Volume 7, Issue s1 - Using Smartphones to Collect Data for Linguistic Research / edited by Adrian Leemann and Nanna Haug Hilton

In the last decade, we have seen a number of studies come to life in which the collection of data for linguistic research has not followed a traditional path of holding in-person interviews or experiments, or using surveys for eliciting judgements, but instead have made use of smartphone technology and applications for collecting data. The current collection is the first to include papers with reflections from the linguistics community about the use of smartphone technology for linguistic research. The scope of the projects presented in this collection is a broad one. They have the mode of data collection, i.e. through a phone, in common, but all present different opportunities and challenges. The studies discussed in this introduction use smartphones to investigate language variation and change, clinical linguistics, psycholinguistics, and the sociology of language. Our hope is that this issue will provide ideas and inspiration, as well as access to readily usable tools, to keep researchers working, in a remote fashion, towards increasing our understanding the human competence of language.

The mobile app Schnëssen establishes a digital and participatory research platform to collect data on present-day spoken Luxembourgish through crowdsourcing and to present the results of data analysis to the general public. Users can participate in different kinds of audio recording tasks (translation, picture naming, reading, question) as well as in sociolinguistic surveys. All audio recordings are accessible to the public via an interactive map, which allows the participants to explore variation in Luxembourgish themselves. In the first year of data collection, roughly 210.000 recordings have be collected covering numerous variation phenomena on all linguistic levels. Additionally, over 2800 sociolinguistic questionnaires have been filled out. Compiling such amounts of data, the Schnëssen app represents the largest research corpus of spoken Luxembourgish.

In recent years, a number of applications on Swiss German have been released. They all crowdsource dialectological data, yet until recently their main focus has been on lexical and phonological features. In 2018, we launched the app gschmöis , an app designed to (1) give users an insight into dialectological research, and (2) collect Swiss German data on all linguistic levels, but with a strong emphasis on morphosyntactic phenomena. At the beginning, three rounds were published. Since then new rounds were published periodically. By including not only lexical and phonological features, we want speakers of Swiss German to become aware of differences in the grammar of various Swiss German dialects.The same linguistic phenomena are covered by multiple questions, and some of the questions are duplicated from more traditional projects, like atlas projects on Swiss German. This allows for comparison across different question types (i.e. translation tasks vs. multiple choice questions), different collection methods (fieldwork methods used in the SDS, written questionnaires used in the SADS, and crowdsourcing via the smartphone application gschmöis ), and different groups of informants (NORMs vs. speakers of Swiss German of various gender and age groups). Furthermore, the older data of previous projects can directly be compared with the crowdsourced data, so that recent dialect change can be detected.

This paper presents the project Stimmen fan Fryslân ‘Voices of Fryslân’. The project relies on a smartphone application developed to involve local communities in the creation of speech corpora, particularly of lesser used languages. This paper lays out the scientific and societal context of the project, showcases the smartphone application and gives an overview of the results from the project that attracted more than 15,000 users. Some key methodological issues are considered, and the paper discusses the role of smartphone technology for citizen science in minority language areas while also showing new maps with distributions of lexical and phonological variation in Frisian.

This contribution sketches the aims, participant recruitment and methodology of the Sprekend Nederland project, the amount and types of data, the research possibilities they offer and suggestions for similar future projects.

Smartphone apps are ubiquitous: in 2018 in Germany alone, 79% of the population owned a smartphone. Virtually everyone in this cohort always carries with them a recording device and a notepad (i.e., a screen), two essential tools that linguists typically use in the field. In the present contribution we discuss how linguists can harness this ubiquity of linguistic tools in the population to capture language variation and change, illustrated through apps that were developed for German-speaking Europe. We present four apps ( Dialäkt Äpp, Voice Äpp, Grüezi Moin Servus , and Deutschklang ) that were developed to (a) engage with the public and (b) to collect linguistic data. We discuss opportunities (e.g., the multimodality of said devices), as well as challenges (e.g., maintenance, updating and the costs involved therein). Finally, we present new findings that have emerged from working with this new paradigm and speculate about future directions and developments in using smartphone apps to collect linguistic data.

In this article we assess the extent to which we can collect plausible data about regional dialect variation using crowdsourcing techniques – the BBC Future Survey – without explicitly gathering any user metadata, but relying instead on background information collected by Google Analytics. In order to do this, we compare this approach with another crowdsourced survey, operated from a smartphone application, which examines the same site – the British Isles – but which explicitly asks users to submit detailed social background information – the English Dialects App (EDA) (Leemann et al. 2018). The EDA has the disadvantage that there is a considerable user drop-off between completing the dialect survey and completing the social metadata questionnaire. The BBC Future Survey, however, only collects information on where users are physically located when they complete the survey – not where they are from or even where they live. Results show that the BBC Future Survey produces a plausible snapshot of regional dialect variability that can complement other more sophisticated (expensive, time-consuming) approaches to investigating language variation and change. We suggest the approach constitutes a digital-era rapid anonymous survey along the lines of Labov (1972), serving similar aims, with similar success, but on a much much larger scale.

LinguaSnapp is a smartphone application developed to support documentation of linguistic landscapes, especially in multilingual urban settings. Since its launch in Manchester in 2016, bespoke versions have also been created for Melbourne, Jerusalem, St Petersburg and Hamburg. We describe the rationale and the operational setup behind the app, and provide examples of how the documentation facility can support analyses pertaining to the relation between language repertoire and spatial demarcations, and how it can be used in public engagement.

Public signage is a central element of the socio-pragmatic organization of everyday practice. The survey of signs and written language in the public sphere has developed into a vital branch of sociolinguistics called “linguistic landscapes”. The paper introduces a participatory research project, Lingscape, that focuses on the documentation and analysis of linguistic landscapes worldwide. Making use of a dedicated mobile research app, the project aims at creating awareness for the semiotic complexity and social relevance of public signage. After the discussion of two basic functions of public signage and two fundamental pragmatic conditions of sign perception, the text briefly introduces the app and project workflow. This is followed by the results of an analysis of the Vienna linguistic landscape using a large dataset collected with the app. This includes a quantitative evaluation of the contributions by different user groups to the (re)construction of a Vienna “crowdscape” as well as a qualitative investigation of the presence and social status of Austrian German in said crowdscape. Finally, the text presents an in-depth discussion about challenges for participatory research in practical work that focuses on the objectives of participatory research and the practical necessities against the backdrop of academic practice.

Smartphone technology is continuously being updated through software and hardware changes. At present, a limited number of studies have been undertaken to assess the impact of these changes on data collection for linguistic research. This paper discusses the potential of smartphones to gather reliable recordings, along with ethical considerations for storing additional personal information when working in other contexts (i.e. healthcare settings). A pilot study was undertaken using the Fitvoice TM account-based application to analyse articulatory proficiency in depressed and healthy participants. Results suggest that phonetic differences exist between these groups in terms of plosive production, and that smartphones are capable of adequately recording these minute aspects of the speech signal for analysis.

Worldlikeness, hosted online at https://worldlikeness.org, is a free Web crowdsourcing experimental tool and database that seeks to make it easier to generate and share cross-linguistic processing data. Adopting HTML5 technology and a responsive Web design using the JavaScript-based Meteor package, Worldlikeness supports the use of text and multimedia stimuli in a user interface optimized for participants using either desktops or mobile devices. Worldlikeness allows individual psycholinguists to collect data on individual languages, store the data in theWorldlikeness server, and share them to help build a cross-linguistic database. As the database grows, outside researchers can freely conduct their own typological cross-linguistic analyses and contribute to our understanding of language-specific and universal properties of language processing. In this paper, we introduce the key features of Worldlikeness and compare online wordlikeness judgments to in-lab data to demonstrate its reliability.