Qualitative Interview Analysis: The Use of Systemic Functional Linguistics to Reveal Functional Meanings

Abstract: Qualitative interviewing is a method in which the interviewer directs the interviewee to answer specific research questions. In this article, I present an example of "Systemic Functional Linguistics" (SFL) analysis of an interview in which I investigated a student's use of an instructional tool called "Schema for the Complete Orienting Bases of Action" (SCOBA). SFL, as a qualitative tool for analysis, provides a functional account of the interview text, revealing semantic meanings through the grammatical and lexical choices of the interviewee. During analysis, SFL revealed different layers of meaning that are mediated by the linguistic choices of the interviewee. SFL helped me to get a fine-grained view of how the interviewee construed her experience as a foreign language learner and how she showed awareness and control of her own language development.

Key words: qualitative interview analysis; systemic functional linguistics; foreign language education; functional meanings; new methods and methodologies

The purpose of this article is to demonstrate the usefulness of the application of Systemic Functional Linguistics (SFL) to qualitative interview texts. SFL scholars view text in a functional and semantic way rather than looking at sentences as formal and syntactic objects (YOUNG, 2009). In SFL, text is understood as the unfolding of culture in a social context. Language is seen as a system that realizes three main metafunctions: representing (ideational), enacting social relations (interpersonal), and organizing these two functions into a message (textual). Researchers such as MARTIN and ROSE (2003) have started to use concepts derived from SFL to analyze qualitative data, both textual and multimodal (KRESS & VAN LEEUWEN, 2001). This type of research has spread not only to the social sciences (ACHUGAR & PESSOA, 2009; GIBBONS, 2006) but to other fields, for example health care (EAKIN & MYKHALOVSKIY, 2002) and journalism (CALDWELL, 2009). [1]

SFL emphasizes the study of language use in context to achieve social purposes, such as explaining to someone your daily routine (recount) or explaining to someone your point of view on a topic (opinion). Speakers and writers have different goals (i.e., a lawyer might want to persuade a judge that someone is innocent during a trial, or a writer might want to express sympathy for somebody's loss) that are recurrently enacted in different text type or genres. Genres emerge from the language choices that speakers make according to the social purpose of the text they are producing. According to SFL each genre has a specific register. For HALLIDAY and MATTHIESSEN (2014 [1985]), register is the variety of language used in a specific situation. There are register variations when the situations variates according to the field or the subject matter of the linguistic situation, the tenor, or who is involved in the linguistic situation and the relations between the people involved, and the mode, whether the language is spoken, written, or multimodal. [2]

I argue that SFL offers the methodological means to obtain a rich and deep analysis of qualitative interview texts. To demonstrate how linguistic metafunctions are realized in the interview text, I present examples of text analysis in an excerpt taken from an interview with a student of Italian as a foreign language (IFL). [3]

First, I review the literature regarding text analysis methods, in particular the most common methodologies used to do interview analysis and how SFL has been used to analyze texts (Section 2). Then, I present the interview text and the methodology of data collection and analysis that I use to demonstrate the use of SFL as an analytic tool for interviews (Section 3). Subsequently, I apply the main concepts of SFL to analyze the same excerpt of interview text to show some of the levels of semantic meaning that could be found in a text using SFL (Section 4). Finally, the article's discussion contains remarks about the results of the study and recommendations for future research (Section 5). [4]

An interview is a topic-focalized conversation. A qualitative interview is a method in which the interviewer directs the interviewee to answer specific research questions, very different in nature from a casual conversation. The analysis of this specific type of text requires researchers to align their theoretical framework to the research design and the methods of data collection (ROULSTON, 2014). [5]

Since the 1990s, research interviews have been considered one of the main methods for gathering qualitative data because they give researchers the possibility to extend their knowledge of how people make sense of the world (SEIDMAN, 2013). Usually, qualitative interviews result in a large amount of detailed and contextually laden data. Establishing the position of the researcher and making sense of the research itself is a significant concern in this type of interview (SILVERMAN, 2000). Most of the data collected from interviews is in the form of texts, either oral or written. Other data can be used when the interview is video recorded or is mediated by another communication channel, such as a video call interview; however, the analysis of text remains one of the focal elements of interview analysis (ibid.). [6]

The nature of the interaction in the interview setting, as NIKANDER (2012) asserts, is specific to this particular kind of qualitative research. Qualitative interviewing is a specific genre that enacts particular patterns of textual meaning in a social context of interaction (MARTIN & ROSE, 2003). Moreover, LAMPROPOULOU and MYERS (2013) have asserted that the analysis of the development of the interview dialogue reveals the change in the roles of the interviewer and interviewee. [7]

Social scientists can understand meaning making processes from different perspectives. Whether it comes from a more linguistic tradition, in which texts are the objects of analysis, or from a more sociological tradition, in which texts are cues to enter into culturally human experiences, scientists have utilized diverse methodologies to interpret these processes (DEPPERMANN, 2013; RYAN & BERNARD, 2000). Within the types of methodologies used for text analysis, some analyze words from a quantitative perspective, while others analyze words from a qualitative perspective. Through methodological tools such as word count and semantic networks, researchers review the data to find word frequency and the relationship between words. These methodologies are considered very efficient and reliable, but the downside is that the context¹⁾ of the interview is not taken into account (GUEST, MacQUEEN & NAMEY, 2012). When using qualitative interview analysis methodologies, the context of the interview is taken more into account than in quantitative methodologies. [8]

Discourse analysis (DA) is an umbrella term for different approaches to text analysis that focus on systematic analysis of recursive cultural and discursive elements with which people make sense of phenomena (NIKANDER, 2012). GEE (2011) argues that each theory of discourse analysis develops a series of tools to analyze language in use. The use of one DA theory instead of another depends upon the nature of the data and of the study. Researchers who use DA try to capture elements of newness in the discourse so they become elements of self-reflection and reflection about the social context (CAZDEN, 2001). Researchers working with DA use a constructivist method to examine the data so that the social and cultural context emerges from the discourse. Social and cultural context is constructed through texts, so the aim of DA as a methodology is to guide the reader of the analysis to the ways in which the text constructs the social context (NIKANDER, 2012). [9]

Conversation analysis (CA) allows the researcher to investigate specifically how sequences of talk are related and how the identities of the speakers are enacted in those sequences (SILVERMAN, 2000). For CA researchers, there are some important premises: GUBRIUM and HOLSTEIN (2000, p.492) state:

"1. Interaction is sequentially organized, and talk can be analyzed in terms of the process of social interaction rather than in terms of motives or social status.

2. Talk, as a process of social interaction, is contextually oriented—it is both shaped by interaction and creates the social context of that interaction.

3. These processes are involved in all social interaction, so no interactive details are irrelevant to understanding it." [10]

Researchers working with CA analyze the moment-by-moment development of the interaction, limiting the context of the interaction to the evidence documented in the text. [11]

Thematic analysis (TA) has been used more as a tool for qualitative analysis rather than as an independent methodology. BRAUN and CLARKE (2006) argued that it should be considered as a theoretical methodology because, although it is a flexible kind of analysis, it is based on thorough theoretical principles. Researchers working with TA look for recurrent patterns in the text or themes that are analyzed and reported in order to give a detailed interpretation of the data (BOYATZIS, 1998). TA mainly involves interpretation on the part of the researchers; it acknowledges the role of the researcher in the selection and interpretation of the themes. In order to obtain reliability and rigor, researchers develop codes to represent identified themes and apply them to the data that is then checked for intercoder agreement (GUEST et al., 2012). [12]

Analysis of personal narratives is a qualitative methodology that has become popular in the last 20 years. It focuses on the life story of the individual. The background of a person can be disclosed in many situations, from personal conversations (e.g., interviews, internet accounts, and investigative accounts) to collective accounts of social and political movements (RIESSMAN, 2012). Storytelling is seen as a relational activity; thus, the way in which the data collection (such as the crafting of the interview) is done, directly influences the results of the analysis process. [13]

For narrative analysts, it is essential to be in some way a part of the context of the interview; thus, "narrative analysts assume that tellers and listeners interact in particular cultural milieus—historical contexts essential to interpretation. Narrative inquiry opens up forms of telling about experience, not simply the content to which the language refers" (p.368). In the storytelling tradition, analysts are aware of the fact that stories are not complete or perfectly focused throughout the narration, and it is their job to point out the meanings the narrators want to communicate. These meanings can be interpreted through different lenses—thematically, structurally, conversationally, culturally, psychologically, and politically/historically—or as a performance (RIESSMAN, 2012). [14]

SFL researchers see text in a functional and semantic way rather than looking at sentences as formal and syntactic objects (YOUNG, 2009). For SFL scholars, text is understood as the way through which a culture unfolds in its social context. Therefore, there is no distinction between structure and meaning in SFL theory. In fact, discourse semantics is realized by lexicogrammatical choices of the speakers or writers of a language and are enacted through phonology and graphology (EGGINS, 2005). Languages are thus organized according to specific functions or metafunctions. [15]

At the level of metafunctions, there are three sets of systems of lexicogrammatical choices (HALLIDAY & MATTHIESSEN, 2014 [1985]), and through the analysis of these systems, it is possible to obtain the different levels of meaning of a text. Table 1 lists the three metafunctions of a text with the three correspondent analysis structures and their meanings.

The first system is the thematic structure that expresses the way texts flow, their organization, what is explicit or implicit in the text, what is new and what is already known, and the cohesiveness and coherence of the text. The second system is the mood, and it expresses the interactional meaning. (Are the participants asking questions, giving commands, or making statements?) The third system is called transitivity and expresses the representational meanings (Who is doing what, to whom, where, and when?) (BAWARSHI & REIFF, 2010). [17]

The thematic system is concerned with the textual metafunction. Through this system of analysis, it is possible to assign the functions of theme and rheme (FRIES, 1997). In functional terms, all clauses are composed of a theme and a rheme (HALLIDAY & MATTHIESEN, 2014 [1985]). Furthermore, both of these elements can sometimes be implicit, particularly in spoken language, as can be seen in the interview analysis presented in Section 4. [18]

The role of theme in the English language corresponds to the initial position in the clause. The meaning of the theme of the clause is the point of departure of the message and tells us what the message is concerned with. As FRIES (1997) states, "[t]he theme of a clause provides a framework for the interpretation of the clause" (p.232). Thus, as a unit of analysis, the theme orients the listener or reader of the text to the meaning of the spoken or written language. [19]

In English, the rheme of the clause generally contains the information unit, which is oriented towards the listener or reader of the text. The new element is thus what is unpredictable in the clause. This unit of analysis is not linked to the lexicogrammar of the language but to the tone group in spoken language. As HALLIDAY and MATTHIESSEN (2014 [1985]) indicate, in the tonic foot of the clause, we can find the "new." In the information unit, what is new is dialectically related to what is known or predictable. [20]

Through thematic progression, the researcher is able to recognize patterns of meaning and to observe the organization of themes and rhemes within clauses throughout a text. Identifying the thematic progression allows the researcher to gather information about the pattern of flow of textual meaning in the interview text (THOMPSON, 2014 [1996]). [21]

The mood system and appraisal analysis are concerned with the interpersonal metafunction. The mood system allows researchers to analyze the clause as an interaction (a clause produced by the speaker or writer to interact with a listener or reader). Mood is grammatically enacted in a subject that corresponds to a nominal group and a finite verb form (HALLIDAY & MATTHIESSEN, 2014 [1985]). [22]

The mood system, then, is a tool to understand the grammatical variations within interactions. In particular, it helps researchers to identify if the clauses are declarative, exclamatory, imperative, or interrogative. The mood system also assists the researcher in detecting the polarity of the clause (whether the meaning of the clause is positive or negative), as well as the modality of the clause. Through the modality of the clause, researchers can assess the evaluation and/or orientation of the listener, in other words, the listener's judgements about the content of the clause (ibid.). To further refine this element of judgment that the mood structure reveals, IEDEMA, FEEZ and WHITE (1994) proposed the appraisal system. [23]

MARTIN and ROSE (2003) define appraisal as a system that focuses on attitude, how feelings, values, and points of view are negotiated between the speaker or writer and the listener or reader. WHITE (2001), in his appraisal outline, explores how researchers started to use appraisal as an instrument for data analysis in order "to explore in what contexts, by what linguistic means and to what rhetorical ends writers pass value judgments, attribute their propositions to outside sources or modalise their utterances" (p.3). Appraisal is a system used to analyze the use of evaluative language, specifically the attitude, the engagement, and the graduation of the language used by the speaker or writer. [24]

Attitude is developed through three different types of meaning: affect, or evaluation by means of emotion (i.e., I love pizza); judgment, or evaluation by means of ethics (i.e., liars should go to prison); and appreciation, or evaluation by means of appearance or aesthetic (i.e., a beautiful vase). Engagement is developed through the use of words or clauses that dialogically include or exclude the listener or reader from the text (i.e., the use of maybe vs. the use of certainly). Graduation is developed through force (words that can lower or raise the tone or intensity of a text) and focus (the language used to convey the preciseness of the meaning of a text). [25]

The transitivity and ergativity systems are concerned with the ideational metafunction. As HALLIDAY and MATTHIESEN (2014 [1985]) assert, the ideational metafunction regards the representation of or the linguistic order that we impose on the flow of events we experience. The grammatical system that corresponds to this metafunction is called transitivity. [26]

Through the transitivity system, researchers explore how the phenomenon unfolds in time through a verbal group that acts as the process of the clause. The types of processes and the participants to those processes are the manifestation of the speaker's or writer's grammatical and semantic choices to represent those processes. In particular, HALLIDAY and MATTHIESEN talk about six overarching processes:

The last type of analysis proposed by HALLIDAY and MATTHIESEN is called ergative analysis. Ergative analysis still regards the ideational metafunction, but in a different way than transitivity. While transitivity enables one to examine an experience as discrete units (how an experience unfolds as process,) ergative analysis allows one to examine an experience as a whole. Taking causation as its first principle, ergative analysis assists researchers in deciphering who is the agent in the text. It helps determine how the agent's voice is effective, depending on whether the process is active, middle, or passive. Through ergative analysis, the analyst can determine if the agent is present, hidden, or absent. [28]

An excerpt from the transcription of a qualitative interview is used to exemplify the application of SFL as an analytic framework. It focuses on a student's evaluation of the effectiveness of the "Schema for the Complete Orienting Bases of Action" (SCOBA) as a learning tool. The SCOBA (Appendix 1) is a goal-oriented graphic representation aimed to materialize a concept, working as a map that guides the student into the mastery of a concept and its application in solving situations (ARIEVITCH & HAENEN, 2005). The SCOBA is part of GAL'PERIN's (1992) pedagogy that organizes classroom activity into a series of mediational steps that foster the internalization of concepts. The interview follows a 6-hour formative intervention in which the didactic tool, the SCOBA, was used to teach the SFL concept of genre and register of a typified situation. [29]

To show how the SFL systems of analysis can be applied to an interview text, I analyzed one excerpt from a qualitative research interview using different systems of analysis. Specifically, I used thematic analysis, mood system and appraisal analysis, and ergative analysis. I decided to use only one excerpt to show how each analysis system added to the meaning the interviewee intended to communicate. Analysts may decide to use only one of these system, all of them, or any combination of them that make sense to answer their research questions. [30]

Data for this research is derived from a larger project assessing the use of a pedagogic tool called SCOBA to teach Italian as a Foreign Language (IFL) to four beginner adult students (FERNANDEZ & DONATO, forthcoming). The participant, an adult female English speaker, is a professor at a university in the northeastern part of the United States where the exploratory study took place. Two participants were present during the interview, the interviewer, who will be referred as T, and the interviewee, who will be referred as S1. The in-depth interview followed a semi-structured protocol (Appendix 2) based on RUBIN and RUBIN's (2012) conversational partnerships. [31]

To investigate how SFL can be applied in practice to qualitative interview analysis, two research questions were formulated:

The same data was analyzed three times to retrieve the three metafunctional levels of meaning. For the textual metafunction, I used thematic analysis. For the interpersonal metafunction, I used mood analysis integrated with appraisal analysis. For the ideational metafunction, I used transitivity and ergativity analyses. The themes and rhemes for the thematic analysis were coded as described in Table 2 below.

To analyze the student's interview excerpt, the following procedures in each system were applied:

Regarding the thematic structure, the interview excerpt presents a thematic progression with constant themes (THOMPSON, 2014 [1996]), in which the same theme (not necessarily in exactly the same words) appears in a series of utterances, and different rhemes are linked up to it. When there was another question or another comment from the interviewee, and the theme changed, the new theme became constant again throughout the clauses. [35]

The identification of the new element in the analysis was very important because, since the genre of the text is an interview, most of the clauses had unmarked themes referring to the interviewee first-person pronoun. Identifying the "new" helped to evaluate elliptical themes that were developed by pronouns or directly by the rhemes. Indeed, through the theme/new analysis, it is possible to establish a connection between the thematic choices of the interviewee and the genre's social purpose: to explore the point of departure and how the interviewee organized the information in order to represent a different perspective rather than that of the interviewer's (RUBIN & RUBIN, 2012). [36]

Thematic analysis gives the researcher a new perspective on the data because it reveals the textual choices of the interviewee. For example, in the first clause, T introduced a new theme, the difficulty of the SCOBA. S1 answered that she found it sometimes difficult. T acknowledged S1's answer with just a recast, saying "OK," and S1 completes her thoughts, adding another rheme, the fact of being novices. Table 3 shows the theme/rheme progression and how the semantic organization of the text into a message indicates the flow of the discourse. The flow of the themes is what guides the listener/reader in interpreting the text. Table 3 also reports when a theme is marked, i.e., the starting point of the clause is less expected by the listener/reader, or unmarked, i.e., the starting point of the clause is expected by the listener/reader.

Theme 1 was introduced by T and was coded as "difficulty working with the SCOBA." Rhemes 1 and 2 were coded the same way, whereas Rheme 3, which becomes Theme 3, and Rheme 4 were coded as "personal path in the SCOBA." The answer that S1 provides in clause 2 is worthy of attention because the SCOBA, which is the focus of the interview, is theoretically conceived as a tool that the students have to use in a personal way to avoid the burden of learning the concepts through rote memorization. Thus, in Clause 4, she introduced the new Theme 3 (based on the previous Rheme 3) in which S1 explained the way she had used the SCOBA: "I just follow this line ..." This is relevant to the investigation because S1 thought that she had not done anything special to learn. According to the theoretical framework of SCOBA use, S1 had done exactly what she had been expected to do: she used the tool to work in tasks rather than memorization. In fact, T, in Clause 5, explains, "well, it's because you find your path," and, in Clause 6, S1 acknowledged the teacher's affirmation, exclaiming, "right; you find your path in it." The rheme of Clause 5 that became the theme of the next three clauses was coded as "student's perceptions of the tool." It is noteworthy that it is the student who introduced the new theme. S1 changed the theme the interviewer had proposed to a more crucial one for research purposes: the student's perceptions of her use of the SCOBA. [38]

Another example of the student's use of the SCOBA can be found in Table 4, Clause 2, when S1 used a clause in which "I" is the topical theme and "feel that I finally nailed something" is the rheme. It is just an embedded clause, but the interviewer deems it important. In fact, in Clause 4, T recast the affirmation in a question: "did you have this impression that you nailed into something?" S1 answered in an unmarked theme ("maybe I didn't know permanently"), and, then, she added a marked theme to let the interviewer understand that she had really felt like that, introducing a "when" circumstance.

The examples in Table 4 are relevant as they show how the discourse had been co-constructed between the interviewer and the interviewee. Even in very simple and short oral answers, there is more meaning than it appears at first glance. [40]

In the case of the above excerpts, the progression develops through constant themes that were coded as "perceptions of the tool." There is an expansion of the themes through the rhemes that gives the listener or reader more information about the interviewee's perceptions. The structure of the interview allowed an exploration of themes that were introduced by the interviewee, not just by the interviewer. This thematic investigation provides insight in a new perspective of the driver of the conversation. The analyst can just use thematic analysis to conduct their investigation or can add another layer of meaning through another system such as the mood and appraisal system. [41]

The mood and appraisal analyses are particularly revealing for this kind of interview because most of the interview text chosen develops interpersonal meanings. The majority of the clauses are declarative and interrogative, as it is expected in an interview, but there are many instances of bound clauses. Bound clauses are clauses dependent on other clauses, i.e., embedded in them (HALLIDAY & MATTHIESSEN, 2014 [1985]). They are important as discourse markers or as markers of interpersonal contact. This means that the interviewer and interviewee were trying to connect to each other and that this connection was, in most occasions, achieved. I analyzed the moves and the type of information exchanged to find in the language use patterns that helped to identify the social roles. For example, most of the time, T initiated the turns, and S1 responded, but there was an instance (Clause 1 in Table 5), in which S1 asked T for a judgment about her comprehension of the language.³⁾

In Table 5, after T answered in Clause 2 with a bound clause, trying to engage S1, S1 continued for another four clauses to integrate her discourse. In fact, in clause 3, she focused the topic ("at least") and expressed judgments ("to be able to pick up some words") and appreciation ("that's been nice"). As can be observed in this example, S1 dominated the exchange, her predominance is a pattern in the text as a whole. In fact, the participation structure is constructed through a pattern in which T initiated the discourse, yet S1 dominated the discourse. In the 96 clauses present in the interview (Appendix 2), S1 produces 48 clauses and T 48, many of which are just bonding expressions to encourage S1 to continue her discourse. [43]

The polarity of the clauses is another type of meaning that can be retrieved through mood analysis. Most of the negative polarity present in the excerpt regards two particular topics: the interviewee's previous experience learning foreign languages and the limited time that the interviewee had dedicated to studying Italian. Using the appraisal analysis to complement the mood analysis, it is possible to realize a more in-depth analysis of the data. For example, if appraisal analysis is added to the negative polarity found when S1 was talking about her previous experience of learning Spanish, the researcher can have a stronger idea of how S1 felt in that situation and how it negatively influenced her first approach to the Italian language.

In Table 6 above, S1 talks using force to underline her ideas. For example, in Clause 4, she expressed the judgment "[no] gained proficiency," using the word "never" that raises a negative judgment to a peremptory state. Then, in Clauses 5 to 7, she declared she had lived in Florida for ten years to give more force to the discourse, and, with an embedded clause, she gave further force and judgment to her discourse to conclude the clause complex with an affective declaration in Clause 8: "so I was concerned." [45]

Force underlaid the whole interview. The interviewee had a strongly negative previous experience learning foreign languages, particularly Spanish. She never judged herself as a proficient Spanish speaker. This previous experience made her express some concerns regarding the new language she was about to learn. The same force is used to indicate the positive experience she had while learning Italian. In Table 7, there is the repetition of a pattern of an attitude that mixes elements of affect and judgment with a positive polarity, revealing how S1, who in the turns before had expressed negative judgments about her performances in foreign language, had a positive affective reaction toward the new learning process. The same pattern occurred while talking about the SCOBA. S1 used force, repeating words, using graduation and other resources, to indicate her positive experience with the tool. Even though she judged her proficiency positively, she still thought she was not self-regulated, that she still needed the SCOBA to operate with affective certainty (Clauses 8-9).

To summarize, mood and appraisal analysis were applied to understand the interpersonal meanings in the text. Mood analysis revealed that the clauses analyzed in the interview were mostly declarative and interrogative, as should be expected in an interview. The presence of bound clauses in this interview demonstrates how a personal connection was made between the interviewer and the interviewee this connection allowed for more open and free interaction. Regarding the polarity of the clauses, instances of negative polarity can be found in two particular topics: the interviewee's previous experience learning foreign languages and the limited time that the interviewee had to study Italian. Using Appraisal analysis to complement the findings from mood helped to show a change to a positive perception of S1's possibility to use Italian supported by the SCOBA. S1's use of force, judgement, and affect reveals how the negative experience she had learning Spanish and the lack of time to study Italian did not hinder S1's positive perception of her Italian language use. Through mood and appraisal analysis, I was able to understand the interviewee's feelings about this new method of language learning. I could stop my analysis here or add another layer of meaning through another system such as the transitivity and ergativity systems. [47]

While analyzing the patterns of choices of S1, I noticed that T asked the student mostly about mental and relational processes (i.e. verbs): student's point of view, her perceptions. S1 responded using mental processes and then confirmed her position in terms of material processes to re-elaborate in terms of relational processes, as can be noticed in the following excerpt from the interview (Appendix 2, Turns 5-6)

"T: Ok, and tell me what was your first impression when you started to learn Italian? (was=relational, started to learn=mental)

S1: I was intimidated because I never gained proficiency in Spanish and I lived in Florida for 10 years where a lot of people speak Spanish so I kind of, I was concerned that I was not be able to do it, so that was my first, ah, perspective (was intimidated=mental, gain proficiency=material, lived=material, speak=material-verbal, was concerned=mental, was able to do=material, was=relational)." [48]

In my analysis S1 first answered opening to the relation with T then explained her mental state giving concrete material examples to return to her opinion opening again to T through a relational process. This pattern gave T an opportunity to abandon the semi-structured protocol to explore some of the processes the student introduced in the interview, e.g., in Turn 10 ("S1: yes, I did very much so, hum I really enjoy it very very much and I mm, I think that we did a lot considering what little time we had to do that") the S1 started to talk about the time constraints for the formation and her inability to study more during that period which will be the topic of the next 4 turns in the interview. [49]

Another point of interest is found in Turns 10 to 26 where S1 tried to explain what she liked most about studying Italian. She used mental processes like "love" and "like" to describe a phenomenon that is metaphorically mental but very material in origins; in Turn 26 she said, "I feel I nailed something." She also signaled this with two temporal circumstances ("when" and "finally") that gave strength to the concreteness of the phenomenon because it is signaling a specific moment in time. Similarly, at the beginning of the text, when she talked about her experience learning Spanish, she used mental processes to describe the concrete phenomenon of not having acquired proficiency in the language. Even in Turn 6 cited above, S1 used spatial "I lived in Florida" and temporal "for 10 years" and to concretize her mental processes. [50]

When asking about the SCOBA, which is the main topic of the educational research, T used a material verb "work" (Appendix 2, Turn 35-36)

"T: yes yes you are picking a lot of language, you are picking a lot of language probably more than what you perceive, but I see in your face when I'm speaking Italian you understand what I am talking about, and how about working with the SCOBA, with the chart?

S1: I think the SCOBAs are helpful, they're helpful they give me voice and they are good scaffold for conversation, ahm." [51]

S1 responds at first using a mental process, "I think," to describe the use of the SCOBA as a phenomenon. Then, she used a material process to express the result of the phenomenon "they give me voice," before concluding with two relational processes, "they are helpful" and "they are a good scaffold," indicating some of the attributes of the SCOBA. [52]

Summarizing, the transitivity analysis of the processes reveals that the interviewer asked the student mostly about mental and relational processes: her point of view, her perceptions. S1responded using mental processes and re-elaborated in terms of material processes. This is worth mentioning because S1 demonstrated her awareness of her learning process, and she explained it in concrete terms. This allowed T to abandon the semi-structured protocol of the interview and follow the thread of S1. Moreover, when S1 answered regarding the SCOBA, she used relational and material processes to explain how she perceived the SCOBA as a phenomenon that had produced concrete results in terms of language learning. [53]

Looking at the interview from an ergative point of view, it can be noticed how S1, during most of the interview, uses an "I" medium, meaning an "I" that actualizes the process. The use of "I" demonstrates that she had been the one who had brought the processes into existence. S1 represented herself with agency throughout the whole interview. She was also the subject of the interview, and she was aware of it, which is important to note because this phenomenon signals that T was focused on S1's answers to conduct the interview. [54]

Another relevant aspect of this interview is the concreteness of the answers of S1. She always answered with details about the circumstances of the processes, particularly temporal and locational. It was not just a circumstance but a circumstance in a particular time and place. This makes the interviewee dependable and gives credibility and thoroughness to the research. [55]

Through its rich range of analytical tools SFL can be used to gather insight into interviewees' perspectives in qualitative interviews. Particularly, SFL helps the researcher to get a fine-grained view of how interviewees construe their experiences (first research question) and how SFL can be a tool to assess FL learners' awareness and control of their own language development (second research question). [56]

To answer research question one about how SFL metafunctions usefulness to enrich the analysis of interview data, the results of this research demonstrate that the different levels of analysis gave a fine-grained view of how meanings were construed by the interviewee. Through the metafunctional analysis of the interviewee's linguistic choices, it was possible to grasp how she had put herself at the center of her learning process. Moreover, it was possible to observe how she had experienced her language learning in a concrete and positive way different from her past experience. Finally, it was possible to assess the importance of the SCOBA, which was the topic of the interview, as an educational tool. The different levels of analysis demonstrated how S1's answers are not only dense in content but are fundamental to the understanding of the interview and in extent to the understanding of the research's outcome as a whole. The whole text of the interview, that at first glance might appear simple, gave the teacher/researcher the means to assess the student's awareness of her own learning process. [57]

The fact that SFL analysis allows researchers to get in-depth information about the interviewee is relevant to answering the second research question I posed: How can SFL be used as a tool for analysis in qualitative interviewing in foreign language education research? SFL can be used to evaluate foreign language learners' language awareness and control (e.g., students' agency in their language use) and to develop language awareness by pointing out problems that interviewees are not aware yet (e.g., the negative influence of previous experience learning Spanish to her perception of learning Italian). [58]

After analyzing the interview through the various SFL systems of meaning-making in English, different levels of meaning were obtained: textual, interpersonal, and ideational. These meanings helped the researcher arrive at interesting findings that can be useful to future research on interview analysis. For example, at the textual level, the analysis revealed that the interviewee had proposed many themes in the interview because she had wanted to express how important learning Italian was for her. At the interpersonal level, the interview text revealed the student's positive appreciation of the instructional tool, the SCOBA. At the ideational level, the student stated in a detailed way how important learning Italian with the SCOBA was for her. [59]

For the above reasons, it may be concluded that SFL is a versatile and flexible methodology for analysis that facilitates qualitative research by highlighting social meanings that are mediated through the linguistic choices made in interviews. Likewise, SFL analysis of qualitative interviews can offer a window to explore foreign language learners' language awareness and control and to act as a diagnostic tool for revealing language learning issues. [60]

Nonetheless, the use of SFL has some downsides. First, the analysis of the data requires a great investment of time depending on the level of depth the researcher wants to reach. Second, the researcher who wants to work with this framework needs to know very specific terminology. SFL terminology sometimes assigns new functional meaning to familiar terminology utilized in traditional English grammar; in fact, SFL terms are written with capital letters to distinguish them from traditional usage. Not knowing the terminology could lead the inexperienced analyst to misinterpretations. [61]

Despite the above-mentioned limitations, this article demonstrates how SFL can be a useful tool for analyzing qualitative interview texts. The purpose of this article was to highlight SFL as a framework of analysis for qualitative interviews in foreign language education. This analytic framework has the potential to reveal different levels of meaning through the linguistic choices of the interviewee. Further research is needed to better understand applications of this methodology for other types of interview (e.g., online interviews and forums) and in other fields of the social sciences (e.g., psychology, sociology, and political science). [62]

I would like to thank S1 for participating to the study. I would also like to thank Dr. Mariana ACHUGAR for sharing with me fundamental insights into Systemic Functional Linguistics. Finally, I would like to thank the blind reviewers of this article for their precious and precise feedback.

1) Each methodology gives a different definition of the word "context" (context of the utterance, context of the interview, social and cultural context of the interactors, etc.), but word counting or semantic networks does not consider any type of context. <back>

2) Since the interview is analyzed using different excerpts, for each excerpt, I start the theme and rheme progression from 1 in order to facilitate the readability of the analysis. <back>

3) Below "XX" (substituting the name of a person) and in the clause below "XXX" (substituting the name of a city) were used to guarantee the anonymity of the interviewees. <back>

Achugar, Mariana & Pessoa, Silvia (2009). Power and place. Language attitudes towards Spanish in a bilingual academic community in southwest Texas. Spanish in Context, 6(2), 199-223.

Arievitch, Igor M. & Haenen, Jacques P.P. (2005). Connecting sociocultural theory and educational practices: Gal'perin's approach. Educational Psychologist, 40(3), 155-165.

Bawarshi, Anis S. & Reiff, Mary Jo (2010). Genre: An introduction to history, theory, research, and pedagogy. West Lafayette, IN: Parlor Press.

Boyatzis, Richard E. (1998). Transforming qualitative information. Thousand Oaks, CA: Sage.

Braun, Virginia & Clarke, Victoria (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101.

Caldwell, David (2009). Working your words, Appraisal in the AFL post-match interview. Australian Review of Applied Linguistics, 32(2), 13.1-13.17

Cazden, Courtney B. (2001). Classroom discourse: The language of teaching and learning. Portsmouth, NH: Heinemann.

Deppermann, Arnulf (2013). Interviews as text vs. interviews as social interaction. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 14(3), Art. 13, http://dx.doi.org/10.17169/fqs-14.3.2064 [Accessed: February 24, 2018].

Eakin, Joan M. & Mykhalovskiy, Eric (2002). Reframing the evaluation of qualitative health research: Reflections on a review of appraisal guidelines in the health sciences. Journal of Evaluation in Clinical Practice, 9(2), 187-194.

Eggins, Suzanne (2005). An introduction to systemic functional linguistics. New York, NY: Continuum.

Fernandez, Loretta & Donato, Richard (forthcoming) Interacting with SCOBAs: Beginner learners of foreign language use of a pedagogical tool. Language and Sociocultural Theory.

Fries, Peter H. (1997). Theme and new in written English. In Tom Miller (Ed.), Functional approaches to written text: Classroom applications (pp.230-244). Washington, DC: Eric.

Gal'perin, Piotr Ia (1992). Stage-by-stage formation as a method of psychological investigation. Journal of Russian and East European Psychology, 30(4), 60-80.

Gee, James P. (2011). How to do discourse analysis a toolkit. New York, NY: Routledge.

Gibbons, Pauline (2006). Changing the rules, changing the game: A sociocultural perspective on second language learning in the classroom. In Geoff Williams & Annabelle Lukin (Eds.), The development of language: Functional perspectives on species and individuals (pp.196-216) New York, NY: Continuum.

Gubrium, Jaber F. & Holstein, James A. (2000). Analyzing interpretive practice. In Norman K. Denzin & Yvonna S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp.487-508), Thousand Oaks, CA: Sage.

Guest, Greg; MacQueen, Kathleen M. & Namey, Emily E. (2012). Applied thematic analysis. Thousand Oaks, Ca: Sage.

Halliday, Michael A.K., & Matthiessen, Christian M.I.M. (2014 [1985]). Halliday's introduction to functional grammar (4th ed.). New York, NY: Routledge.

Iedema, Rick; Feez, Susan & White. Peter R.R. (1994). Media literacy. Sydney: Metropolitan East disadvantaged schools program, NSW Department of School Education.

Kress, Gunther & van Leeuwen, Theo (2001). Multimodal discourse: The modes and media of contemporary communication. Oxford: Oxford University Press.

Lampropoulou, Sofia & Myers, Greg (2013) Stance-taking in interviews from the Qualidata archive. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 14(1), Art. 12, http://dx.doi.org/10.17169/fqs-14.1.1813 [Accessed: February 24, 2018].

Martin, Jim R. & Rose, David (2003). Working with discourse. New York, NY: Continuum.

Nikander, Pirjo (2012). Interviews as discourse data. In Jaber F. Gubrium, James A. Holstein, Amir B. Marvasti & Karyn D. McKinney (Eds.), The Sage handbook of interview research. The complexity of the craft (2nd ed., pp.397-413). Thousand Oaks, CA: Sage.

Riessman, Catherine K. (2012). Analysis of personal narratives. In Jaber F. Gubrium, James A. Holstein; Amir B. Marvasti & Karyn D. McKinney (Eds.), The Sage handbook of interview research. The complexity of the craft (2nd ed., pp.367-379). Thousand Oaks, CA: Sage.

Roulston, Kathryn (2014). Analyzing interviews. In Uwe Flick (Ed.), The Sage handbook of qualitative data analysis (pp.297-326). Thousand Oaks, CA: Sage.

Rubin, Herbert J. & Rubin Irene S. (2012). Qualitative interviewing. The art of hearing data. Thousand Oaks, CA: Sage.

Ryan, Gery W. & Bernard, H. Russell (2000). Data management and analysis methods. In Norman K. Denzin & Yvonna S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp.769-801). Thousand Oaks, CA: Sage.

Seidman, Irving (2013). Interviewing as qualitative research. New York, NY: Teachers College Press.

Silverman, David (2000). Analyzing talk and text. In Norman K. Denzin & Yvonna S. Lincoln (Eds.), Handbook of qualitative research (2nd ed., pp.821-834). Thousand Oaks, CA: Sage.

Thompson, Geoff (2014 [1996]). Introducing functional grammar (3rd ed.) New York, NY: Routledge.

Young, Richard F. (2009). Discursive practice in language learning teaching. Chichester: Wiley- Blackwell.

Loretta FERNANDEZ, Ph.D. works as faculty at Duquesne University, Pittsburgh teaching Italian as a foreign language. Her current research focuses on foreign language pedagogy, sociocultural theory, systemic functional linguistics and qualitative research methods.

Duquesne University
Department of Modern Languages
Room 204 Fisher Hall, 600 Forbes Ave.
Pittsburgh, PA 15282, USA

Fernandez, Loretta (2018). Qualitative Interview Analysis: The Use of Systemic Functional Linguistics to Reveal Functional Meanings [62 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 19(2), Art. 6, http://dx.doi.org/10.17169/fqs-19.2.2663.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License

Analysis Systems	Metafunction	Levels of Meaning
Thematic structure	Textual	Text flow
Mood	Interpersonal	Interactional meaning
Transitivity	Ideational	Representational meaning

Codes	Description of the Codes
Difficulty working with the SCOBA	Student mentions issues during instruction
Personal path in the SCOBA	Student mentions conceptual orientation
Student perception of the SCOBA	Student expresses feelings about the SCOBA
Student perceptions of culture	Student mentions culture related to language
Student use of the SCOBA	Student talks about the SCOBA use during instruction

Clause	Speaker	Excerpt	Type of Theme	Progression
1	T	and did you find difficult to follow the SCOBAs?	Unmarked theme	Theme 1, Rheme 1
2	S1	Sometimes	Elided theme "I found it difficult"	Rheme 2
3	T	Ok
		sometimes I think it's	Unmarked theme	Theme 2, Rheme 3
4	S1	because we are such novices
		like sometimes I just kind of hey
		I just follow this line let's go by, instead of really making choices	Unmarked theme	Theme 3, Rheme 4
5	T	well it's because you find your path
6	S1	right you find your path in it	Marked theme	Theme 3, Rheme 4

Clause	Speaker	Excerpt	Type of Theme	Progression
1	S1	and I love	Unmarked theme	Theme 1, Rheme 1
2	S1	when I feel that I finally nailed something	Marked theme	Theme 2, Rheme 2
3	S1	like when we were practicing different aspects of the field	Embedded clause
4	T	did you have this impression that you nailed into something?	Unmarked theme	Theme 1, Rheme 2
5	S1	maybe I didn't know permanently but	Unmarked theme	Theme 1, Rheme 3
6	S1	When we were working on the alphabetic stuff and counting stuff,	Marked theme	Theme 2, Rheme 4
7	S1	it it just feels nice	Marked theme	Theme 1, Rheme 4
8	S1	to be able to do something ...	Embedded clause

Clause	Speaker	Excerpt	Mood Subject Finite	Type of Clause	Appraisal
1	S1	you feel like our comprehension is increased greatly	you feel	Bound	Affect, judgment, force
2	T	oh yeah		Bound	Engagement
3	S1	so that's been nice to be able to at least pick out words	that's	Declarative	Appreciation, judgment, focus
4	S1	like when you were talking to XX	you were	Declarative	Engagement
5	S1	and you wanted to know where was she from	you wanted	Declarative	Engagement
6	S1	I know she's from XXX	I know	Declarative
7	T	yes yes you are picking a lot of language	You are	Bound	Affect, judgment, force