LINGUIST List 2.768

Sun 10 Nov 1991

Disc: Human Subject Research, I

Editor for this issue: <>

Directory

1. Vicki Fromkin, Re: 2.763 Query: Human Research
2. , Human Subjects Research
3. , re: Message 2,763
4. "Barbara.Abbott", 2.763 Query: Human Research
5. Michel Jackson, Human Subjects Research (LONG)

Message 1: Re: 2.763 Query: Human Research

At ucla all research involving human subjects -- whether tape recordings,
experiments, grammaticality judgments, etc etc etc -- is supposed
to be done only after approval of the Human Subjects Protection
Committee. We submit forms and protocols and statements to
be read to and/or by subjects and under certain circumstances
signed by subjects. Usually this only pertains to research
that comes under a grant proposal, which proposal must go through
the HSRC before being submitted to granting agency. But, other
research also comes under these regulations. For example, students
in the huge Psych 10 (intro) courses serve as subjects for many
psych experiments. If you wish to recruit subjects from this pool
you must submit approval forms to HSRC and to Psych Dept Human
Subjects Committee.
I agree that this often seems like and probably is silly re certain
studies but in the long run I personally feel it is better to err
re time and effort wasted than in doing anything which could possibly
be unethical. It is probably the case that the procedures have been
overly complex recently due to the various 'science fraud' cases and
those involving subjects being uninformed or misinformed. Also probably
true that institutions are driven less by ethical aims than by need to
protect themselves against litigation. But I would rather see this
extreme than the lack of concern symbolic of the past.

This will be one of many similar responses: all human subjects
research these days must go through a human subjects committee
approval procedure, no matter how benign or banal. I thought
everyone knew that, but evidently not. At least you don't need
approval for political correctness, yet.

As a psycholinguist, I have also had some problems with Human
Subjects Committees. It seems to me that the problem is this:
Certain sorts of research do not need clearance. These are data
collection in which (1) the subjects are informed that they are
participating in an experiment or that the material collected may be
used as part of a research publication, (2) no personal data is
collected, (3) no records are kept of the participants' identities,
and (4) the procedure is neither emotionally or physically stressful
or invasive.Under these circumstances, no privacy is violated and
no clearance is necessary. The problem is that some researchers do
collect data not within these bounds, and universities are extremely
wary of being sued.
A second problem is the matter of an informed consent procedure. The
university at which I used to reside insisted that I had to use a
standard form which stated that the iundersigned that the undersigned
had been informed of any risks that might result from the
experimental situation. I was asking prople to read sentences on a
computer screen, for goodness sake! There weremn't any dangers that
I could inform them of. Thus if some nut case decided that I had
damaged them, they could also simply decide that I had failed to warn
them of the dangers. Unless there is some actual risk, why sign such
a statement? Presumably here, the university was again afraid of
liability, but the procedure does not appear to me to actually avoid
problems.
I don't actually mind the presence of Human Subjects Committees, but
I do object to having blanket policies that tie up non-invasive, non-
privacy threatening research. Especially since the waiting period
for clearance stretches in accordance with the number of researchers
required to submit applications.
Laurie A. Stowe

AT Michigan State there is a committee that has to approve research
involving human subjects. MA thesis and PhD dissertation research that
involves human subjects does have to be approved at least -- I'm not
sure about more informal class projects (I don't know how they'd
check on that).

> I would like to open a new thread for discussion. Faculty
> and students here in Linguistics at SIU are being hassled
> (at least that's our perception of it) by the Human
> Subjects Committee. To what extent have people on this
> net been asked to clear their research through such a committee?
> What I have in mind is ordinary linguistic and ESL research--
> [specific examples and much else deleted]
I suspect that what is happening is that the same standards that are
applied to other fields have been extended to you. In my experience,
toeing the line in phonetics would mean:
- getting informed consent for every recording (i.e. a signed form
telling the subject what the purpose of the recording is, telling them
that no harm can come from it, and guaranteeing their anonymity,
etc.), palatograph, laryngographic recording, etc.
- getting approval for any procedure performed _for research
purposes_.
This is a tricky area. "Normal classroom activities" are perhaps
exempt from human subjects regulation EXCEPT when it is for research.
The reason, as i understand it, is that classroom activities carry a
benefit for those involved (i.e., it contributes to student
education). Research activities do not necessarily carry a benefit to
anyone but the researcher. An example from an allied field: in Speech
& Hearing here at Ohio State, "real-ear" acoustic measurements with a
probe microphone inserted in the ear canal (only slightly invasive -
less invasive than having a pediatrician shove an otoscope into
your ear) are routine clinical procedures. Done nearly daily,
certainly dozens of times monthly, no risk to clients with or without
hearing losses, etc. HOWEVER, a "real-ear" acoustic measurement as
part of a dissertation, research project, etc., REQUIRES SEPARATE
APPROVAL FROM THE BIOMEDICAL HUMAN SUBJECTS RESEARCH COMMITTEE. Which
was indeed forthcoming reasonably rapidly.
The same principle applies to questionnaires, interviews, and just
about anything else i can think of.
Is this policy reasonable? Personally, i feel rather ambivalent. To
the extent that "we linguists" claim to be scientists, it seems
obvious that we should adhere to the same ethical standards as other
scientists. (Granted that there have been ethically awful miscarriages
in the past: consider, say, lobotomies. The point of HSR regulations
is to attempt to prevent recurrences of the same). To the extent that
our research poses no risk to the subjects, we should be able to
demonstrate that fact, and approval should be routine (as in my
experience, it is). But the questions still have to be asked.
On the other hand, it does seem silly to have to get an informed
consent just to have someone sit in a sound booth to make a recording.
Unless that person were one of the last few speakers of a language, &
wound up feeling as though i or someone else had made a reputation and
accrued fame and fortune as a result of exploiting their knowledge of
the disappearing language ... such a situation is to my knowledge
completely hypothetical, but it is an example of the kind of situation
that "informed consent" requirements are meant to prevent.
Practically, there are several options available to you. Some
institutions have 'blanket approvals' for procedures which are routine
within a department or other administrative unit. Our department's
'blanket' includes descriptions of standard psychoacoustic and speech
pereption paradigms (play recorded or synthesized beeps or syllable to
listeners who make same/different or identification judgements);
electrophysioogical research (surface electrodes (nothing breaks the
skin) on the neck (auditory brainstem), scalp (motor planning), or
lips (motor activity); "real-ear" measurements; and a few other
things. Such blanket approvals are of varying validity (e.g.,
grant-funded research has to be approved seperately here & does not
fall under the blanket. But apparently departmental activities such
as student research (MA thesis, qualifying papers, etc.) with no
external funding does not require separate approval).
If you don't have external funding, don't care, and don't ever expect
to get any, i suppose you could go ahead and do what you want anyway.
i don't recommend this AT ALL, but i do know some people who have
coasted along this way happily for years.
The least palatable option is to go ahead & get piecemeal approval for
every project that comes up. You might ask what people in Psych or
Sociology or perhaps Anthro do; they are facing the same problems.