This document contains two sets of helpful suggestions for AAI interviewers.

The first (by Mary Main) covers general is-

sues of interviewing. It is followed by a set of more specific suggestions (by Judy Corwell & Dominique Treboux) about
probing with follow-up questions, interrupting, and some useful Do’s and Don’ts for interviewers.

This material is not a substitute for training in AAI administration procedure. However, it can help consumers of AAI
based research appreciate the level of interview information and detail underlying AAI scores. It can also help consumers
of AAI based research make important decisions about the adequacy of procedures in various reports they may encounter.

EW

Avoiding Common Errors In Conducting

The Adult Attachment Interview
Mary Main, Ph.D.

The Adult Attachment Interview requires a special kind of interviewing skill. As noted earlier, it is nether a
fully clinical interview nor a questionnaire, and interviewers accustomed to clinical or questionnaire format
may be in special need of extensive practice prior to conducting their first "real" interviews. The interviewer
should be prepared to work with speakers who are irrelevant to the topic, those who speak at excessive length,
and those who are unusually succinct. Here, we review these as well as other potential difficulties and biases.

Difficulties Arising From A Clipped, "Questionnaire" Format

The informationally oriented or "questionnaire/survey" interview format leads interviewers to utilize a

"clipped" manner of speaking in order to encourage participants to respond briefly (often by answering "yes" or
"no"), and then move on to new topics. Administration of the Adult Attachment Interview, in contrast, requires
a continuing awareness that the analysis of the interview will depend upon a study of the participant's natural
phrasing, including his or her hesitations. For this reason, participants must not be interrupted as they attempt to
formulate their thoughts. They must be given time to reflect, and the interviewer must not hasten to new ques-
tions. Interviewers who are genuinely interested in the story the participant is telling will find it natural to give
the participant time to answer thoughtfully, and to expand gradually upon a given topic over the remainder of
the interview.

Two particular speech tendencies are associated with the "clipped" or "questionnaire" approach to the inter-
view. These are (i) the use of "cut-off' speech acts, and (ii) a failure to affirm continuing interest as the partici-
pant is speaking.

"Cut-off" speech acts. Responding to the Adult Attachment Interview is difficult for many participants, and
many of the questions require considerable reflection. Therefore, the participant will often need time to pause,
to stumble, to correct and to think out loud. Not only should the interviewer encourage continuation of the par-
ticipant's speech through active affirmations of interest (below), but she or he should also listen to the
audio-tapes to make certain they do not contain what we have termed "cut-off' speech acts or markers (George
et al, 1985). Thus, some interviewers unconsciously note the end of a participant's first response to a question
with remarks such as "mmHMM!", "great!", "uhHUH", "ohKAY". Interjections such as these can indicate to
the participant, "Right. We've covered that area, now let's move on to the next", terminating the natural flow
and development of the participant's ideas. Even when the content of the speech act appears to be positive
("great!"), the participant will understandably feel that what she thinks, remembers or feels is not really of in-
terest to this interviewer:

Now I wonder if you could tell me, to which parent did you feel the closest, and why you felt
closest to that parent?

Well, I would have thought, when I was a child I might have thought I was closer to my mother,
 because I saw more of her. . (2 sec pause]. Um/

/UhHUH, great! Why isn't there this feeling with your father?

Note that this participant has not only not been given a full conversational turn, but may have been considering
contrasting the way she had seen her relationship to each parent during childhood with the way she would see it
presently. By interrupting her the interviewer has not only indicated disinterest, but--insofar as the participant may
have been beginning to engage in a change of mind regarding the parent she felt closest to--also may have taken
away the opportunity for the interview analyst to later credit her with metacognitive monitoring.

Failure to affirm continuing interest as the participant is speaking. At times a respectful silence on the part of
the interviewer is warranted, i.e., as the participant actively searches for thoughts or for words, or is clearly en-
gaged in personal feeling or reminiscence. For many conversations, however, soft, semi-verbal acknowledgments
of the partner's speech are a component of the listening attitude. These interjections ("oh", "not once?", "mmmm",
"you went to her?"), may not even be conscious on the part of many interviewers, but their absence can make the
continuation of speech difficult for the participant. On the audio-tape, these interjections on the part of the inter-
viewer may be only barely audible.

Finally, although seen superficially as affirmations of interest and/or concern, "canned" phrases such as "thank you
for sharing that with me" should be avoided. In general, such phrases affirm neither the interviewer's interest or the
participant's experience.

Difficulties Arising From an Overly Informal Style

As noted earlier, in clinical and cognitive terms the objective of the Adult Attachment Interview is to "surprise the
unconscious". Few research participants will have had occasion to be addressed with the set questions and probes
which form the basis of this interview, and the words, hesitations, exact phrasings, and speech errors which occur
in response to the interview questions form the basis of our understanding of a given speaker's "overall state of
mind with respect to attachment" (Main, 1995).

The participant's mental state is obscured to later analysis, however, when the interviewer is overly informal, vio-
lating guidelines by extensively joining in on or initiating conversation regarding matters other than the set inter-
view topics. When marked, even a single such move on the part of the interviewer inevitably reduces the coder's
ability to apply the potentially powerful linguistic analysis of the interviewee's first (and perhaps somewhat
"surprised") response to the interview query. Because common conversation follows rules actively developed by
both partners, the participant's own or individually guided speech errors and tendencies become obscured.

Note that in earlier editions of these guidelines (George et al, 1984, 1985) we advised that . an interviewer may
occasionally remark on their own experiences as an aid to assisting the participant to recognize the kind of memo-
ries or experiences which were being sought. This was not intended to encourage a sharing of experiences between
interviewer and participant, and the remarks made in a few interviews in the initial Bay Area study were brief and
to the point (as, "Well, it could be an experience that you wouldn't necessarily consider rejecting looking back on it
now. For example, I can remember feeling rejected when my parents didn't get me a pony"). Remarks of this kind
are in fact not necessary, and.are no longer used. Needless to say, remarks that go beyond the ultimately fairly im-
personal example given above should never have been utilized, since an interviewer must not begin a semi-inti-
mate sharing of experiences with the participant.

Extreme Violations Of Protocol

A few interviewers are informal to an extreme. Some attempt pursuing topics of their own, inserting queries re-
garding e.g., sibling relationships or sibling rivalry'. Others attempt to expand upon topics mentioned by the par-
ticipant as, "You said earlier your sister was depressed. Did you experience your sister's depression as traumatic in
any way?", or attempt to combine these questions with questions which are standard to the interview, as "What ef-
fect do you think your sister's depression had on the development of your adult personality?", or "We've been dis-
cussing loss through death, but now I'd like to ask you whether you experienced your sister's moving to the East

2
Coast as a loss. Also, was her depression some kind of loss for you?" A few interviewers have attempted to step
outside the purposes of the interview in a semi-therapeutic manner, as, "I wonder if you've ever thought there
might be some kind of connection between the way you say you worry about your child and those early losses you
experienced", or, "I wonder if you ever experienced any anger with your mother for letting your father behave like
that".

These extreme protocol violations are very serious. Any interviewer who adds queries or comments of this kind
beyond a single initial practice interview should not be retained by the researcher.

Interviewing Subjects Who Present Special Difficulties

Interviewing participants speaking on irrelevant topics and/or at excessive length. The speaker's initiation of irrele-
vant diversions from the topic set by the interviewer are highly informative, as are tendencies to speak beyond
one's conversational turn. Knowing that she or he must artfully permit the participant to fully reveal his or her
speech tendencies, the skillful interviewer allows the speaker diversions from topic, as well as lengthy turns. In the
following example, the interviewer--patiently not interrupting for several minutes, then once politely giving the
speaker opportunity to return to the original query--acts correctly:

And ‘difficult’. You used the word `difficult' in describing your relationship with your mother in
childhood. Any incidents or memories that would tell me why you chose that word to describe
your relationship in childhood?

My mother was a very difficult person. And I've made an effort to be, um, very nice to my mother I
don't mean nice nice like uh, well I've been nice nice to her. 1 the most important decision that I
made was that she was not my mother and, um, I could see her as a separate person (Mm-hmm)
and um, very much a character [laugh] (Mmhmm) um, I was bullied by her when 1 was, up to
when I was 12, but I'm not bullied by her anymore. (Mm-hmm mm-hmm) And, those are things
that happened over years y'know like uh, after I got away from her and got married. . . (2 sec)
(Mm-hmm) and um, I just 1, the few times that we've seen her, um, each time it's gotten much
easier and much easier, I told her I guess about 2 weeks ago 3 weeks ago, excuse me, she was
angry that, I had her to dinner and she was angry that I served creamed onions, because she is
sometimes allergic to onions, like, she gets these big rashes and she makes a big deal about it,
and I said, its up to you to remember your allergies, it isn't up to me, and . . . (3 sec) and she's a
very resentful person and these are the--little, little incidents that grow in her mind and they be-
come more and more important and she calls up about them, and she screams about them, until
she has, uh, well, I think she has some kind of, she's kind of borderline, a borderline disorder...
(3 sec)

And would you say that she was difficult in that same way in your childhood, too?

Definitely, and what I said to her was, I said, Lookit, y'know nobody tells me what to serve to my
family. It's a family matter, and you're over here as my guest, and I want you here, I really do,
but you're not going to tell me, and I said, I think when you don't hear from me, if I don't call
you, for whatever reasons, you think you find something that you can churn up I said that's just
I'm busy I have, not only myself but other people to ca-take care of, all the time. And um, that's
what's happening. (Mm-hmm) And when I when I s -- when I get off the phone I always feel like
that's terrific y'know it's one more thing I mean this person, whether she's my mother or what, all
that stuff, there's always that conflict that goes on and 1 I feel very clear about it and 1 know that
it'll happen again, the borderline stuff will come out again, her stuff will come out, but for me,
I'll just it'll happen it'll come it'll get resolved faster, (Mm-hmm) and won't bother (typist can't
hear] as much. I think that I'm I'm it's I wonder whether my sister, she's just like my mother, I
wondered if my sister would have treated me the same as my mother does, that is, if my sister
was alive. I don't think so, I mean I think my sister really loved me, but she died, many many
ye--many years ago, and she, with that she, I think 1 idealized her a little bit, I mean, if she lived,
she might have been more like my mother is now, and treated me more like my mother does now,

3
 but maybe not. And 1 don't care for my mother the way I cared for my sister, because she didn't
take care of me the way my sister took care of me, cause she was never there. So my mother
wants me to love her now, but she hasn't earned that, she's not going to get it. (Mm-hmm) So I try
with her, but I have to try, and I just can't feel about her the way she wants me to, (Mm-hmm) I
mean the the conflicts that arise are that my husband really likes my mother, (Mm-hmm) So I get
to witness the fact how he doesn't see what I see or know what I know, which is very difficult and
painful. (Mm-hmm). But, it's, it's typical the way she behaves in front of him, she's different when
he's around, but I'm different from her, I've never left my, never neglected my child, my son .... (2
sec) under any circumstances, . . (2 see) which is what had happened to me as a child. . . (2 sec)
So I was lucky to be able to be different from my mother.

Mmmm. Well now, the second word you used to describe your childhood relationship with your
mother was `strong'. Can you think of a memory or an incident that would illustrate why you
chose strong' to describe the relationship?

What the transcript coder will want to know is whether the speaker has violated the maxims of quantity and rele-
vance, in the face of an interview which proceeds at a reasonable pace with a series of set questions. The analyst
will be better assured that the speaker is in violation of cooperative discourse if the interviewer does seek one or
two reasonable junctures to attempt to guide the speaker back to the immediate topic (as does the interviewer
above), or to move forward to the next question (or to the interviewer's own conversational turn). If the interviewer
fails to at least once politely remind the speaker of the original topic, the speaker will gradually become less mark-
edly in violation of cooperative discourse, since the interviewer will have inadvertently "licensed" this change of
direction. If, on the other hand, the interviewer interrupts too frequently or insistently, the opportunity for correctly
scoring and classifying the interview may be lost. As noted earlier, the interviewer quoted in the passage above has
behaved appropriately, and should not have spoken more than she did. It should be noted that she waited to
"remind" the participant of the original topic until there was a reasonable break in the participant's speech (a
3-second pause), and she then waited past another long speech passage and a few brief pauses for a "concluding"
sentence ("So, I was lucky .to be able to be different from my mother") before moving on to the next topic.

Interviewing Speakers Who Introduce Extraneous Topics In A

Manner Inviting Interviewer Participation

It is easy for speakers with tendencies to introduce irrelevant topics to lead new interviewers to violate protocol by
briefly following the speaker into irrelevant areas. This may be constitute a sympathetic attempt to alleviate the
speaker's evident anxiety and desire to avoid the topic. These well-intended errors may be comical:

....and so, my mother was... a sweet, mild, uh, sort of a relationship, you know. That very light fix-
ture there, it, I think I've seen one like that before, somewhere...

I have too. You used sweet, and mild, and...?

I wonder where it was, maybe when I was a little kid, they had them. And the color, it looks familiar
too.

Yes, well, we have sweet and mild. Its one of the very best colors for light fixtures, I think. Any-
way, they've been that color ever since I came here.

Errors of this kind are easy to make, but the interviewer should have found a way to acknowledge the participant
without joining in on the new topic of conversation. . This would have permitted the participant to continue with
the irrelevant topic, without inadvertently "licensing" the change. A better interviewer response would have been
as follows:

....and so, my mother was... a sweet, mild uh, sort of a relationship, you know. That very light fixture
there, it, I think I've seen one like that before, somewhere...

4
 Oh (spoken in a manner indicating mild acknowledgment). You used sweet, and sold, and..?

I wonder where it was, maybe when I was a little kid, they had them. And the color, it looks familiar
too.

Yes, it could be. . ( said mildly again with brief pause to permit speaker to continue on irrelevant
topic, but without joining in on it. If speaker does not continue on irrelevant topic, the inter-
viewer continues with) . . Now we have the first two adjectives or words you've chosen, l think,
sweet, and mild.

This interviewer has succeeded in gently reminding the participant of the topic, while permitting the speaker to
continue with his or her violation of relevance.

Interviewing Speakers Who Are Exceptionally Succinct

The interviewer not infrequently is faced with a speaker who does not actually refuse the interview queries (e.g.,
by repeating "I don't remember"), but rather responds with answers which are exceptionally succinct. In this case,
the interviewer still politely seeks to obtain the full information sought in the interview guidelines, and also to clar-
ify points which may be obscure. At the same time, as always, the interviewer waits through long pauses for the
speaker to complete his or her thought. The following interviewer is doing well with a speaker of this kind.

Did you ever feel rejected as a young child? Even if now you height not think you really were be-
ing rejected--well you might, but the point is 1 wonder whether as a child you ever felt re-
jected?

Yeah.

How old were you?

Ahm, ahm, twelve, twelve years old.

And, what happened that made you feel rejected?

Ahm, I wrote a couple of letters back, a couple of letters to uh, to my parents that I didn't want to
stay there at the school, and uh

You mean you wrote to them from the school?

From the school at Place 7 (yeah), and I can't even remember uh, I got like one letter back for, for,
for Christmas vacation and that was, and I can't, I'm amazed, and 1 asked them, I said, "How
come I have to stay here?". They never gave me a response.

So they didn't answer your letters?

Right.

How did you feel about that? What did you do?

Ahm, un--I just, 1 didn't feel wanted, or loved.

You, you felt unwanted and unloved.

Yeah.

Why do you think your parents did that--do you think they realized how it was making you feel? .

5
 . . . . (5 sec) Probably my mom, yes. Probably your mom did? Yeah.

But your dad maybe didn't?.

No.

Okay. Can you think of any other times when you felt rejected as a child?

. . . (17 sec) That's, that's about it.

That's the main one?

Yeah.

Other Common Errors and Difficulties

The errors discussed above consist in general biases frequently found in new . interviewers, and problems pre-
sented by speakers of particular kinds. In addition, there are other relatively common errors and difficulties.

"Translating" the participant's responses and making reassuring interpretations. Clinicians and others accustomed
to "mirroring" or "reflective listening" may tend occasionally to attempt to show understanding of a speaker's re-
sponses by repeating them back in altered form, or by completing sentences or ideas which seem to have been left
incomplete. Responses of this kind, however well meant, are ruinous to the Adult Attachment Interview. Occa-
sional repetition of a speaker's exact words can be acceptable as indicating a natural affirmation of interest.

So, that time 1 went to her . . . . (4 sec pause].

That time you went to her.

And, then it didn't work out the way I hoped.

However, any alteration of a speaker's word choice (as, alteration or reinterpretation of suggested adjectives), and
any attempted completion of the speaker's unfinished responses will leave the interview analyst unable to deter-
mine the speaker's mental state at a given moment:

When he would scream at me like that I felt so, so. . [2 sec pause] well, 1 just felt like..

You felt so angry.

Uh,1, guess, yeah, sure, angry.

While the speaker is having some difficulty in finding the appropriate word, the interviewer has committed a seri-
ous error. The speaker might have eventually said that he found himself confused, upset, or frightened, or simply
(equally informative to the eventual interview analysis) may have found himself unable to complete the sentence.
This serious error may have even further repercussions, in that the rest of the interview may be somewhat altered.
In further discussions of his feelings regarding the father over the course of this interview, the speaker now has the
task of defending, or expanding upon, an "anger" towards the father which he may not have felt, and/or remem-
bered, and/or intended to disclose.

Another error common to new interviewers consists in attempting to verbally support the participant's views of
their experiences or of their parents. The interviewer may indicate empathy and interest non-verbally; or by the soft
under-riding affirmations of continuing interest (`mmm' .... 'oh') discussed above. He or she must not, however,
verbally endorse statements of attitude made by the speaker ("he sounds like an awful father/ they sound like

6
lovely parents").

The above represent a set of injunctions to interviewers neither to change, nor especially endorse, the speaker's dis-
course or apparent attitudes. This does not mean that the interviewer must remain silent when not formally query-
ing or probing, as the example taken from an interview with an overly succinct speaker (pp. 34-35) has shown.

Failure To Respond Appropriately When a Subject Inadvertently

Disrupts The Ordinary Order Of Interview Content

Sometimes a participant begins to answer a question which is to appear later in the interview. The interviewer who
is familiar with the protocol will be able to allow the research participant to complete their thought, while simulta-
neously indicating that the topic will arise again at a later point. When the speaker has completed her thought, then,
the interviewer can simply add that she or he will want to be asking more about the topic later, and the participant
should not be encouraged to continue to digress beyond this point. The question pertaining to this topic should then
be raised when it appears in its prescribed order, with an acknowledgment that it has been discussed previously.
This acknowledgment should however be phrased in an open-ended and non-leading manner. As an example, the
interviewer might introduce the later question as, "You touched on this earlier, but..." or "You may have touched
on this earlier, but...".

Note of course that a participant who has just explained that she felt rejected throughout her childhood
should not be asked, "Did you ever feel rejected as a young child?" Rather, the interviewer begins, "I know we've
already discussed this, but nevertheless could I ask you again, did you ever feel rejected as a young child?". The
interviewer in this way acknowledges that this is a question which she or he "has" to ask because it is part of the
interview protocol, and goes on to ask it as it appears within these guidelines (in this instance, for example, the in-
terviewer might go on to say "Of course, looking back on it now, you may realize it was not really rejection, but
what I'm trying to ask about here is whether you remember ever having felt rejected in childhood").

In certain rare cases--normally, those involving experiences of trauma--it may be necessary to allow the research
participant to discuss a particular topic to some extent before telling them that it will be brought up again later. It is
critical under these circumstances that the participant does not feel in any way diminished after having raised an
topic which both creates feelings of vulnerability and is highly personal.

Failure to probe and/or to "follow-up" on participant responses. This is a separate issue from the
need for continuing, under-riding acknowledgment that the participant "has the floor" and that what she is saying is
continuing to be of interest (item 2, above). Often a research participant will make a statement which needs clarify-
ing for the purposes of the interview, for example by responding briefly "I withdrew" to "What did you do when
you felt rejected?". Clearly, the interviewer needs to ask what this research participant actually did, since "I with-
drew" might have any of several meanings.

In general, the skillful interviewer uses not only the set probes, but also his or her own spontaneous probes and fol-
low-up questions as a technique in his or her effort to maintain a natural, semi-conversational. These probes and
follow-up questions indicate that the participant's experiences and feelings are of genuine interest.

Tendency To Probe Negative More Than Positive Events, Or Vice Versa

On reviewing their protocols, interviewers are not infrequently surprised to note a bias towards probing either
negative or positive adjectival descriptions for the parents, and/or experiences. The most obvious example is a ten-
dency to avoid the follow-up probes to the abuse questions, and/or, somewhat more subtly, to avoid the natural in-
ferences (and queries) which would follow when a subject (perhaps in some other section of the interview) de-
scribes an incident clearly bordering on abuse. Other interviewers tend to become most interested, and probe most
deeply whenever negative events involving the parents are described or implied.

These biases are not confined to abuse incidents, since quite a few interviewers begin with a tendency towards pro-
viding more follow-up probes for negative than for positive adjectival descriptors for the parents. This is one of the
several difficulties which interviewers and researchers should be alert to in reading through the prospective inter-
viewer's early protocols.

7
 Reminders About Interviewing
Judith Crowell, M.D. and Dominique Treboux, Ph.D.

With all memories - find out what the parents did and how the parent responded.

If relevant information is presented earlier in the interview and not mentioned when a question about the topic is
asked, previous comments can be used as probes.

Any information given about a caretaker should be probed.

Adjectives: Get adjectives which describe the relationship. It the adjective does not seem to describe the relation-
ship (my mother was beautiful), ask if the word is "related to" their relationship.

Ask for a specific memory or incident.' If the subject gives you a vague response or a scripted response,
probe once (Can you think of a specific memory or a specific incident). If they still can't come up with
one, go on to the next adjective.

Initial upset question:

Make sure you understand what the subject did. If they say they cried or had a
tantrum, probe to find out more specifically they did. Did they go to their room
and cry, what is a "tantrum", etc.

Other adults: Find out who they were and why/how they were influential.

Loss: Probe all losses mentioned.

Changes in relationship and current relationship: Get information about both parents.

ON PROBING
How To Probe

Basically, all the questions you need are on your AAI prompt sheet.

Bear in mind probes are meant for clarification purposes;

We do not need specific details (i.e., the type of house they lived in, if they had uncles of aunts and how often they
saw them, how they were spanked etc.,)

The best probes are when the interviewer says little, gives quick, short prompts where the function is to get the par-
ticipant going again. If you sound like you are saying a mouth full or a full paragraph you are off target.

Good: What happened? What did your mother do? Your father do see PS25 p.7

Probing vs. Leading:

LEADING: statements/prompts/questions which require the participant to agree with you or require a yes/no re-
sponse or give interviewers' interpretation of experience

S: I think I wasted a lot of my years Leading: Do you think that was because there was a lack of communication
between you and your parents? Non Leading: Why do you think you wasted a lot of years? or why do you think
that is.

8
When To Probe:

1. Need more information

Example l: one subject mentions that her father was away often: Probe: need to find out where he was, why
was he away and how often he was away, and for how long.

Example 2: one participant recalls an incident concerning her father getting mad because a neighbor never
picked up the kids, the reason the neighbor did not pick them up it was because he was drinking, the
daughter never told the father that the neighbor would drink and none of the kids wanted him to pick
them up. Probe: Why did she not tell her father?

Example 3: one participant describes a separation incident when she hurts herself with her grandmother,
there are no follow-up questions Probe: what did grandmother do, what did parents do when they came
back?

2. When subject makes a declarative statement or uses an adjective to describe themselves

Example 1. I am totally different than my parents, I am totally different with my children than my parents
were. Probe: How are you different from your parents?

Example 2. I am overwhelming with my children. Probe: What do you mean by overwhelming?

Example 3. I am violent with my children. Probe: What do you mean by violent?

The following questions are prone to leading probes:

Effects

Parents' behavior/

Questions which do not require probing

The relationship question: Just ask the question as is. There are no wrong answers. Do not probe, no not comment
on ability to recall.

ON INTERRUPTING

Do not interrupt the subject: let them finish the sentence or thought

Example 1 in the relationship question if the person is talking about when she is a teenager let her finish
and then ask about an earlier time. If you interrupt you are risking being both rude and losing valu-
able information. Of course, if the participant is going on and on about something that seems totally
irrelevant, do interrupt and bring her back on track

If the participant brings up a topic or something unusual, let them finish their thought. Do not interrupt but do not
go on with the topic if it seems totally irrelevant.

Example 1. One participant wanted to talk about the death of her dog under the loss question

Example 2. One participant talked about her mother's suicide attempt.

9
 EMPATHIC STATEMENTS

if the subject says that an experience was particularly difficult or moving to them, do not rush on to the next ques-
tion but acknowledge what the participant said (i.e., that sounds difficult, that must have been hard). However,
make such statements only if the subject has verbally said that the experience was painful, difficult, hard etc. If the
participant does not comment on the experience in terms of her feelings, neither should the interviewer.

NOT TO DO'S

Never interject unnecessary comments. (i.e., that sounds exciting) Do not assume you know what the subject is
feeling (that sounds horrible) Do not sum up (i.e. regarding illness So it was a time when they took really good
care of you.) Don't make judgments (i.e., so it was a nice time, so she was nurturing) when the subject has not used
the word; Don't sum up. Do not think you can help the subject remember by anchoring in time (i.e., how about
nursery school, how about when you had your first play) Do not put words in their mouth. (i.e., do you mean...)
MISC

A. Always ask the question as on AAI, if the subject does not come up with anything then you can use some infor-
mation that the participant gave you earlier in the interview.

Example 1. The participant has told you that the she had been sick as a child at the beginning of the inter-
view. Do not start the illness question with this particular incident, but if she doesn't bring up the sub-
ject herself, then ask "what about..."

B. Always pause after each question. Give them time to collect themselves. If the subject does not answer the ques-
tion either derails by laughing and dismissing question (i.e., says that is a difficult question and laughs) or goes off
onto another topic, be sure to ask the question again.

C. As much as possible get an incident for each adjective, but do not make them feel bad if they can' t (i.e., that's
okay maybe you will be able to do it later, this becomes easier with time) or comment on their ability to recall.
D. If participant asks you if what she says makes sense say yes if true, otherwise "I' m not really sure or I don't
think I quite understood". Do not repeat what the subject said or interpret for her. bb not ignore the question either.

E. Re: Changes in relationship and current relationship. Be sure to ask both questions and for both parents. If par-
ent is deceased ask about relationship up to the death.

F. Re: Other adults, be sure to ask how adult was important/special/influential.

10