4.3 Hidden Worlds: Challenges Related to Investigating Beliefs

Much of educational research has traditionally focused on teacher actions which represent observable and measurable phenomena that lend themselves easily to empirical research. Beattie (1995) notes that the major goal of this research was to determine the effects of teacher actions and performance on student achievement. Brophy and Good (1986) describe this type of research as process-product studies. Such research typically concerned itself with causality and was conducted using quantitative research methods. Fang (1996) notes that this research was conducted under laboratory or contrived conditions and describes it as follows:

[Such] research assumed that the relationship between teachers' actions and their observable effects is linear and unidirectional. Data analysis was relatively decontextualized and objectified in a search for positive, generalizable principles which can be used to formulate teaching and learning theories. Statistical procedures used in these studies were often correlations and analysis of variance. (p.48)

In recent years, advances in cognitive psychology have resulted in a shift away from process-product studies towards a growing interest in understanding teachers' thought processes (Fang, 1996). This tendency has led to various attempts at investigating teacher beliefs. However, focusing on what happens inside teachers' heads presents certain obvious research challenges. Pajares (1992) argues that the construct of belief does not lend itself easily to empirical investigation. Beliefs have a very covert nature, might not be observable and even the teachers holding them may not recognize them (Milne Taylor, 1995). Teachers' beliefs are often tacit and inarticulate (Driver Erickson, 1983). They belong to the area of thought processes that occur inside heads and are thus unobservable in the same way that behaviour would be (Clark Peterson, 1986).

Kagan (1992) has identified a number of challenges related to studying beliefs. She notes that they cannot be inferred directly from teacher behaviour, because teachers can follow similar practices for very different reasons. Furthermore, notes Kagan, "much of what teachers know or believe about their craft is tacit. For example, teachers are often unaware of their own beliefs, they do not always possess language with which to describe and label their beliefs..." (p.66). She therefore cautions against a style of direct questioning such as "What is your philosophy of teaching?" and characterizes it as "an ineffective or counterproductive way to elicit beliefs" (p.62).

Varying approaches have been taken by researchers in their attempts to understand teacher beliefs. One of the more common approaches is the use of Likert-type questionnaires to which teachers indicate agreement with direct statements on beliefs. Another approach is that of forced-choice options that require teachers to endorse a particular option. Such approaches, however, are unlikely to provide an accurate indication of teachers' beliefs. Munby (1984) points to a number of drawbacks to such approaches to studying beliefs:

While the items of the instrument generate a response, they may be doing so not because the teacher would necessarily have thought of the belief represented by the items but because the test developer did. In other words, the scores represent what the teacher says is believed when he or she is physically presented with various beliefs of interest to the researcher (and possibly identified by many other teachers), and these do not necessarily correspond to the beliefs which are paramount to the individual teacher's handling of the immediate and unique professional environment. (p.29)

Like Kagan, Woods (1996) emphasizes the tacit nature of beliefs and the difficulties of accessing them. Even in an interview situation, direct questioning can produce misleading results:

...beliefs (and their interrelationships) may not be entirely consciously accessible, and teachers may, in responding to questions about generalized beliefs, answer according to what they would like to believe, or would like to show they believe in the interview context. When a belief or assumption is articulated in the abstract as a response to an abstract question, there is a much greater chance that it will tend more towards what is expected in the interview situation than what is actually held in the teaching situation and actually influences teaching practices. (p.27)

A further problem arises, according to Woods, in relation to the way in which the questions may be posed. Terms like whole language can have varying meanings for different individuals and their use by the interviewer may be misleading. As Woods explains "...language teachers know a lot about themselves in language teaching situations, but although they have had the experiences, they may not have categorized and labeled them. Therefore using abstract questions, symbols and categories that the interviewer might feel comfortable with may not allow themselves to express what is important to them" (p.27).

Fang (1996) expresses a similar concern about the approach of much of the research on teachers' thought processes and beliefs. He posits that studies of teacher cognition and beliefs must address the personal experiences of teachers and their influence on shaping these beliefs. He recommends approaches such as life history, narrative, and autobiography in order to capture the complexities of teacher beliefs. Woods makes similar recommendations and proposes investigating the contexts of what he refers to as teacher stories about events, behaviours and plans. Munby (1984) argues that special attention needs to be given to providing teachers with opportunities to talk about fundamental beliefs.

The present study recognizes the challenges related to studying beliefs. It is in recognition of these challenges that the research was designed. Emphasis in the research design is on providing a means or mechanism to allow teachers the opportunity to bring to the surface their own theories, personal knowledge, or beliefs about the teaching of FSFL in online learning environments. The study has taken a broad focus as a means of providing an initial foray into an area about which little is yet known. Expressed in this study therefore is an interest in understanding the beliefs of both teachers who do and do not operate in OLEs. The broad range allows for an understanding of the beliefs of those who may value and those who may not value the use of the Internet in the teaching of FSFL. The study allowed for participation from a wide range of teachers from diverse educational and geographical backgrounds. At the same time, the study provided an opportunity to probe more deeply into the beliefs of a certain number of the study’s participants.

An approach often taken to the study of teachers’ beliefs is to focus in on a small number of three or four teachers and to probe their beliefs in-depth. The present study takes a very different approach. The study’s data collection techniques allowed the opportunity to probe in-depth the beliefs of certain participants. The teachers involved in the study were able to participate in the large group discussion where many issues were raised and a broad range of topics discussed. However, at the same time, these teachers’ beliefs were also probed in a more in-depth fashion through individual dialogues conducted via e-mail for the discussion participants and via telephone for the questionnaire participants. The range of topics generated allowed the researcher to capture a broader range of beliefs than might have been possible if only three or four individuals participated in the study. The following section provides a brief overview of the data collection techniques.