Validating a Qualitative Interview Schedule on Multilingualism and Second Language Vocabulary Attrition and Maintenance

This paper focuses on the validation process of a qualitative interview schedule designed to investigate the nuanced dynamics of multilingualism and second language vocabulary attrition and maintenance. The interview schedule was developed to gain insights into the complexities of the participants’ language experiences, identities, as well as vocabulary learning and retention strategies. Second language speakers of English with diverse linguistic backgrounds were selected for the study to ensure the reliability and validity of the interview. The piloting stage played a pivotal role in laying the ground for refined interview questions, where data authenticity was ensured through the creation of a comfortable environment that helped participants provide genuine responses and avoid offering socially desirable ones. The interview questions were piloted multiple times to identify and resolve any inconsistencies in the participants’ responses. The validated interview schedule can serve as a dependable data collection tool and prompt researchers to consider the implications of second language vocabulary attrition and maintenance for pedagogical practices.


N
umerous theories have been put forward in the fields of language acquisition and language attrition to decipher and understand the mechanisms of language development and language deterioration, respectively.Specifically, second language vocabulary acquisition and maintenance and second language vocabulary attrition, which are the focus of the study, play an essential role not only in the formation of solid language faculties for the L2 learners, but also in the establishment of proper pedagogical materials which are vital to fulfill their needs.Observing the phenomena of second language vocabulary acquisition and attrition, one becomes naturally inclined to contemplate and endeavor to explain the correlation between them.While a substantial body of research has been conducted in these fields (Olshtain 1989;Schmid 2006;Wei 2014;Schmid 2022;Ding 2021), they have received little attention in the Moroccan and Hungarian research domains.Thereafter, this study aims to depict the validation process of an interview schedule on multilingualism in relation to second language vocabulary attrition and maintenance in the Moroccan and Hungarian contexts.This is one of the data collection tools of a larger study that aims to shed light on the various factors leading to second language vocabulary attrition, the applicability of theories of acquisition and attrition in the above-mentioned contexts, the research participants' practices to promote second language vocabulary maintenance, and the implications and impacts of these findings on language pedagogy.
Of the numerous lexical and grammatical areas that can be studied in second language acquisition and attrition, only vocabulary will be given central focus in the larger study.This is because the lexicon is said to be the first linguistic area that is commonly and rapidly affected by attrition, even more than grammar (Kuhberg 1992).Furthermore, it was concluded that production skills (i.e.speaking and writing) are at a higher risk of attrition than receptive ones (i.e.listening and reading).Albeit vocabulary is not considered a skill, productive vocabulary knowledge is still at a higher risk of attrition than the receptive one (Hedgcock 1991).Productive and receptive vocabulary knowledge, also known as active and passive knowledge, can be defined from a pedagogical perspective as the ability to recall and correctly use a word in a written text or speech, and the ability to understand a word in its spoken or written form, respectively (Pignot-Shahov 2012).Furthermore, vocabulary attrition is claimed to result from a lack or reduction of access, meaning the inability to recall a word either due to memory decay or interference from other learning (Cohen 1989).Accordingly, the attention is directed to the measurement of both productive and receptive vocabulary attrition and maintenance.Three factors will be investigated to detect attrition: factors of language knowledge and use, individual factors, and factors of input (Bardovi-Harlig & Stringer 2010).To measure the factors of language knowledge and use, change in the vocabulary, accuracy, and fluency will be quantitatively examined to account for any language production issues that could relate to attrition.Concerning the change in vocabulary, the number of retained or lost words per task, and the total number of retained or lost words shall be inspected.For accuracy, the number of errors per response, and the number of error-free responses will be analysed.As for fluency, the number of recalled words per task, filled and unfilled pauses, false starts, repairs, the number of pauses between utterances, length of unfilled pauses, elapsed time between question and response, and length of speaking time are going to be examined.
Although these quantitative measures are needed to delineate and assess the production of and access to vocabulary items, they are not indicative of the reasons for their attrition or maintenance.To unveil these reasons, a qualitative measurement of individual factors will be put forward.This incorporates age, gender, motivation, strategic competence (i.e.verbal and non-verbal strategies to compensate for breakdowns in communication: paraphrase, repetition, hesitation, guessing, etc.), the multilingual background of the participants comprising the origin (nature), function (proficiency and use), competence, and identification with the L2 (Bardovi-Harlig & Stringer 2010).
Another qualitative measure concerns the factors of input.This includes the duration and nature of the extensive program of instruction, the duration and nature of the instruction during the disuse period, the duration and nature of the reduced input and use.This is closely related to the concept of "learning situation level" of motivation (Dörnyei et al. 1994), which can be divided into three components: "Course-Specific Motivational Interest (relevance, expectancy, satisfaction), Teacher-Specific Motivational Affiliative Drive (authority type, modelling, task presentation, feedback), and Group-Specific Motivational Goal-Orientedness (norm & reward system, group cohesion, classroom goal structure).The study by Dörnyei et al. (1994) serves as groundwork for understanding the impact of the learning situation on motivation and how instructional practices and classroom dynamics impact learners' motivation.
For the purpose of this qualitative paper, only one individual factor will be studied: the multilingual background of the target population.This is assumed to be an important indicator of the factors leading to second language vocabulary attrition or maintenance.To test this hypothesis, the data collection tool deemed to be the most effective and suitable is the in-depth interview.This is because "the depth of the conversation, which moves beyond surface talk to a rich discussion of thoughts and feelings" (Maykut & Morehouse 1994, 76) gives a much greater opportunity to gain full access to the interviewees' multilingual profiles and backgrounds.More so, because the definition of a multilingual person depends heavily on the person's perception of multilingualism and self-identification not only with regard to the languages spoken, but also to the identities they bring along.
To be able to form a comprehensive image of the multilingual profiles of the interviewees, several open-ended questions are needed to elicit answers regarding the various dimensions involved in multilingualism, including the origin (nature), function (proficiency and use), competence, and identification with the spoken languages, all of which are tackled in detail below.Accordingly, the main research focus is on Moroccan and Hungarian students' experience of English language vocabulary attrition and maintenance in light of their multilingual profiles.

Literature review
This section of the paper examines the existing literature to explore the conceptualizations and definitions of multilingualism in order to shed light on nuances associated with this dynamic linguistic phenomenon, and to develop a theoretical framework for scripting the qualitative interview schedule.
There are two distinct, but not completely separate dimensions of multilingualism; the first one is individual, relating to a single person's multilingual ability, the second one is societal, relating to a society's overall multilingual state (Cenoz 2013).For the interest of this paper, some societal aspects will be taken into account as they are part and parcel of a person, but the focus will mainly fall on the individual facet of multilingualism since the studied phenomenon of language attrition requires an indepth qualitative investigation on the individual level.
An important distinction to be made when tackling multilingual matters is that of the "multilingualism" versus "bilingualism" dichotomy.Whilst there is no universal agreement on the exact difference between the two terms, different scholars adopt various positions.For instance, Cenoz (2013, 5) concludes that some traditionally use "bilingualism" as a "generic term" to refer to research involving two rather than multiple languages, but with the possibility of including more than two (Cook & Bassetti 2011).Some follow a mainstream position and use "multilingualism" as a "generic term" to refer to two or more languages, with bilingualism or trilingualism being examples of multilingualism (Aronin & Singleton 2008).Yet others use bilingualism and multilingualism as two distinct terms where "bilingualism" refers to the use of two languages, and "multilingualism" refers to the use of three or more languages (de Groot 2011).Moreover, while in sociolinguistics, bilingualism and multilingualism are generally seen to synonymously denote more than one language, especially when tackling the societal level, in psycholinguistics the exact number of spoken languages needs to be identified, especially when tackling individual matters such as language acquisition and language loss.For the purpose of this study, bilingualism and multilingualism will be used as differentiable terms with "bilingualism" referring to the exclusive knowledge/use of two languages, while "trilingualism" denoting the knowledge/use of three, and "multilingualism" the knowledge/use of more than three languages.This is because the exact number and nature of the spoken languages of the research participants are of key significance in this study.
A four-way definition of bilingualism is proposed by Skutnabb-Kangas (1990, 11) to shape the given conception of the term.This includes "origin, competence, function, and identification".Origin relates to whether an individual was born into a bilingual situation, and how the languages are accordingly used.As for competence and function, the former revolves around the individual speaking more than one language at a certain level, and the latter around the frequency of use of the languages.Identification concerns whether an individual internally identifies themselves as bilingual and part of the two languages' culture, and whether they are externally, by other members of the society, identified as native speakers of the two languages.
Furthermore, two noteworthy dimensions, "proficiency" and "use", involved in the definition of bilingualism are pointed out by Bassetti & Cook (2011, 1) who conclude that the scholarly definitions take two directions.One group of definitions "consists of a maximal assumption where being bilingual means speaking two languages with equal fluency in every situation".The other group "takes the minimal view that bilingualism refers to any real-life use of more than one language at whatever level".Taking this line, Bloomfield (1933) for instance states that "nativelike control of two languages" (56) is a necessity, while Weinreich (1953) asserts that it is "the practice of alternately using two languages" (1) that is most important, and Haugen (1953) claims that "the point where a speaker can first produce complete meaningful utterances in the other language" (7) is where bilingualism begins.De Bruin (2019) adds that bilinguals can show dissimilarities in various aspects such as "age of acquisition, language proficiency, use, and switching practices in daily life" (200), and that even when two bilinguals have attained native-like mastery in both languages from an early stage, they can still show significant differences in their actual language use.Hoffman (1991) takes the view that both groups of definitions are flawed, for, on the one hand, it is wrongful to presume that a bilingual individual's competence ought to be equal to that of two monolingual individuals, and on the other hand, it is not sensible to measure the "use" of a language of bilinguals in comparison to that of monolinguals, especially that factors such as "codeswitching, translanguaging, and translation" are only specific to bilinguals (23).Bassetti & Cook (2011) proceed to raise additional issues with these definitions, namely the fact that language skills(i.e.listening, speaking, reading, and writing) are not given due emphasis (2).In fact, a speaker's productive (speaking and writing) and receptive (reading and listening) skills in a second language are not necessarily concurrently proficient.This means that some bilingual speakers may have a solid receptive grasp of one language, but still not be able to fluently produce it.
Similarly, in the words of Wardhaugh ( 2015), "most people who are multilingual do not necessarily have exactly the same abilities in all the languages (or varieties) they speak; in fact, that kind of parity may be exceptional" and "the level of competence in a code is, of course, developed based on the need of the speaker to use a language in a particular domain or for a particular activity" (84).Whilst this interestingly suggests that the development of competence in a language depends heavily on the need for its use, an even more interesting new feature that appears in this claim is the term "varieties".One may naturally wonder if individuals speaking two language varieties or more are considered multilingual.
"Bidialectalism" is the term coined to refer to this phenomenon.Waleed & Mubarak (2019) summarize the existing scholarly definitions of bidialectalism moving from Chambers and Trudgill (1998) defining it as "speaking a dialect in addition to a standard language", through Crystal (2004) describing it as "the use of two distinct dialects (of the same language) for different social purposes" to Crystal (2008) extending it to a "speaker's ability to use two or more dialects, and to know how to code-switch appropriately between these different varieties" (23).Bilingualism and bidialectalism are distinct complex fields, yet it cannot be denied that both play a major role in affecting a multilingual individual's linguistic and sociolinguistic state.For this reason, this study will not discriminate between standardized and non-standard varieties.The full linguistic background of the target group shall be accounted for along with their language varieties based on their language use in and out of the EFL/ESL setting in order to form a comprehensive image of the linguistic background of the participants.
Coming back to the dimensions involved in defining bilingualism, de Groot (2011) notices that some scholars classify bilinguals based on the "relative competence in both languages", in that there are some who are "balanced bilinguals who possess similar degrees of proficiency in both languages" and others who are "dominant (unbalanced) bilinguals […] with a higher level of proficiency in one language than in the other" (4).Also, this dimension is directly linked to and varies in accordance with the context; specifically, how much (exposition), where (natural or formal setting), and when (age) the languages are acquired.
Speaking of the context, Cenoz (2013) explains that bilingual individuals may acquire the languages either "simultaneously or successively by being exposed to two or more languages from birth, or successively by being exposed to second or additional languages later in life" (5).The former case concerns individuals who are labeled "early bilinguals".This means that their acquisition of the mother tongue and the second language happens either at the same time (simultaneously) or one before the other (successively) during childhood.The latter case also concerns "late bilinguals" falling into the categories of "adolescent bilinguals" and "adult bilinguals" acquiring the additional languages in different stages of life starting from puberty.As was mentioned previously, age is generally an essential factor in the study of multilingualism.
In a similar vein, de Groot (2011) mentions "compound bilingualism" that occurs in a natural context, for example at home, in which the two languages are spoken interchangeably.This type of bilingualism is differentiated from "coordinate bilingualism" that emerges under a firm separation of the domains where the two languages are used, for example, either at home or at school and in public places respectively (5).
Finally, another crucial aspect of bilingualism relates to the social status of the languages spoken by a bilingual individual.Along these lines, de Groot (2011) also differentiates between "additive and subtractive bilinguals".The former blooms when both the native and the second languages have a high social value and are both used frequently, and the latter emerges when one of the acquired languages, mostly native, is looked down on and devalued socially, discouraged to use, and forced to disappear (5).This is directly linked to the social aspect of multilingualism and brings into light phenomena such as "language shift", which is concerned with groups or communities shifting to the explicit use of one dominant language, "language maintenance", which is related to the continuous use of two languages, and "ethnolinguistic validity" which is associated with the likelihood of the maintenance of a language (Wardhaugh 2015, 83).For the purpose of this paper, this aspect is especially significant, for it has a direct influence on the overall cognitive competences of the bilingual individuals, which then influences their language use.This is relevant here because the current study aims to investigate whether some of the multilingual participants identify with these types of bilingualism.
To sum up, as each individual and each purpose needs a relative definition, it remains essential to be aware of and able to detect the diverse and unique background and features of bilingual individuals before starting an interview in order to be able to establish a correlation between dimensions of multilingualism and aspects of second language vocabulary attrition.

Validating the interview schedule
As it is arduous to grasp the emotions, cognition, and behaviors that occurred at some point in past time, the organization of the world, and the attachment of meaning to the world from different individuals by mere observation, one may naturally resort to tools that ensure access to these complex yet significant elements (Patton 2002).One of these tools deemed to be effective is the qualitative interview, for one major purpose of interviewing is to gain access to the other person's perspective (Patton 2002).Qualitative interviewing establishes a predetermination whereby the other's perspective is seen as meaningful and prone to be made explicit by means of posing relevant questions.In light of the complex nature of the phenomenon of multilingualism, the wideness of its scope, and its altering definitions from one individual to another, choosing the qualitative interview has the purpose of gaining a deep insight into the interviewees' perceptions on multilingualism, their language history, experiences, thoughts, and even feelings, all of which will contribute to framing their multilingual profiles in regard to the origin (nature), function (proficiency and use), competence, and identification, of the spoken languages and their possible implications on second language vocabulary attrition or maintenance.
The approach to an interview is as crucial as its selection as the main data collection tool.Maykut & Morehouse (1994) summarize the different descriptions given to interviews in the form of a range of formats moving from "structured" to "unstructured" (76).The actual structure of the formats is said to depend on the degrees of the development of the interview questions prior to the actual interview time, which leads to a categorization of three formats: "the unstructured interview, the interview guide, and the interview schedule" (76).
On the one hand, the unstructured interview, or in Patton's (2002, 342) words the "informal conversational interview" is most fit for emergent studies which require a maximal suppleness in gathering data and looking for information in whatever course of action that seems fitting.With the focus of inquiry clearly set in mind, the questions are only posed and formed while the interview is being conducted (Maykut & Morehouse 1994, 78).
On the other hand, the interview guide and interview schedule fall into the structured category where the former is made of a sequence of general inquiries that provide the researcher with the freedom to delve into various topics with the interviewees, and the latter consists of a comprehensive collection of questions and prompts (Maykut & Morehouse 1994, 78).
While the interview guide is prepared beforehand to guarantee that basic inquiry lines are equally tracked with each interviewee, but with the possibility of introducing new questions onsite, the interview schedule, which follows the standardized openended interview approach, is a well-prepared set of detailed open-ended questions which acts as the same stimulus for all the participants (Patton 2002).Although all these formats differ in terms of structure, they all share one crucial feature: the presence of open-ended questions which aim at revealing what needs to be known about a studied phenomenon (Maykut & Morehouse 1994).
Relating these formats to the study of the multilingual profiles of the research participants, it appears that using an interview schedule prepared in advance for an in-depth structured interview is the most suitable for examining the aforementioned foci of inquiry: origin, identification, competence, and function.This choice is relevant because the information sought involves events about which little is known to the interviewer, because it is crucial not to miss out on any information related to these specific foci, and because it is important to ensure that the same set of questions are similarly posed to all the participants, so that the gathered data is consistent.

The setting
The participants were selected from the following two universities after receiving authorization and consent to conduct the study: Faculty of Humanities and Social Sciences, Pázmany Péter Catholic University, Budapest, Hungary.Faculty of Letters and Human Sciences, Sidi Mohamed Ben Abdelah University, Fes, Morocco.
Interviews were carried out online using Zoom to facilitate the process of recording and transcription.

The participants
To select the research participants, purposive sampling proved to be the most suitable sampling strategy.This is because "it seeks only to represent itself in a similar population, rather than attempting to represent the whole, undifferentiated population" (Cohen et al. 2007, 113).Using the purposive sample approach, a group of second-year BA students of English Studies from the aforementioned universities will be selected to participate in the study.The reasons why second-year students have been chosen include their high likelihood of staying engaged in the longitudinal study, their already established familiarity with university life and acquaintance with the research world, their less busy schedule compared to their third-year peers, and the high probability of finding them on campus in case of contact loss.Moreover, the two nationalities have been selected based on the researcher's access to both the Moroccan and Hungarian sites.Accordingly, the target group will subsequently be divided into two homogenous sub-groups; a Moroccan and a Hungarian one.Bearing in mind that the participants are native speakers of Arabic and Hungarian, respectively, they constitute a suitable target group because they are multilingual individuals who have expectedly reached at least an intermediate level of proficiency in English as a foreign language.

The interview design
The research method follows Maykut & Morehouse (1994)'s general procedure for developing an interview schedule.The different stages of their procedure are as follows: writing down the focus of inquiry brainstorming key words related to the focus of inquiry grouping categories of inquiry selecting relevant categories to be included in the interview developing open-ended questions for each category of inquiry using index cards arranging the categories into a sequence drafting the interview piloting the interview making revisions beginning the actual interview It is worth noting that since the interviewees are participating in a longitudinal study, this interview only serves to collect data about the participants for the time being.Later on, the collected data will be used to make inferences about the relevance of the multilingual profiles to second language vocabulary attrition and maintenance.Moreover, since this paper reports on work in progress, only the first eight stages of interview schedule development will be discussed below.

Developing categories of inquiry
Having arranged various brainstorming sessions, ideas about the focus of inquiry (i.e. the multilingual profiles of the individuals) were drawn based on the theoretical framework established with the help of the literature.Then, similarities in these ideas were clustered together and grouped into categories of inquiry.For example, questions exploring the participants' spoken languages and how they became multilingual were all identified and grouped under the category "origin".The categories chosen for the interview are labeled (1) "origin", exploring themes related to native and second languages, (2) "function", exploring the frequency of language use, (3) "competence", exploring the participants' self-assessment of their linguistic competencies, and (4) "identification", exploring the participants' identities in relation to language.These categories serve to account for the diverse and unique background and features of bilingual individuals and to establish a correlation between these characteristics and aspects of second language vocabulary attrition.
8. Developing open-ended questions Patton (1990)  Amongst these open-ended questions, only the sensory type was disregarded, for it is not relevant to the focus of inquiry.Each of the rest was used when investigating the established categories of inquiry.For instance, the question "what language(s) do you personally identify as your second language and why?" aims at revealing background and value answers, and the question "what feelings do you hold towards this second language?"aims at unraveling the participants' affective states in relation to their linguistic identity.

Drafting the interview
A first draft of the interview was formed including an introduction of the interviewer, a statement of confidentiality, an informed consent form to be signed by the interviewees, a request for permission to record the interview, and a statement about the goal of the interview.Here, it is important to note that to minimize the observer's paradox, the participants were not told that the study is explicitly meant to examine second language vocabulary attrition and maintenance, instead, they were informed that it is a mere analysis of their multilingual profiles.The interview draft was then edited to become the official interview schedule after the pilot phase.To have a look at the interview questions, see the Appendix.

Piloting the interview
The interview was piloted with five Moroccan students.This proved to be sufficient to gather adequate data about the foci of inquiry.Some questions were omitted to avoid repetition, and others emerged during the interview.To ensure the internal validity of the schedule, data authenticity was maintained throughout piloting.For example, participants were incited to provide authentic and genuine responses based on their unique language experiences and encouraged to avoid providing socially desirable responses.This was facilitated through a secure non-judgmental interview space that allowed participants to express their thoughts freely.Accordingly, the collected data reflected the participants' legitimate language experience and provided in-depth insights into multilingualism and second language vocabulary attrition.
Reliability was enhanced through the careful piloting of the interview schedule, the highly structured formatting of the interview design, and the consistent wording and sequencing of the questions throughout the five pilot sessions.More elaborately, the consistency and replicability of the interview questions were continually assessed when conducting the five pilot interviews allowing the rephrasing of ambiguous questions and facilitating the elicitation of consistent responses across interviews.
The piloting stage has played an important role in refining the official interview checklist and identifying inconsistencies in the wording of questions so as to ensure clarity and coherence.Moreover, feedback from the participants regarding their understanding of questions and suggestions for improvement were considered, making sure that the yielded responses were comprehensive and accurate.The insights gained from the pilot study has ultimately led to the validation of a robust and reliable tool that explores the relationship between multilingualism and second language vocabulary attrition.

Conclusion
The present paper has reported on the validation of an interview schedule designed to investigate the intricacies of multilingualism in order to provide an understanding of the relationship between the linguistic background of bilingual EFL students and their vocabulary attrition and maintenance in English.The final version of the interview schedule makes it possible to effectively capture rich and nuanced insights into the participants' language experiences.Participants with diverse multilingual backgrounds have been selected to ensure reliability and validity, which enables this tool to elicit detailed narratives and explore factors influencing vocabulary attrition and maintenance.For example, participants of the pilot phase have illustrated the challenges they faced as regards maintaining vocabulary during prolonged disuse periods, and in the absence of repeated exposure to the target language.Also, they have provided insights into their linguistic identity outside of the socially constructed ones.Additionally, the paper has demonstrated the vital role of qualitative research in allowing a deeper understanding of the social and individual dimensions of vocabulary attrition and maintenance.By using this interview schedule, the participants' unique vocabulary maintenance strategies can be explored.Indeed, the narratives provided during the pilot phase have shown that mnemonic techniques, extensive reading, contextual guessing, and word formation strategies are popular and effective in vocabulary maintenance.
The validation of this interview schedule makes it a reliable tool that can be used by researchers exploring multilingualism and vocabulary attrition.The interview questions enable uncovering and exploring rich data comprising authentic examples and stories.Using this tool in future studies can potentially inform educational practices to support learners in maintaining their vocabulary proficiency and avoiding occurrences of attrition.

Introduction
Greetings!Thank you very much for your willingness to take part in this interview.I will be your interviewer for this study.My name is Hanae Ezzaouya.I am studying for a Ph.D. at the Doctoral School of Linguistics in Pázmány Péter Catholic University in Budapest, Hungary.I am currently conducting research on language pedagogy.You are one of the five students who agreed to take part in this longitudinal study and to participate in this interview from your university.