The presence of a video camera initially could have been disruptive for some, but none of the children mentioned any concerns to me and all of the parents, although more self-conscious, complied with my request to videotape them. I offered examples of some possibilities if they did not independently suggest an activity or routine. The options included shared book reading (Beals, de Temple & Dickinson, 1994), such as reading a story to the child or helping with homework, interactive play (Nickse & Quezada, 1994) which could include family members engaging in a commercial game, sport or other activity together or a mealtime exchange (Beals et al, 1994) or the preparation of part of the meal together. The activities chosen by the participants are presented in Table 3 on the next page.

A guide was developed for analyzing shared book reading and is contained in Appendix H. This included adult book reading style, child behaviours, purpose of parent-initiated dialogue and purpose of child-initiated dialogue. These categories were based on Wade and Moore's (1996) focus points for observing children during book sharing and Kook and Vedder's (1995) variables used in coding book reading episodes. Snow et al's (1991) observation categories and Bottorff's (1994) hints for effectively using videotaped recordings were used for analyzing other types of parent-child interactions.

Parent-child interactions

A second type of naturalistic observation looked more specifically at the nature of the adult-child interactions surrounding literacy events (Mason & Allen, 1986; Scarborough, Dobrich and Hager, 1991; Snow, 1991; Teale & Sulzby, 1987) to determine how adults foster literacy development. The observational data from parent-child interactions were intended to complement the interview data and reveal aspects of behaviour in the participants that would be unable to be recognized or obtained in an interview or from a written form.