Preliminary Investigation of Some Influences on Student Teachers' Self- Efficacy for Teaching with Computers

Although many teachers now have access to computers for teaching and learning and profess a belief in the value of computers for education the impact on the day to day work in classrooms is still limited. Even where teachers believe in the use of computers and have the necessary technical skills they may be reluctant to implement them in their classrooms. Increasing teachers' self-efficacy beliefs for teaching with computers has been proposed as a means of influencing their behaviour. This paper reports on a preliminary study of some factors that might be associated with self-efficacy for teaching with computers. The absence of significant relationships with factors such as Pupil Control Ideology and Innovativeness suggests that it may be possible to intervene to increase graduating teachers' self-efficacy for teaching with computers. In Australia, as in the USA, computers are now widely available in schools of all types. The arrangements for access vary. Some schools place the computers in classrooms and others create computer laboratories. Whatever the arrangements in their school, there are now very few teachers who could seriously claim that they have no opportunity to use computers in support of teaching and learning. There is also widespread support for the increased use of computers among policy makers, educational administrators and the community. In Queensland, the government has provided financial support for Schooling 2001, a policy framework which includes as a goal that "computers should be used in every subject area at every level of education" (Education Queensland 1998). Despite the levels of opportunity and official encouragement implied by these developments, there continue to be concerns expressed about both the frequency and success with which teachers employ information technology in their classrooms (Cuban 1998). Recent online discussions among students returning from teaching practice have included comments about the limited use being made of computers in many classrooms. However, students observed that even teachers who did not use computers in their teaching expressed belief in the importance of computers in education. Surprisingly, on some accounts the rate of uptake of information technology in classrooms does not appear to be notably higher among beginning teachers than among their more experienced colleagues (Oliver 1993). Various reasons have been postulated to explain the apparent reluctance of teachers to embrace IT. On the whole teachers appear to share the broader community enthusiasm for IT in education as a "good thing" but they report a lack of confidence in their capacity to use IT effectively for teaching and in many cases attribute this to inadequate preparation (Albion 1996; Handler 1993). A previous paper (Albion 1999) discussed the importance of teacher beliefs in relation to their adoption of IT for teaching. Self-efficacy for teaching with computers (SETWC) was suggested as an area of belief which was related to teacher behaviour in the longer term, was measurable using existing scales and was capable of being influenced by teacher educators in the context of preservice and inservice courses. Problem-based learning (PBL) has been identified as an instructional design methodology that might offer particular advantages in relation to development of self-efficacy (Albion & Gibson 1998) and a multimedia package using this methodology has been developed (Gibson & Albion 1999). Factors Influencing Self-efficacy for Teaching with Computers If self-efficacy for teaching with computers is a significant determinant of teachers' behaviour in respect of teaching with computers and if PBL is an effective instructional design for influencing self-efficacy, then multimedia using PBL as a basis for design warrants investigation as a means of increasing beginning teachers' SETWC. As part of the development of the multimedia package it was proposed to test this hypothesis by measuring students' SETWC before and after the use of the materials. The Microcomputer Utilization in Teaching Efficacy Beliefs Instrument (MUTEBI) was selected as a measure of SETWC. MUTEBI was initially developed and used in the context of science teaching (Enochs et al. 1993) but it is not specific to that context. It comprises two sub-scales which correspond to the two components of the selfefficacy construct identified by Bandura (1977), namely efficacy expectations, which was labeled by Enochs as self-efficacy (SE), and outcome expectations, labeled as outcome expectancy (OE). Teachers' beliefs and behaviours in respect of computers do not stand in isolation from their other beliefs. Previous studies have demonstrated that teachers' use of computers may be influenced by, among other things, their beliefs about student centred pedagogy (Honey & Moeller 1990) and their inclinations towards innovativeness (Marcinkiewicz 1994). Moreover, students who reported more custodial orientations on the pupil control ideology (PCI) scale also reported lower self-efficacy for science teaching (Enochs et al. 1995). It is often suggested that the use of computers in teaching leads to more student-centred approaches to teaching. Thus, the possibility of a relationship between PCI and SETWC should be considered. The effectiveness of the multimedia package in increasing students' SETWC may depend, at least in part, on the existence and relative strength of these other influences on SETWC. Early knowledge of such factors might guide refinement of the materials design. Hence an investigation of the relationships between SETWC and other aspects of student teachers' belief systems is warranted. Because the multimedia materials were to be developed using an untried design, plans were made to conduct trials of a very early prototype with a small group of students who were representative of those who would use the final product. This group provided a suitable opportunity to trial the instrument that would be used to measure SETWC and to undertake a preliminary investigation of potentially influential factors. It seems reasonable to assume that positive attitudes towards computers and strong self-efficacy for computer use would be prior conditions for positive self-efficacy for teaching with computers. Scales for the measurement of attitudes towards computers and self-efficacy for computer use have been developed and used with teacher education students (Kinzie et al. 1994). The attitude scale has two sub-scales that relate to comfort/anxiety and usefulness of computers. The self-efficacy scale comprises several sub-scales that relate to categories of computer use including word processing, spreadsheets, databases and so on. Subsequent studies have confirmed the reliability of the attitude scales and of a slightly modified version of the self-efficacy scale comprising seven sub-scales (Albion in press). The scales described in the latter paper were used in this study. Scales for the measurement of teacher self-efficacy have been developed and used in several published studies. For the purposes of this study the instrument described by Guskey and Passaro (1994) was selected because it was the most recently published validated scale for teacher self-efficacy. It comprises two sub-scales. The internal sub-scale appears to represent perceptions of personal influence in teaching and learning while the external sub-scale represents the influence of elements beyond the direct control of the teacher. Two other factors were chosen for examination. Pupil control ideology (Graham et al. 1985) is a unidimensional construct which has been found to be related to self-efficacy for science teaching (Enochs et al. 1995). As noted above, PCI could influence SETWC through the relationship between custodial orientation and student-centred classes. It was measured using the instrument published by Graham (1985). Innovativeness has been found to be related to computer use in teaching (Marcinkiewicz 1994). It was measured using the unidimensional instrument published by Hurt et al. (1977). All instruments used in this study were derived from previously published studies. The body of the questionnaire comprised 134 items representing 15 scales and sub-scales as shown in Table 1. Each item was presented as a Likert scale item with the extremes of the range identified as Strongly Disagree on the left and Strongly Agree on the right. The number of points on the range for each scale varied from 4 to 7 according to the usage reported by the originators of the scales. Table 2 records the maximum value attainable on each scale.