Introduction

The focus of this paper is the think aloud (TA) study, which is part of a more extensive study investigating reading skills of advanced learners of English. I concentrate on the dialogic nature of this study, as revealed in the process of analyzing protocol data.

In the process of conducting the study, a number of interrelations between various factors were observed: e.g. between learners' first language (L1) and their foreign language (FL), reader and text or reader and TA method. This paper investigates the dialogue between research and theory that became evident during conducting the study, especially at the stage of analyzing data. It explores a symbiosis between the methodology applied and the very construct the study investigates – reading. In other words, the paper discusses how the TA method used in this study influenced the conceptualization of reading.

With regard to the relationship between theory and research, protocol analysis can be used both inductively i.e. as an exploratory methodology, and deductively i.e. in order to test hypotheses about reading that emerge from initial explorations (Pressley and Afflerbach, 1995). Erickson and Simon (1980) and Afflerbach (2000) emphasize the role that protocol analysis plays in the exploration of the very construct it is used to investigate. Protocol analysis may be first used to initiate the construction of theories concerning reading; to break ground for new understandings of reading.