academe.  There are such pressures.  But I've also
worked for an association in the private sector with an advocacy
component.  The pressures are many times as intense.>>>>

David:  No disagreement.

>
> > You will of course argue that "the poll is bias" b/c of who
> > conducted it.
>
> Polls are not biased because of who conducts them.>>>

 David:

This view would probably apply to reserach in general and would
also  depend on who you ask. For example, many academics hold
the view (or at least clam to hold the view) that a research finding
will speak for itself, whereas other feel that anyone "tainted" by a
particular association to some group or theoretical framework are
not be trusted.

To make an illustration closer to home: although I am a long ways
from calling myself a academic, I had previously cited a link to a
"nazis" website  to Marx-mail to dispute a point brought up by Lou
P. on German-Jewish capitalist concentration in a particular
industry in pre-war Germany. Despite the fact that the information
was itself written by a Jewish (non nazis) academic (and then cited
by the Nazis webmaster), rightly or wrongly I was accused by
some people of holding nazis sympathies. Although this wasn't a
poll and although i obviously hold no such sympathies ,  seems to
me that this a perfect e.g of what you claim_never_ happens in
academia -- i.e., flawed by association.

No?




 Polls are biased because of flaws in their methodology.>>>>

 And for some people the flaw is a matter of how the question(s) is
framed, which is itself sometimes a matter of debate.


  Flaws in their methodology can be deliberately, or even
unconsciously, selected for by who conducts them.>>>

 Yes.

>
> Let's take the Annual Survey of Jewish Public Opinion as an
example.
>
>
> The respondents to the 2000 survey had a median age of 55, and
37% of them were over 60.  This is similar to that Greenberg thing
from mid-2002, also from AJC, and it may be that they got the
sample from the same place.
>
> What is the source of the respondents to the Annual Survey of
American Jewish Political Opinion?  According to the AJC, the
2002 responses came from "1008 self-identified Jewish
respondents selected from the Market Facts consumer mail
panel."  Market Facts, Inc., which has been bought up by
Synovate, had a "Consumer Mail Panel" composed of 600,000
people who would answer mailed questionnaires for market
research, mostly about products and stuff.  Possibly they get
coupons for it - I am not sure.  Basically these are people who have
time to read their junk mail, which explains the high median age>>>

Most 55 and 60-year-old people I know don't spend their time
reading junk mail. They work. I suspect that the behaviour you
attribute to 55 and 60-years olds is indicative of that of senior
citizens or persons not working. I would suspect that more often
than not most 55 or 60-year old (white and Jewish) males and
females in today's world are professionals, especially since Jews
are generally quite liberal on issues of gender.


 and they also like to answer surveys, which often means that they
have more extreme opinions than the rest of the population. >>>>

 I'm not sure why it would  "often mean" this. For example, I
supposedly hold "extreme opinions" (at least, according to friends
and family)  yet I'm NOT in my 50s or 60s and don't answer junk
surveys. So I wonder why in your opinion people  in this age
bracket are more likey to hold "extreme opinions".   For example, if
you sampled 600 000 above-mid-age black persons about their
opinion of some issue, are they more likely to "have extreme
opinions"? And what btw is your definition of "extreme opinions"?

 (Suppose for example you did a survey of 'people who call into talk
radio programs.'  It would be biased for the same reason.)

Yes.

>
> I don't know how the people identified themselves as Jewish - it
would make a difference whether the question were about ethnicity,
religion, or "Are you Jewish? (Yes/No)" <<<<

I suspect you're wrong. I think that for the majority of Jews, being
Jewish is HUGE Part of their identity. So when asked on a survey,
I would imagine that many secular Jews (which make up a
significant percentage of the Jewish population in NA ) are likely to
self identify as Jewish.

Just take internet sites for "Jewish singles" (I've perused a few
myself): you'll notice that the options given in choosing partners
are: ethnic background (e.g. Askenazi/white, Sephadic, Mixed,
other, etc) and religion (secular, reform, conservative, orthodox,
etc). A fare share of the people who post messages to the list also
self identify as secular.


 My guess is that it was a multiple-choice question on religion.  I
would assume the religion question was optional, in that you didn't
get thrown out of the panel if you left it blank. So these are people
who are not only religious Jews but they felt like telling it to Market
Facts rather than believing it wasn't their business and leaving it
blank. Well, if you limit "Jews" to visible temple-going Jews and
further skew it towards aging people who like to answer opinion
surveys, you are biasing the opinion distribution considerably.
 In addition, these people are on a Panel, which means that they
are repeatedly surveyed by Market Facts.  Market Facts says that
its respondents "have an established relationship with us."  This is
an annual survey.  The AJC have been doing it for years.>>>>

You are again assuming things that aren't at all evident IMO.



  I don't know how many people on the panel self-identified as
Jewish, but I think it's possible that there is a lot of duplication of
respondents from one year to the next. >>>>

 You think it's possible, but you really don't know.


 This would mean that if someone objects to being surveyed by the
American Jewish Committee, they get thrown out of the Panel.
This may be another source of bias.>>>


You're going from unproven assertion in the sentence above  to a
"factual " statement in the next...


>
> Note that in all the above paragraphs I say 'this may be a source
of bias', 'maybe they did that' and so on.  This is one of the
differences between academic research and commercial market
research: the academic research tends to be more transparent,
and more information is provided about where the sample came
from.>>>>

 Ok.


  I will tell you this: I bet nobody would dare to publish a study in
the American Sociological Review purporting to describe "American
Jewish opinion" based on data from the Market Facts consumer
panel!>>>>

 I took a methods class where the Prof showed e.g.s of pretty
horribly methologically flawed research that had been accepted into
academic journals. Point: I don't think it's as uncommon as you
would make it out to be...


>
> Finally, here is the thing: the AJC has been using this same
source of opinion data for years.  If they had gotten results they
didn't like in the first year, they might have jumped to a different
research organization. >>>>>

 What results are they? i'm not sure whether it serves more of their
interests to present findings suggesting that most Jews support
Israel or support attacking Iraq since it just seems to prove the
(racist) adage  that Jews act in their common interests. Yet, this is
matter of debate...



 Instead, they have stayed with the same organization and the
same panel, possibly including the same respondents, over the
years.  This means they are happy with what they are getting.


This is also a potential source of bias.>>>

Key word here is: "Potential". Yet, above you seem to applying this
is "fact".


>
> Just some things to think about,

 Indeed.

DOQ



> Lou Paulsen
>


~~~~~~~
PLEASE clip all extraneous text before replying to a message.