Applied Corpus Linguistics最新文献

The vast majority of corpus annotation projects goes through a piloting phase in which the annotation scheme is gradually shaped through iterative annotation cycles until its final version is produced and applied to the collected data. The differences in annotators’ choices are usually recorded and reflected by the ‘Inter-annotator Agreement’ (IAA) that serves as a proxy to understand and resolve the raised issues. However, little has been reported on how to formulate a systematic approach to: (i) tracing the source of the differences in the annotators’ choices and (ii) provide attainable solutions that would considerably increase IAA. In this paper, the ‘Greek Learner Corpus II’ (GLCII) -the largest online greek learner corpus will serve as a basis to shed light on two commonly met types of ambiguity in error annotation that are closely related to target languages in which syncretism is ubiquitous in grammar (e.g., Greek and Romanian): a classification level and an identification level ambiguity.

Stimuli norming (the process of controlling experimental items to minimise bias) is important for the validity of psycholinguistic experiments. Survey norming (asking large numbers of people to rate or otherwise define the items) is typically used for this purpose but requires large samples. Clinical populations are not always large, nor easy to reach. Clinical participants often have ongoing symptomatology, and some cohorts experience language and communication difficulties. We present a corpus-linguistic method suitable for clinical populations for which survey norming is difficult or inappropriate. We also include the experiment generated, which measures metaphor-creation behaviour in schizophrenia to test Cognitive Constraint Theory (CCT) in clinical and nonclinical populations (see S2.1). We describe the design rationale before outlining the design stages in tutorial form. This allows us to show readers why the approach was needed and support them to consider and respond to the challenges that we encountered. We conclude that it is easier to consider norming and design practices in parallel when experimental units are defined linguistically. Corpus stimuli norming provides a versatile alternative when survey norming is prohibitive, especially in speech pathology.

First year writing (FYW) courses aim to prepare students for disciplinary writing. However, research suggests that FYW often fails to provide sufficient preparation for writing across genres and disciplines (Leki, 2007). A register-functional approach to corpus linguistics has elucidated key differences across disciplines and genres for both published and student academic writing (Biber and Gray, 2016; Staples et al., 2016; Staples and Reppen, 2016). To date, however, no studies have compared these features across FYW and First Year Engineering (FYE) writing.

This research uses a corpus of FYE and FYW texts developed by the authors. The subset for this study includes papers written by undergraduate students majoring in Engineering and taking FYE and FYW courses in the same semester. Technical Briefs (TB) and Design Reports (DR) were selected from the FYE corpus and Rhetorical Analysis (RA) and Research Reports (RR) from the FYW corpus. We investigated the situational context and normed frequencies of linguistic features hypothesized to show similarities and differences.

Our situational analysis shows key differences in characteristics of the RA and TB, particularly regarding audiences (clients for the TB, and instructors for the RA) and the object of analysis (advertisements for the RA and mathematical models for the TB). There were more similarities between the RR and DR, including a shared focus on a solution to a problem and the presence of both a methods and results section. Results from the linguistic analysis show the impact of the situational characteristics. For example, conditional clauses and premodifying nouns were used at similar rates of occurrence in the DR and RR, reflecting their inclusion of research questions and their sharing detailed information about the problem and solution. Implications of these findings for teaching in these contexts will be discussed.

This study presents a corpus-assisted ecolinguistic analysis of the evolving discursive representations of tree/s and forest/s in US American discourse from 1820 to 2019 in the approximately 475-million word Corpus of Historical American English (Davies, 2010). To explore these entities and their depictions in prevailing discourse, this study performs a diachronic collocation analysis of adjectives occurring with the terms across the span of the corpus. The analysis identified the 100 most frequent adjective collocates appearing with the singular and plural forms of tree/s and forest/s and calculated Kendall's tau correlation coefficient scores using decade-by-decade per million use rates in order to empirically assess the strength of trends in language use. The findings indicate a divergence in broadly positive and negative representations over the time span as adjectives construing poor health and lack of vitality are rising while adjectives conveying positive attributes of size, beauty, and wellbeing are declining. In addition, adjectives reflecting experiential engagement with tree/s and forest/s have progressively been replaced by a discourse of scientific identification and governmental dominion.

This corpus-based study examined the vocabulary in a 2.7-million-token corpus composed of digital science resources for middle school (6–8 grade) students in the United States. The findings of this study show that to reach the suggested 95%–98% lexical coverage thresholds of the Digital Science Corpus (DSC) that are conventionally deemed to facilitate minimal and optimal reading comprehension (Laufer, 2020), middle school (MS) students grade 6–8 must recognize the first 6,000 and 14,000 most frequent word families in the BNC/COCA (Nation, 2012), respectively, plus proper nouns and marginal words. The results of the lexical analysis across the three sub-corpora in the DSC suggest that the Life Science sub-corpora has a considerably larger vocabulary load than the Physical Science and Earth and Space Science sub-corpora. Additionally, while 98.60% of the most frequent 1,000 BNC/COCA word families occurred at least six times in the DSC, the 2,000–7,000 BNC/COCA word families provided significantly fewer opportunities for repeated occurrence. Since more than half of the words in the 5,000–7,000 BNC/COCA bands occurred five times or less in the overall corpus, most words across these bands do not have high enough frequency in the digital science resources to allow MS students to learn them incidentally from reading the texts found in digital science resources. Several pedagogically relevant suggestions for middle school science teachers are discussed.

One of Randi Reppen's major contributions has been her pioneering corpus research into school children's writing. In this paper, I will discuss how such research can contribute to both theory and educational practice. I will then look at two sets of unresolved methodological issues in this area: the issue of defining appropriate linguistic and textual categories, and the issue of drawing valid developmental inferences.

The issue of categories arises because corpus analysis depends on abstracting away from specific instances of language use in specific texts to make claims about the use of linguistic categories (e.g., noun phrases, low-frequency vocabulary) in textual categories (e.g., stories, science reports). Such abstraction enables researchers to draw out patterns of language variation that are difficult to spot by other means. But it also raises the problem of how to define categories that are reliably operationalizable, that capture consistent developmental patterns, and that are theoretically and educationally informative.

The issue of drawing valid inferences stems from the fact that corpus data record the products of complex, contextually contingent writing processes, involving the interaction of many variables. Capturing the combined outcomes of these complex processes promotes ecological validity. However, it also creates challenges for researchers who want to draw conclusions about specific aspects of the writing process, such as writers’ knowledge of vocabulary or grammar, or their emerging awareness of audience.

This paper will discuss these issues in detail, illustrating their impact and suggesting ways forward for educationally informative corpus research.

This methodology-focused paper reports how I compiled and analysed a 12-million-word corpus of threaded online discussions by employing Corpus Workbench tool (CWB, Evert & Hardie, 2011) and combining corpus analysis with micro-analysis drawing on the principles of digital Conversation Analysis. The tool not only affords an efficient retrieval and analysis of a large dataset, but also, more importantly, facilitates exploration of a corpus of online discussions based on different variables (e.g., topics of discussions, role of internet users, types of postings) and units of analysis (e.g., subforums, threads, postings). Examples are presented to illustrate how I used this tool to investigate various aspects of online discussions, and extract threads surrounding a particular topic or language practices for micro-analysis. I propose internet users’ interactions in online discussions can be further explored in the field of corpus linguistics by using this tool and a synergy of corpus linguistics and an interactional approach.

A corpus linguistic approach has been applied to examine the representation of donkeys in public discourses for an international equid welfare charity (The Donkey Sanctuary) with a view to improving the British public's understanding of the roles of donkeys, in Britain and worldwide. By increasing understanding of public perceptions, this study aims to support improvements in donkey welfare through targeted education.

The study explored patterning in public discourses about donkeys (online and print news and social media) using corpus linguistic (CL) techniques and tools, supplemented with methods from discourse analysis. The findings highlight key representations of donkeys in public and media discourses that are not present in informed discourses about the animals. In this paper, we examine the results of the corpus study from the perspective of one current aim from The Donkey Sanctuary's public engagement strategy: to promote understanding of donkeys as sentient beings with the capacity to experience a wide range of emotional responses to events or situations.

We found that donkey experience is more subtly represented in the discourses than other aspects of donkey lives, such as actions and behaviours, which have more obvious, overt representations. The results demonstrate the value of applying the CL framework for researchers and practitioners involved in textual analysis for charity communications and public awareness campaigns. We discuss the implications that our findings have for donkey welfare—and animal welfare more generally—as well as what such a methodology could offer other organisations providing public education and/or relying on philanthropic support from the public.

The effective implementation of health policies addressing opioid addiction may be jeopardized by the complex and sometimes mismatched beliefs and discourses held by policymakers, agency administrators, case managers, and ultimately target populations. Policies must be “aligned” socially across service levels, but misalignment by well-meaning stakeholders becomes a potential hindrance to implementation at different administrative levels. This observation motivates the study to ask, what do health policies and agents actually say? Data for this study come from policy documents (n = 100; words = 571,481) and ethnographic interviews (n = 29; words = 171,492) collected from rural, older adult health service offices. Results and analysis focus on comparing linguistic features, keywords and collocations, between policy texts and agents’ talk. Findings show a complex, socially mediated relationship between priorities and stances in official documents and the enacting agents, especially regarding the causes and effects of the opioid epidemic.