73 

The Impact of a Digital Speaker on a Teacher's Interaction with 
a Child with Limited Functional Speech 

Elsa Popich and Erna Alant 

Centre for Augmentative and Alternative Communication, 
Department Qf Communication Pathology, 

University of Pretoria 

ABSTRACT 

This paper describes the impact of the implementation of a digital speaker and teacher training on the quality of a teacher's 
interaction with an AAC user. The data was obtained by recording and analysing classroom story time interactions. The types 
of utterances directed at each child in the class, the types of questions used as well as the messages recorded onto the digital 
speaker were determined for each research phase. There were four major phases in the research, namely, a pre-implementa-
tion phase, a post-implementation phase, a post-training phase and a post-withdrawal phase. The results indi.cated that the 
implementation of the digital speaker and the training of the teacher resulted in certain changes in the way that the teacher 
interacted with the AAC user. Utterance types such as answering and imitating were used more frequently after the imple-
mentation of the device. There was also a much greater variety in the teacher's use of questions as well as in the messages that 
were recorded onto the digital speaker after the five training sessions were completed. This study highlighted the importance 
offormal training for the communication partners of the AAC user. One of the most important communication partners of the 
school-aged AAC user is the classroom teacher. 

OPSOMMING 

Hierdie studie beoog om die impak van die implementering van 'n digitale spreker en die opleiding van die onderwyser op 
die kwaliteit van die onderwyser se interaksie met die AAK-gebruiker te bepaal. Die data is verkry deur die opneem en 
analisering van klaskamer-storietyd-interaksies. Die tipes interaksies wat aan elke kind in die klas gerig is, sowel as die 
tipes vrae wat gebruik is asook die boodskappe wat opgeneem is op die digitale spreker gedurende elke navorsingsfase is 
bepaal. Daar was vier fases in die navorsing, naamlik, 'n voor-implementeringsfase, 'n na-implementeringsfase, 'n na-
opleidingsfase asook 'n na-ontrekkingsfase. Die resultate het daarop gedui dat die implementering van die digitale spreker 
en die opleiding van die onderwyser gelei het tot sekere veranderinge in die onderwyser se interaksies met die AAK-gebruiker. 
In~efaksie-tipes soos antwoorde en nabootsing het meer dikwels voorgekom na die implementering van die digitale spreker. 
Dimr was ook 'n groter verskeidenheid boodskappe op die digitale spreker na die vyf opleidingsessies voltooi is. Hierdie 
navorsing beklemtoon die belang van formele opleiding vir al die mense wat met die AAK-gebruiker kommunikeer. Een van 
die mees belangrike mense ,wat met die skoolgaande AAK-gebruiker in interaksie tree, is die onderwyser. 

, 

KEYWORDS: classroom i~teraction, device implementation, teacher training, use of questions 

Literature has indicated:that}eachers adjust to children 
with limited speech by changing their interaction patterns 
(Light, 1988; Cicognani & Zani, 1992). Teachers take more 
turns and initiate conversations more often. They use more 
direct questions, attention directing utterances and re-
quests, and less answers, imitations, praise and higher-
level questions with the children who have limited or no 
functional speech (Light, 1988; Cicognani & Zani, 1992). 
Owing to the exclusion from interaction and the teacher's 
use of certain types of interactions and questions, the child 
who has limited or no functional speech (LNFS) can adopt 
an exaggerated passivity and develop an attitude oflearned 
helplessness (Basil, 1992). 

Aprevious study was done (Popich & AI ant, 1997) which 
examined the verbal interactions which occurred between 
a teacher and two groups of children (children with LNFS 
as well as children who were speaking). Descriptive data, 
generated by analysing ten lessons, suggested that the 

teacher's interaction with the children with limited or no 
functional speech (LNFS) differed, in terms of quantity and 
quality. The teacher directed approximately 10% less in-
teraction at each of these (children with LNFS) three chil-
dren, when compared with the number of interactions that 
the teacher directed at each of the five speaking children. 
However, as can be expected, she also did not spend an 
equal amount of time interacting with each of the children 
with LNFS. Her interaction with these children was domi-
nated by questions, attention directing and requesting. 
Interaction types, such as answering and imitating did not 
occur at alI in the teacher's interaction with the children 
with LNFS. The children with LNFS therefore had differ-
ent learning experiences in the classroom to the speaking 
children. These results formed the basis for the cilrrent 
research study as the researchers wanted to determine 
whether these patterns in the teacher's interaction with 
the individuals with limited or no functional speech wou~d 

Die Suid-Afrikaanse Tydskrifvir Kommunikasieafwykings, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



74 

be influenced by the implementation of a voice output de-
vice. 

There is little doubt as to the positive effects of the im-
plementation of AAC systems on children with LNFS. Chil-
dren who use AAC in the classroom have demonstrated 
better communication competence, initiate more frequently, 
use more complex sentences and demonstrate more con-
trol in interactions (Smith, 1994). There are also a greater 
number of AAC users who now participate in education and 
recreation and the number of AAC users being employed is 
also increasing (Beukelman, 1991). 

AAC gives the user access to communication and inter-
action and increases participation and learning. A greater 
ability to interact results in improved social abilities, an 
increase in social acceptance by the community and an 
improvement in the quality of life. Increases in participa-
tion and learning result in progress at school and a higher 
level of education for the individual who has limited or no 
functional speech. 

However, in a classroom situation, where the goal is not 
only to interact or to learn language, but also to use lan-
guage to acquire knowledge and skills, one is no longer only 
concerned with how the implementation of an AAC system 
such as a voice output device (VOCA) will affect the child's 
interaction, but also with how the device will affect the 
child's primary interaction partner in the school setting, 
namely the teacher. If a VOCA is implemented in a class-
room the teacher takes on many new roles and responsi-
bilities. The teacher will need to adapt the curriculum and 
write objectives for the AAC user. Materials such as device 
overlays will have to be made and the teacher will have to 
record messages onto the device., Furthermore the tea,cher 
will be responsible for assessing progress, providing ongo-
ing skill development, identifying vocabulary and commu-
nication needs as well as many other administrative tasks 
(Locke & Mirenda, .1992). These additional roles and re" 

'. sponsibilities may affecttheteacher's attitude towardsAAC 
. implementation as well as the AAC user. 

The teacher is a significant interaction partner and will 
greatly influence the success or failure of the AAC device 
as a communication tool in the classroom ( Dalton & 
'Bedrosian, 1989). The teacher may, however, not be fully 
equipped for many ofthe new responsibilities resulting from 
AAC device implementation and training is critical in or-
der to ensure success (Locke & Mirenda, 1992; Baker, 1993). 
The majority of teachers working with children who use 
AAC devices desire further inservice training and work-
shops (Locke & Mirenda, 1992). Teachers primarily feel the 
need to increase their knowledge on AAC techniques and 
AAC devices but also to improve their skill in management 
issues such as identifying appropriate vocabulary. Research 
(Soto, 1997) has found a direct correlation between teach-
ers' perceptions oftheir own levels of skill with their atti-
tudes towards AAC implementation. Furthermore teach-
ers' perceptions of their own skills had a direct influence 
on their impression of the AAC users' abilities to improve 
their communication effectiveness. 

. The effects of training on the teachers' ability to adapt 
to the presence of an AAC user in the class, will depend 
greatly on the content ofthe inservice training. If the train-
ing concentrated on the development of skills, such as the 
understanding of technical aspects of the AAC device or on 
the necessary preparations (for example making an over-
lay) for the use of the AAC device then one would expect 
changes to occur in these areas. On the other hand iftrain-

Elsa Popich anll·Erna Alant 

ing emphasised the development of strategies, such as the 
teacher's style of interaction and use of questions, then 
changes in this area would be expected. For the benefit of 
the AAC user, however, it is necessary to pay attention to 
the development ofthe teacher'soperationiil skills and the 
development of strategies in order to ensure the'most ef-
fective implementation of AAC (Bl~ckstbne, 1986). It would 
be important to determine whether the!provision oftrain-
ing, to the teacher, would result in an increased utilisation 
of the AAC device in the classroom as well as whether it 
would result in changes in the teacher's interactions with 
the AAC user. ' 

Previous studies have found training in the use ofques-
tions to result in significant positive changes in the ,teach-
er's questioning techniques (Greenberg & Woodside, 1994). 
Questioning is one of the most important and most fre-
quently used teaching techniques. Teachers use questions 
to stimulate thought, to assess progress, to motivate chil-
dren to pay attention, to maintain control, to provide rep-
etition and emphasise key issues (Camp, 1993). The types 
of questions teachers use have a signifi~ant influence on 
the amount of initiations made by the children, the types 
of responses given, the amount ofle,arnirtg a~d the level of 
thought obtained (Cicognani . &, Zani, 1992~)"owell & 
Lawton, 1992; Greenberg &Woods}de, 19!,)4). Iilsufficient 
research has, ho,,:,ever, been done ~ri' the effects of training 
on teachers' use of questions wi'th ¥C users,'changes in 
general interaction styles and the e.ffe~t of training on t,he 
utilisation of AAC devices in,classrooms. 

METHOD 

AIM 

This study aims to describe the impact of AAC device 
implementation and teacher training, in a classroom situ-
ation, on the teacher's interactions with the AAC'~ser. The 
types of utterances that were directed at the speakingchil~' 
dren were compared to the utterances that were directed 
at the children with LNFS and the AAC user. Furthermore" 
the use of questions and the selection of messages on the 
AAC device, by the teacher, during each of the four rese1rch·· 
phases were determined. The results of each of the rese~rch ': . 
phases were compared in order to ascertain the effe~t of ' ,'. 
the implementation of the VOCA, the teacher training ~es- " 
sions as well as the period of withdrawal. 

RESEARCH DESIGN 

Although there were nine children in the class of whom 
three had limited or no functional speech, only one AlAC 
device was implemented. The focus of the study was on the 
teacher's interaction with the individual who was using the 
AAC device. A single case quasi-experimentaL design with 
multiple measurements (ABA with withdrawal) was used 
as it is ideally suited to studying the effect of individual 
variables on communicative behaviour (in this case class-
room interaction) and provides empirical data for clinical 
interventions. // . 

There were four major phases in the research, namely 
A1, B1, B2 and A2. During each research 'phase six meas-
urements were made of the teacher interacting with the 
whole class and six measurements were made of the teacher 
interacting with the AAC candidate on an individual basis, 
during a period of two weeks. It was necessary to use six 

The South African Journal of Communication Disorders, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



The impact of a digital speaker on a teacher's interaction with a child with limited functional speech 75 

measurements to ensure that consistent baseline data was 
obtained. Group as well as individual sessions were used 
in order to determine whether the teacher's interactional 
style was influenced by the situation. 

Ir..WLEI\IENTATHIN TIt,\II\'I;'\(;SI· ...... '\!lJ:'\.... \\TIIIIJI{.\\\,\L 
('FTHE V(ICc\ WITI[TIIE"E.\CIIER fIFUINTr\CT 

SIX SIX SIX SIX SIX SIX SIX SIX 
GIU>Uf' INI)JVJl)UAL GROUP INDIVI])UAL GROUI' IN[)[VIIJUAL GROUP !Nr)JVlllLi,\i_ 

1 Ll I rLr 
TWELVE TWELVE TWELVE TWELVE 

MEASUREMENTS MEASUREMENTS rvlEASURErdENTS MEASUREMENT" 

1 1 1 1 
"RE pelST P(IST I'OST 

I 
II\'II'LEMENTAT((IN 'I\II'LEI\IENTATI( IN T1{AININ(; IVITHIll<AW,IL I 

I'UASE PHASE I'II,.\SE PHASE 

AI III I\~ A2 

FIGURE l:'The research phases 

During the measurements ofthe first phase the child was 
not using the digital speaker in the class (pre-implementa-
tion phase, A1). After this phase was completed the digital 
speaker was implemented in the classroom. Prior to this the 
AAC user had only used the digital speaker during speech 
therapy sessions. Although the AAC user was trained to op-
erate the device the teacher had not received any training 
yet (post-implementation phase or, alternatively, pre-train-
ing phase, B1). After this phase five training sessions, of 
approximately one hour each, were held on five consecutive 
days. After the training was completed more measurements 
were made (post-training phase, B2). A withdrawal phase of 
eight weeks followed (post-withdrawal phase, A2). A pilot 
study was conducted in order to determine the effectiveness 
of the' recording and analy~is procedures. 

// 

SUBJECTS . 
Description of the teacher i 

The teacher was a fifty-one (year old,female wi th approxi-
mately twenty years teachirtg experience, seventeen of 
which had been in teaching ~hildren with disabilities in 
the same school where the research was conducted. Her 
professional preparation included a three year teaching 
diploma. 

Description of the AAC user 

The AAC user was a 4,8 year old male. He had no sib-
lings and his home lC!-nguage was Afrikaans. His primary 
disability was described as a developmental delay with 
ataxia as an additional complication. 

Before the digital speaker was implemented he was ex-
posed to PCS (picture communication symbols) during 
therapy and could use it to express a few basic needs. He 
did not use it in the classroom, however, but depended on 
vocalisations and gestures to communicate. His expressive 
vocabulary was less than 15 words and consisted of the 

words mother, grandmother as well as his teacher's name. 
He voluntarily used vocalisations to express desires, gain 
attention, maintain contact during an interaction, express 
denial and protestations, to obtain an object as well as to 
comment on an event. He visually attended to others, 
tracked moving objects and maintained eye-contact during 
interactions in all environments and situations. He also 
sometimes turned his head away as a signal that he did 
not wish to engage in interaction. He often used intentional 
gestures to manipulate the environment around him and 
he would respond to the use of gestures by others. He pri-
marily used gestures to gain attention, direct attention to 
a specific object or event and to communicate a basic need. 
The use of recognisable facial expressions also occurred 
spontaneously and appropriately in all situations. He of-
ten used facial expressions to indicate a response to ques-
tions and statements. He spontaneously initiated interac-
tions with peers and certain, familiar adults but did not 
exhibit appropriate turn-taking behaviours. Furthermore 
he was not able to repair conversational breakdowns or 
provide revisions of unclear messages. 

According to the Word-Findings Vocabulary Scale -Re-
vised, (Renfrew, 1972) he obtained a score of zero for his 
expressive vocabulary as he could not name any ofthe items 
in the test. The speech therapist and the'teacher agreed 
that this was a reliable reflection of his expressive language 
abilities. According to the Peabody Picture Vocabulary Test 
-Revised, (Dunn & Dunn, 1981) his score placed him in the 
lowest possible category. The speech therapist, however, did 
not consider the test to be an accurate reflection of his re-
ceptive vocabulary as he did not co-operate during the evalu-
ation. However, an in depth analysis (by the researcher) of 
his functioning within the classroom context (in terms of 
the vocabulary he could understand as well as the types of 
commands that he responded correctly to) placed him on a 
level more or less equivalent to that of a three year old 
child. A large discrepancy, therefore, existed between his 
expressive language abilities and his receptive language 
abilities. Hearing tests revealed normal, bilateral hearing. 

Description of the whole class 

This study was conducted at a school for children with 
special needs. The school caters for Afrikaans as well as 
English pupils and has three levels, namely a beginner's 
phase, a junior phase as well as a senior phase. For the 
purpose of this study each of the children in the class was 
described by a speech language pathologist (Table 1). These 
descriptions were important as they formed the basis for 
the differentiation between the children who had limited 
or no functional speech and the children who were speak-
ing. The information was obtained by means of observa-
tions and formal testing as well as from the school records. 
As shown in Table 1, there were three children who had 
less than fifteen words in their verbal expressive vocabu-
lary (children with LNFS). AnAAC device was implemented 
for one of these three children (child A). 

PROCEDURES 

Teacher training procedures 

Training took place at the school where the research was 
completed. Prior to the training sessions consent was ob-
tained from the school principal for the participation of the 

Die Suid-Afrikaanse Tydskrifvir Kommunikasieafwykings, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



76 

teacher, by means of a formal letter. The letter explained 
the teacher's participation in the research project and the 
importance of the training sessions for the benefit of the 
AAC user. An informal, friendly atmosphere was created 
during the training with the use of flowers, refreshments 
and an informal conversational style (Knowles, 1983). The 
training involved the teacher as well as the AAC user. The 
other children in the class were not involved in the train-
ing sessions so that the trainer could focus on developing 
the teacher's skill in interacting with the AAC user. 

Five training sessions of one hour each were held on 
five consecutive days from llhOO - 12hOO. The time of day 
was chosen to coincide with the break time so that the 
teacher was without responsibilities and could give her 
undivided attention to the training. The sessions took place 
in a separate room that was familiar to the teacher as well 
as the AAC user. Training was done on a one-to-one basis 
to accommodate the teacher as it was felt that she would 
be more comfortable with this arrangement. The training 
sessions included some theoretical discussions as well as 
some practical sessions where the teacher had to apply the 
theory in a real-life situation with the AAC user. This was 
done to ensure that the teacher would be able to use her 
newly acquired skills in the classroom. 

Content of training sessions 

A description of the content of the training sessions is 
given in table 2. From this table it is clear that the training' 
sessions concentrated on the development of operational 
as well as strategic skills. The focus was on the practical 
implementation in the classroom context. 

Data collection procedures 

During each phase six recordings were made of the 
teacher interacting with the whole class and six recordings 

TABLE 1: A summary of parameters for each child 

Elsa Popich and Erna Alant 

were made of the teacher interacting with the AAC user on 
an individual basis. Each of the recordings was approxi-
mately fifteen minutes in length and took place between 
nine and nine thirty in the morning. The curricular activ-
ity with which the class was busy during that time was 
"story time". This activity was chosen because it provided 
the researcher with the best opportunity for observing the 
teacher in interaction. The recordings took place on Mon-
days, Wednesdays and Fridays. All the messages that were 
recorded onto the VOCA were selected by the teacher in 
order to suit the story. It was decided that the teacher should 
select the messages to be recorded onto the device so that 
any changes in the types of utterances recorded could be 
measured and discussed. The messages were recorded onto 
the digital speaker prior to the start of the sessions. Dur-
ing the second research phase (post-implementation phase, 
B1), the speech therapist recorded the messages onto the 
device. Mter the training sessions were completed, how-
ever, the messages were recorded on to the device by the 
teacher. The group as well as individual recordings were 
done in the AAC user's classroom. The group recordings 
were completed first, after which followed the individual 
recordings. The classroom assistant engaged the other chil-
dren in activities such a; drawing in the work area in the 
west end of the classroom while the teacher and the AAC 
user remained in the "story time" area in the east end of 
the room for the individual sessions. A video recording was 
made while an audio cassette recorder, which was attached 
to the teacher with a harness, was used to make an audio 
soundtrack. Mter the recordings were completed they were 
transcribed and analysed by the researcher as well as by 
an independent transcriber. . 

MATERIAL AND EQUIPMENT 

The stories "Goldilocks and the three bears" (Creative 
Child Press, 1990), "Where is Otto" (Hill, 1992) and "Otto 

PARAMETERS DESCRIPTION I , 

Child A B C D E F G H I 
I 
I 

Chronological age I 
(Jan 1996) 4,8 3,0 3,0 4,0 3,6 3,4 3,10 3,5 3,0 

I 
I 
I 

Home language Mrikaans Afrikaans Mrikaans Mrikaans Mrikaans Mrikaans Afrikaans Afrikaans Mrikaa~s 
I 

Primary disability Develop- Cerebral Cerebral Develop- Cerebral . Develop- Cerebral Cerebral Mentally 
mental Palsy Palsy mental Palsy mental Palsy Palsy Impaired 
delay delay delay 

Additional Ataxia Squint Squint Cleft lip None None Epilepsy None None 
complications and palate 

Primary means of AAC (digital) Full Phrases Full Phrases Vocali- Single word Full Gestures 
communication speaker) sentences sentences sations utterances sentences 

More than 
./ 

Verbal Expressive Less than More than 20-40 More than Less than 20-40 more than less than 
vocabulary 15 words 50 words words 50 words 50 words 15 words words 50 words 15 words 

Use of gestures for ,/ 
symbolic commu- Yes No No Yes Yes No No No Yes 
nication 

./ 

The South African Journal of Communication Disorders, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



The impact of a digital speaker on a teacher's interaction with a child with limited functional speech 77 

goes for a walk" (Hill, 1992) as well as pictures from the 
stories were used during the training sessions. 

The researcher made use of overlay grids with 4 spaces 
for pictures while making overlays during the training ses-
sions as well as during the classroom interactions. 

A list oftypes of interactions as well as a list of types of 
questions was used for training the teacher in the use of a 
variety of utterance types as well as for analysing the data 
that was obtained during the forty-eight classroom inter-
actions. 

TABLE 2: Contents of individual training sessions (occurred between phase B1 and B2) 

Day People Aims Theoretical Discussions Practical Sessions Equipment 
Present Used 

1 - teacher - improve operational - technical working of the - the teacher and trainer - grid 
- trainer skills with regards to digital speaker went through the story - story book 

the digital speaker - needs and abilities of and selected relevant - pictures 
and making overlays the individual AAC user pictures for the grid and - digital 

- increase awareness of different types of then recorded messages 
different utterance 

- speaker 
types interaction and adjusted the volume 

- teach strategies for 
adapting to the 
individual child 

2 - teacher - improve skill at using None - a recording was made of - story 
- trainer different utterance the teacher telling the book 
- AAC user types AAC user a story. - story 

- teach strategies for - the recording was used to overlay 
selecting appropriate prompt a discussion on - tape 
overlay items the use of different recorder 

utterance types and - tape 
overlay items - digital 

speaker 

3 - teacher - improve skill at using - the importance of the - the teacher went through - grid 
- trainer a variety of questions use of questions in the the story and selected - story book 

- teach strategies for classroom relevant pictures for the - pictures 
using questions for a - different types of grid, recorded the mes- - digital 
variety of reasons questions sages and adjusted the speaker 

volume 
- the trainer only prompted 

4 - teacher - improve skill at using None - a recording was made of - storybook 
- trainer different utterance the teacher telling the - story 

/ 
/ - AAC user types AAC user a story, using overlay 

/ - teach strategies for the story overlay - tape 
selecting appropriate - the recording was played recorder 
overlay items back and the teacher's - tape 

use of different question - digital 
types and overlay items speaker 
was discussed 

5 - teacher - im~rove~technical . None - the teacher made a story - grid 
- trainer skills with regards to overlay without assist- - storybook 
- AAC user the digital speaker ance - pictures 

and the making of - a recording was made of - story 
overlays as well as the the teacher telling the overlay 
use of various types of AAC user a story, using - tape 
utterances and the story overlay recorder 
questions - the recording was played - tape 

- teach strategies for back and the teacher's - digital 
adapting the digital ' use of different utterance speaker 
speaker to the child as and question types was 
well as selecting discussed 
appropriate items for - the number of items on 
the overlay the overlay, the length of 

the messages as well as 
how the AAC user coped 
was discussed 

Die Suid-Afrikaanse Tydskrifuir Kommunikasieafwykings, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



78 

Twenty-five TDK D60 tapes, a Phillips D628 cassette 
recorder, a Panasonic NV-S20B video camera, forty-eight 
VHS video cassettes as well as an Aiwa condenser lapel 
microphone were used to record the training sessions as 
well as the classroom interactions. 

A Blackhawk digital speaker from Adam Lab was used 
as the voice output device for the AAC user, during the train-
ing sessions as well as during the classroom interactions. 

DATA ANALYSIS PROCEDURES 

Transcriptions 

The audio soundtrack was transcribed verbatim and the 
video screen display was used to clarify to whom each state-
ment had been addressed. The external rater simultane-
ously made an independent transcription and analysis, and 
points of disagreement were reconsidered until 100% agree-
ment was achieved. In order to ensure that the raw data 
was transcribed consist.ently correctly by both transcrib-
ers, certain transcription rules were followed (Stuart, 
Vanderhoof & Beukelman, 1993). Repetitions of words were 
included; vocalisations that were not actual words were 
represented and transcribed in a consistent form, e:g. 
mhmmm, uhhuh, huhuh, mmm, uh, ah, aw, whoop; num-
bers were typed as proper nouns; contractions were typed 
as such that the proper form was spelled out only when it 
was spoken that way, e.g., don't was typed as don't and do 
not was typed as do not; standard abbreviations were in-
cluded, e.g., Dr.lMrs.; during the transcription of a commu-
nication segment that was unintelligible and the entire 
segment was skipped even when a few intelligible words 
were available. A spell checker was used on all transcripts 
before proceeding with further analysis. 

Analysis of categories 

After the audio recordings of the interactions had been 
transcribed verbatim the researcher determined the types 
of verbalisations that the teacher directed at each child. 
Twelve types of verbalisations had been defined (adapted 
from a study by Romski, Sevcik, Reumann & Pate, 1989), 
and the teacher's verbalisations during each session were 
categorised accordingly. The teachers choice of utterances 
to be recorded on the digital speaker, for the AAC user, for 
each story session, was also categorised according to the 
twelve types of verbalisations. The twelve types of 
verbalisations are as follows: 

1. Questioning: A sentence adapted by order of words, 
punctuation or intonation to elicit information (e.g. "Is 
it a shirt? Sharon? Tell me ... "). 

2. Attention directing: To guide the child's thoughts to a 
specific topic (e.g. "Look here"). 

3. Answering: A reply to a child's verbalisation, but not 
merely negating or affirming (e.g. ''Yes you may"). 

4. Requesting: Asking for an action, object or comment, 
. but not a question (e.g. "Fetch me the cow"). 

5. Imitating (the child): To mimic the child's utterance, 
with or without expansion (e.g. "I want" "I want"). 

6. Naming: To designate an object or action (e.g. "It's a 
shirt"). 

7. Negating: To imply that what the child said or did was 
wrong (e.g. "No it's not a shirt, Uh-uh"). 

8. Affirming: To imply that what the child said or did was 

Elsa Popich and Erna Alant 

right (e.g. "You are very clever"). 
9. Greeting: A salutation (e.g. "Hello friends"). 
10. Self-repetition: To reproduce one's own verbalisations 

in more or less the same form (e.g. "Whatis it? What is 
it?"). 

11. Informative: To state facts or opinions about a certain 
topic (e.g. "They run with these pants"). 

12. Uncodable: The utterance does not fall into one of the 
above communication function categories(e.g. "Huh?" 
or an incomplete sentence). 

Utterances categorised as questioning were analysed 
further according to Bloom's taxonomy, (Wilson, Lanza & 
Barton, 1988). The categories within Bloom's taxonomy are 
as follows, namely: 

1. Levell - knowledge: The simple recalling of information 
that was given earlier (e.g. "What happened in this pic-
ture?"). 

2. Level 2 - understanding: The awareness of interpreta-
tion abilities. This includes explaining the meaning of a 
word as well as answering why questions (e.g. "What is 
this?"). 

3. Level 3 - application: Applying what is already known 
or previously learnt in a different context (e.g. "Do you 
like porridge?"). 

4. Level 4 - analysis: The ability to break information down, 
into its separate elements to determine the interrelat-
edness of the elements (e.g. "Why were the bears cross 
with Goldilocks?"). 

5. Level 5 - synthesis: The integration of elements into a 
meaningful whole (e.g. ':What did Goldilocks do in the 
house?"). 

6. Level 6 - evaluation: Requires a well motivated critical 
evaluation of or reflection on the material (e.g. "What 
should Goldilocks have done when she saw that nobody 
was horneT'). 

Statistical analysis 

A bar chart was used in order to display the utterance 
types that the teacher directed at the speaking children: in 
comparison with the utterance types that she directed lat 
the children with LNFS. A bar chart was also~ used to co~­
pare the types of questions that the teacher directed at the 
AAC user during the different research phases. EWMA 
charts were used to display changes in the teachers use lof 
answers 'and imitations during the four research phas~,s. 
An EWMA chart for quality control detects a shift in tl'ie 
standard deviation of 0,50 within 12 observations and can, 
therefore, highlight where significant changes occurred. Pie 
charts were used to display the types of messages that were 
recorded, by the teacher, onto the AAC device. 

RESULTS AND DISCUSSION 

TYPES OF INTERACTIONS USED 

Figure 2 gives an overview of the types of utlerances' 
that the teacher directed at the children in the class. The 
figure indicates ~hich three utterance type's were directed 
most frequently at each of the individuals. In other words, 
of all the utterances directed at a rra~ticular child during 
the research (100%) the proportions of the three utterance 
types that were most frequently directed at the child are 

,/ 

The South African Journal of Communication Disorders, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



The impact of a digital speaker on a teacher's interaction with a child with limited functional speech 79 

given in Figure 2. The teacher's interactions with each of 
the children with LNFS (child A, child F and child I) is 
characterised by the frequent use of requesting (such as 
"Show me") and attention directing (such as "look there"). 
Although child A, the AAC user, had access to a device dur-
ing three of the four research phases the teacher's interac-
tions with this child still contained a large proportion of 
requesting and attention directing. The teacher's interac-
tions with the speaking children (child B, child D, child E 
and child H), however, is characterised by the frequent use 
of answering (such as "because ifhe doesn't put ajersey on 
he will get cold") and affirming (such as "yes, he is a naughty 
dog"). There is, therefore, a trend in the differences between 
the types of interactions that the teacher directs at these 
two groups (children with LNFS and speaking children) of 
children. Although the teacher's interactions with the AAC 
user (child A) resembles her interactions with the other 
children with LNFS the teacher also directs a large pro-
portion of affirming utterances at the AAC user. This is an 
utterance type that was frequently directed at the speak-
ing children. The teacher's interactions with the AAC user, 
therefore, resemble the interactions with the children with 
LNFS as well as the interactions with the speaking chil-
dren. The teacher's interactions with the children with 
LNFS as well as with the speaking children did not change 
significantly during the four research phases. 

The results of this study correlate with previous research 
findings which found that the quality of interactions di-
rected at speaking children differs from the quality of in-
teractions directed at the children with LNFS (Mirenda & 
Donnellan, 1986). A previous study that was conducted by 
the authors indicated that the type of interactions used by 
the teacher is affected by the child's level of verbal output 
(Popich & Alant, 1997). Research has also indicated that 
teachers use more direct questions, attention directing and 
requests as well as less answers, imitations and praise with 
children with LNFS (Romski, Sevcik, Reumann & Pate, 
1989; Cicognani & Zani, 1992). 

Although the teacher's interaction with the other chil-
dren in'the class was consistent over the entire period of 
research, the teacher's interaction with theAAC user (child 

/ 

Children with limited or nolfunctional speech 

30 

25 
20.6 20.69 

21.43 21.69 21.08 

20 

15 
15.06 

10 

5 

0 

A F B D 

A) was not consistent over the entire period of research. In 
the teacher's interaction with theAAC user, the use of some 
utterance types such as answering and imitating varied 
during the four research phases. The EWMA chart was used 
to display patterns in the teacher's use of answering and 
imitating with the AAC user in the group sessions and in 
the individual sessions. The EWMA chart for quality con-
trol should detect a shift in the standard deviation of 0,50 
within 12 observations. Increases or decreases in the fre-
quency of occurrence of a certain utterance type are there-
fore displayed. 

Figure 3 displays patterns in the teacher's use of an-
swering and imitating with the AAC user. The x-axis rep-
resents the measurements over time (twenty-four individual 
sessions as well as twenty-four group sessions were ana-
lysed). Measurements one to six occurred during phase one 
(pre-implementation phase), measurements seven to twelve 
occurred during phase two (post-implementation phase), 
measurements thirteen to eighteen occurred during phase 
three (post-training phase) and measurements nineteen to 
twenty-four occurred during phase four (post-withdrawal 
phase). They-axis represents the percentage of answers or 
the percentage of imitations that were used during each 
classroom interaction. In other words, of all the utterances 
that were directed at the AAC user during a specific ses-
sion (100% for each session) the proportion of answers or 
imitations that occurred during each session is presented 
in this figure. The percentages for each session can be com-
pared with the percentages obtained during other sessions 
in order to determine trends. Trends in the teacher's use of 
answering utterances with the AAC user during the indi-
vidual sessions can be seen in the graph. There is a dra-
matic increase in the use of answering from measurement 
thirteen. The five training sessions were held in between 
measurements twelve and thirteen. This means that the 
teacher used answering more frequently after the training 
sessions. A similar change in the use of answering can be 
seen in the teacher's interactions with the AAC user dur-
ing the group sessions. During the second half of the re-
search, in the individual sessions as well as in the group 
sessions, the teacher used answering utterances more fre-

Speaking children 

25.61 

E H 

o Requesting 
D Attention directing 
• Affirming 

~Negating 

!Il Questioning 
• Answering 

• Informative 

FIGURE 2:An overall comparison between the types of interactions directed at the children with limited or no 
functional speech ~nd the speaking children 

Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



80 

quently. Contrary to the change in the teacher's use of an-
swering the use of imitating utterances already increased 
after the implementation of the AAC device. There is an 
increase in the teacher's use of imitating utterances dur-
ing individual sessions as well as group sessions from the 
seventh measurement. In other words, not only did the voice 
output device enable the AAC user to interact, but it also 
resulted in an increase in utterances that were as a result 
of the AAC user interacting (teacher imitating the AAC 
user). 

TYPES OF QUESTIONS USED 

Questioning is one of the most important and most fre-
quently used teaching strategies. It is, however, important 
to look at the types of questions that are used as they could 
significantly influence the amount of initiations made by the 
children, the types of responses given, the amount oflearn-
ing and the level of thought obtained (Cicognani & Zani, 

Phase 1 Phase 2 

OJ 30 
I: 

§ 25 
'E 

20 
Cl 

15 I: .;: 

" Q) ~ 10 I fIl 
I: 

5 <I: , 
;,!! 

B 
0 

0 

Elsa Popich and Erna Alant 

1992; Fowell & Lawton, 1992; Greenberg & Woodside, 1994). 
Figure 4 displays the teacher's overall use of different 

types of questions for each research phase. There is no sig-
nificant difference between the teacher's use of questions 
in the pre-implementation phase and the post-implemen-
tation phase. During both these research phases the teacher 
only used three types of questions, namely, knowledge, 
understanding and a small percentage of application ques-
tions. According to Bloom's taxonomy these three types of 
questions are the lowest three cognitive levels of question-
ing. After the teacher received training, however, there was 
a noticeable change in the use of questions. During the post-
training phase and the post-withdrawal phase the teacher 
used questions on all six cognitive levels, with the majority 
of the questions on level 3 (application). Although the 
teacher used all six types of questions in the post-training 
and post-withdrawal phases only a small percentage were 
on the higher cognitive levels (level 4, level 5 and level 6). 

Phase 3 Phase 4 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 

Measurements over time 

---Answering: individual sessions u Answering: group sessions 

---Imitating: individual sessions - - 'Imitating: group sessions 

i 
FIGURE 3. The teacher's use of answering and imitating with th eAAC user during individual and group sessions 

70 .-----------------~~-------------------------------------------: 
~ 60 ~~~----------------­
e.... 50 c: 
g 40 
(5 30 
g- 20 
a: 10 

o 
Pre-implementation Post-

implementation 
Post-training Post-withd rawal 

/ 

.-------------------------------------------------------------------~/ 
• Level 1 questions: knowledge 

III Level 3 questions: application 
III Level 5 questions: synthesis 

III Level 2 questions: understan~ing 
• Level 4 questions: analysis/·· 

Jjij3 Level 6 questions: evalu6tion· 

FIGURE 4. The teacher's overall use of different types of questions with the AAC user during each phase 
/ 

The South African Journal of Communication Disorders, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



The impact of a digital speaker on a teacher's interaction with a child with limited functional speech 81 

MESSAGES RECORDED ONTO THE DIGITAL 
SPEAKER 

During each of the research phases where the AAC user 
was using the VOCA in the classroom (phases B1, B2 and 
A2) the teacher had to decide what messages to record on 
to the device. The messages that were recorded on to the 
device would determine the AAC user's access to relevant 
material so that he could participate in classroom discus-
sions. The presence of certain types of messages such as 
answering messages as well as the variety of message types 
can be seen as a means of determining the extent to which 
the device was being utilised in the classroom. 

Figure 5, Figure 6 and Figure 7 display the types,of 
messages recorded onto the digital speaker during the post-
implementation phase, the post-training phase and the 
post-withdrawal phase during group as well as individual 
sessions. During the post-implementation phase (Figure 5) 
the teacher recorded a limited variety of messages (nam-
ing, informative and questioning) and the largest propor-
tion was naming (79%). After the teacher received train-
ing, however, a larger variety of messages were recorded 
(Figure 6) and the percentage of naming messages de-
creased to 2%. The largest proportion ofthe messages, dur-
ing the post-training phase (Figure 7), was answering (29%). 
After the period of withdrawal the answering messages 
decreased to 7% and the naming messages increase again 
to 13%. Furthermore, a couple of previously unused mes-
sage types (negating=2% and greeting=2%) also occurred. 
The increase in the number of different message types with 
each consecutive research phase indicates that the AAC 
user had access to a greater variety of utterance types and 
could express himself better. Furthermore, the increase in 
the number of different message types in the final two 
phases of research correlates with the increase in the teach-
er's use of a greater variety of question types. After the 
training sessions the teacher was, therefore, more aware 
of the number of different types of messages that could be 
recorded as well as the range of questions that could be 
used in order to facilitate the AACuser's language devel-

/ 
opment. 

,/ 

CONCLUSION 
i 

The quality of interaction it hat was directed at the speak-
ing children was different to; the quality of interaction that 
was directed at the children with LNFS. The teacher's in-
teraction with these childre~ was characterised by the fre-
quent use of requesting and att~ntion directing while the 
interaction with the speaking children was characterised 
by the frequent use of answering and affirming. The im-
plementation of the AAC device (phase B1) resulted in an 
increase in the teacher's use of imitating and answering 
with the AAC user. After the teacher had received training 
(phase B2) there was' afurther increase in the use of an-
swering. These responses by the teacher, to the AAC user's 
initiations can be seen as encouragement to tl:te AAC user 
to participate. According to a study done by Hart and Risley 
(1995) children who receive input that is responsive and 
encouraging perform significantly better. The implemen-
tation of the digital speaker and the training of the teacher 
in the use of different utterance types resulted in the crea-
tion of a richer language environment for the AAC user. 

Although the implementation of the AAC device resulted 
in an increase in the teacher's use of imitation and answer-

Informative 
19%· 

Questioning 
2% 

FIGURE 5: The types of messages recorded into the dig· 
ital speaker during the post. implementation phase 

Requesting 
20% 

Answering 
29% 

Naming 
2% 

Attention directing 
3% 

Uncodable 
12% 

FIGURE 6: The types of messages recorded onto the 
digital speaker during the post· training phase. 

Uncudablc 
10% 

r\lIenlinn dirccling 
H% 

DAttention directing 

IIIIAnswering 

II Requesting 

DNaming 

II Negating 

DGreeting 

IIIIlnformative 

IIUncodable 

DQuestioning 

FIGURE 7: The types of messages recorded onto the 
digital speaker during the post·withdrawal phase 

Die Suid-Afrikaanse Tydskrifvir Kommunikasieafwykings, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)



82 

ing with the AAC user the teacher's interaction with the 
other children in the class remained the same throughout 
the four research phases. As mentioned previously the AAC 
user was not the only child in the class with LNFS. The 
fact that the teacher's interaction with the other two chil-
dren with LNFS did not change as a result of the training 
that the teacher received is interesting. This confirms the 
notion that skills acquired while working with a particular 
child with LNFS do not automatically carryover to other 
children. Although one could argue that the other students 
with LNFS did not have the same access to messages as 
they did not use VOCAs, the trend in the behaviour of the 
teacher could indicate the limitations of technology focused 
training for implementation. The question of how to en-
sure that AAC implementation is beneficial for the whole 
class must be considered! 

Not only was the teacher's use of interaction types in-
fluenced by the training sessions but the teacher's use of 
questions was also influenced. The teacher's use of ques-
tions on higher cognitive levels was positively influenced 
by the training sessions. This means that implementation 
of an AAC device alone will not result in the best possible 
interactional environment. Training of the interactional 
partners should be part of the implementation process. This 
underlines the importance of training for the teacher of an. 
AAC user in order to increase the AAC user's exposure to 
interactions that result in maximum cognitive development. 

The teacher did not spontaneously record a variety of 
messages onto the digital speaker. The AAC user only had 
access to a limited variety of utterance types until the 
teacher received training. One can see, by the teacher's 
choice of a greater variety of messages on the AAC device 
that, by the final two research phases, the teacher had be-
come more sensitive to the needs.of the child. Access to a 
limited variety of utterance types could impair the AAC 
user's participation and development. It is, therefore, criti-
cal that the teacher of an AAC user should recei ve adequate 
training on the selection of appropriate items to be included 
on the device display. Research has found that AAC users 
in residential settings direct less than 5% of their interac-
tions at communication partners who have not received 
formal training (Dalton & Bedrosian, 1989). Formal train-
ing of communication partners, therefore, forms a critical 
part of the AAC implementation process. 

ACKNOWLEDGMENTS 

The financial assistance of the Human Sciences Research 
Council (HSRC) towards this research project is gratefully 
acknowledged. Opinions expressed and conclusions arrived 
at are, however, those of the author and are not necessarily 
to be attributed to the HSRC. 

REFERENCES 

Baker, L.A. (1993). Educational challenges: strategies for theAAC 
student. Communication Outlook, 15(1): 13-16. 

Basil, C. (1992). Social interaction and learned helplessness in 
.. severely disabled children. Augmentative and Alternative Com-

munication, 8: 188-199. 

Elsa Popich and Erna Alant 

Beukelman, D.R (1991). Magic and cost of communication com-
petence. Augmentative and Alternative Communication, 7(1): 
2-10. 

Blackstone, S.W. (1986). Augmentative communication: an intro-
duction. American Speech-Language-Hearing Association, 
U.S.A. 

Camp, W.G. (1993). Improving your teaching: questioning tech-
niques. The Agricnltural Education Magazine, 66(5): 17-23. 

Cicognani, E. & Zani, B. (1992). Teacher-children interactions in 
a nursery school: an exploratory study. Language and Educa-
tion, 6(1): 1-12. 

Creative Child Press (1990). Goldilocks and the three bears. 
Playmore, Inc. Publishers, New York. 

Dalton, B.M. & Bedrosian, J.L. (1989). Communicative perform-
ance of adolescents with severe speech impairment: influence 
of context. Journal of Speech and Hearing Disorders, 54: 403-
421. 

Dunn, L. & Dunn, L. (1981). Peabody Picture Vocabulary Test -
Revised. American Guidance Service: Minnesota. 

Foweli; N. & Lawton, J.T. (1992). Discrepancies between ques-
tions and responses during small-group instruction in two 
preschool programs. Early Childhood Research Quarterly, 7(3): 
415-439. 

Greenberg, K.H. & Woodside, M.R (1994). Differences in the de-
gree of mediated learning and classroom interaction struc-
ture for trained and untrained teachers. Journal o/Classroom 
Interaction, 29(2): 1-9. 

Hart, B. & Risley, T. (1995). Meaningful differences in the every-
day experiences of young American children. Paul H. Brookes, 
Baltimore. . ~ 

Hill, E. (1992). Wdar's Otto? Dieter Zimmerman, Johannesburg. 
Hill, E. (1992). Otto gaan stap. Dieter Zimmerman, Johannes-

burg. 
Knowles, M. (1983). Andragogy: an emerging technology for adult 

learning. In Malcolm Tight (Ed.), Adult Learning and Educa-
tion, 1. 

Light, J. (1988). Interaction Involving Individuals Using Aug-
mentative and Alternative Communication Systems: State of 
the Art and Future Directions. Augmentative and Alterna-
tive Communication, 4(2): 66-72. 

Locke, P.A. & Mirenda, P. (1992). Roles and responsibilities of 
special education teachers serving on teams delivering AAC 
services. Augmentative and Alternative Communication, 8(3): 
200-209. 

Mirenda, P. & Donnellan, A. (1986). Effects of adult interaction 
style on conversational behaviour in students with severe com-
munication problems. Language, Speech and Hearing Serv-
ices in Schools, 17(2): 126-139. 

Popich, E. & Alant, E. (1997). Interaction between a teacher and 
the non-speaking as well as speaking children in the class-
room. The South African Journal of Communication Disor-
ders, 44: 31-40. I 

Renfrew, C.E. (1972). Word-finding Vocabulary Scale. Renfrew 
Publications, Oxford. i 

Romski, MA., Sevcik, RA., Reumann, R & Pate, J. (1989). Young-
sters with moderate or severe mental retardation and severe 
spoken language impairments 1: Extant communicativelpat-
terns. Journal of Speech and Hearing Disorders, 54: 366-~372. 

Smith, M.M. (1994). Speech by any other name: The role of dom-
munication aids in interaction. European Journal of Disor-
ders of Communication, 29(3): 225-240. I 

Soto, G. (1997). Special education teacher attitudes towardAAC: 
preliminary survey. Augmentative and Alternative Commu-
nication, 13(3): 186-197. 

Stuart, S., Vanderhoof, D. & Beukelman, D.R (1993). Topic and 
vocabulary use patterns of elderly women. Augmentative and 
Alternative Communication, 9(2): 95-107. 

Wilson, C.C., Lanza, J.R & Barton, J.S. (1988). Developing higher 
level thinking skills through questioning techniques in the 
speech and language setting. Language, Speech and Hearing 
Services in Schools, 19(4): 428-431. 

The South African Journal of Communication Disorders, Vol. 46, 1999 

R
ep

ro
du

ce
d 

 b
y 

Sa
bi

ne
t G

at
ew

ay
 u

nd
er

 li
ce

nc
e 

gr
an

te
d 

by
 th

e 
P

ub
lis

he
r 

( 
da

te
d 

20
12

)