A number of limitations in this pilot study should be taken into consideration when looking at its findings.

The match between each communicatively impaired child and a younger child at the same level of verbal comprehension was made on the basis of the age of the younger child. and not on the basis of his or her performance score on the relevant test. This match must therefore be regarded with some degree of caution (Jordan and Murdoch, 1987).

Another limitation to the current study is the small number of subjects involved. This was due to the restricted time and resources available for the recording of the data and to the time-consuming nature of the analysis. The analysis of conversation makes considerable demands of time and expertise (Edelsky, 1981; Inglis, 1987). Adams and Bishop (1989) looked at "use of unestablished reference" (p. 237) by 124 children, each audio recorded in conversation with an adult investigator. However, because of the time-consuming nature of such an analysis, they focussed only on pronouns and demonstratives, since these could be identified using computerised search procedures. Investigations on the scale of that carried out by Adams and Bishop but looking at every instance of unestablished referent, could be undertaken using the P.E.R. as a framework for identification and analysis, and are essential before the findings of the current study can be confirmed.

In addition, the quantitative part of the current study looked only at those instances of unestablished referent where the subject was speaker. Taking the pragmatics-as-cause-effect approach, it has been argued that unestablished referent is mutually constructed in the interaction between conversational is thus limited by its failure to include those instances of unestablished referent where the subject was addressee.

The limited resources available also affected the collection of data for the current study. The role played in establishing referent by the visual context and by non-verbal communication, especially by gesture. makes the use of video recording mandatory for an analysis of this kind. However the use of only one camera restricts what can be recorded of conversational partners' dual roles, as both speakers and addressees. In the current study a specific omission on account of this limitation was the recording of eye gaze (see 3.4.6.c). Other omissions would have been avoided if a second investigator had been available to operate the camera (preferably by remote control, as less inhibiting to the children). In particular this would have avoided the limited recording of non-verbal aspects of the adult's contributions to the conversation. This occurred because, when the children left the table where the conversation started, the adult followed them with the camera hand-held and was therefore not in vision. While the informality of "photographing Louie" in this way seemed conducive to the relaxed participation of many of the children, it was not conducive to a high technical level of video recording, especially in combination with the extreme brightness of the Queensland sun and the background noise provided by the more vociferous local insects. Limited resources also made the transcription of the data from the video tape more arduous than necessary. It is recommended that transcription of this kind be done if at all possible on a three or four head VCR (which has the facility of a steady single frame image as needed) and with a high fidelity sound system, preferably with a head phone socket.

Go to Chapter 6
Return to Contents
Hosted by