Reconceptualizing and decentering think-aloud methodology in qualitative research

Abstract

Most applications of think-aloud protocols have been conducted from theoretical perspectives that prioritize knowledge that is predictable and controlled by the researchers. In this article, we present an augmented form of the think-aloud method in which we aim to gain situated and participant-generated knowledge. The context for our study is examination of the problemsolving processes used by engineering students. We illustrate how our adaptation of traditional think-aloud protocols provides insights into participants' thoughts and beliefs and how such thinkalouds can increase social scientists' understandings of complex phenomena such as learning or problem solving. In contrast to a typical focus on researcher-defined processes or an analysis of the products generated by students, our approach to think-aloud utilizes think-aloud procedures in combination with follow-up interviews to expand participants' perspectives and investigate their experiences more deeply. Keywords constructivism, problem solving, think-aloud methodologyThe fields of social research and qualitative research include various methodological and theoretical traditions, many of which have been developed in different disciplines (see e.g.

Positive thinking and physical activity motivation for one individual with multiple sclerosis: A qualitative case-study

“…The think-aloud process is an established method of collecting qualitative data (Al-Janabi et al 2013, Anderson-Lister and Treharne 2014, Lundgrén-Laine and Salanterä 2010, Koro-Ljungberg et al 2013, McGavock and Treharne 2011 with some strengths and limitations. Think-aloud interviews are less like a natural conversation than semi-structured interviews and require considerable attention in answering the questionnaires as well as verbal questions, which can be fatiguing (although this was not the case for Norman).…”

“…In this article we describe an in-depth qualitative analysis of a single case, which focuses on our interpretation of Norman's experience (Braun and Clarke 2013, Radley and Chamberlain 2001, Koro-Ljungberg et al 2013, Thomas 2010, Willig 2008. We used the concurrent think-aloud method, a valuable method in health research (Adamson et al, 2004), to gather qualitative and quantitative data, and understand Norman's perspectives in-depth (Lundgrén-Laine and Salanterä 2010).…”

"Hún gæti alveg verið múslimi og allt það": Ráðning fólks af erlendum uppruna til íslenskra fyrirtækja

“…Rannsóknin byggir í grunninn á eigindlegum viðtölum, en innan eigindlegrar rannsókn-ar-hefðar á "að hugsa upphátt" aðferðafraeðinni, þar sem viðmaelendur hugsa upphátt um ákveðið ferli (Koro-Ljungberg et al 2013;van Someren et al 1994). Eins og fyrr var minnst á hefur þessi aðferð verið notuð við ólík rannsóknaviðfangsefni og þar á meðal hvernig einstaklingar komast að niðurstöðu um ákveðna hluti, eða við úrvinnslu ákveðinna verkefna (Atman et al 2007;Hoppmann 2009;Koro-Ljungberg et al 2013).…”

Exploring the Feasibility of an Educational Computer Game as a Novel Means of Assessing Problem Solving Competencies

“…Use of metacognitive strategies during problem-solving has classically been studied in-situ using think-aloud protocols 4,14,15 though there are some documented limitations of the procedure in balancing obtaining rich, meaningful data without significantly disrupting the participant's thought process itself. Psychometrically validated instruments (e.g., the Critical Thinking Dispositions Scale 16 ) can also be used but are limited due to their reliance on self-report Likert-scale item response separate from a specific problem-solving context (in contrast to using an anchoring problem exercise in a think-aloud protocol).…”

“…Most methods of researching problem-solving process involve self-report data which, while validated, still have limitations associated with the difference between individual perception and observable action. Building on the work of others to address such limitations, 4,5 we prioritize research protocols that move beyond pairing of achievement (right/wrong answers) and self-reported reasoning process (think-aloud protocols) to include observable data during problem-solving process that can corroborate (or refute) the self-report data.…”