The creation of Grammar and Beyond involved many different aspects of research on language use, corpus research, and what we know about language learning and teaching experience. This paper focuses on how information was selected and used in the creation of Grammar and Beyond.
Three principal layers of research informed the selection of material in Grammar and Beyond:
1. Published, corpus-based research on language use
2. In-depth field research (classroom visits, surveys, and extensive piloting of the materials)
3. Extensive original research using Cambridge corpora and also some specialized corpora
PUBLISHED RESEARCH AND FIELD RESEARCH
The most distinctive feature of Grammar and Beyond is the extensive corpus research that infused its creation and content. This corpus research used previously published research about specific linguistic features and the context of use for those features.
The field research for Grammar and Beyond, of course, included the traditional field testing of units and revising based on the results of in-class pilot testing. It also included extensive testing of topics, types of exercise, format, and surveys of students and teachers about topics and tasks. But that is only the beginning of what makes Grammar and Beyond a unique resource for both students and teachers.
ORIGINAL CORPUS RESEARCH
In addition to previous research, specific vocabulary and grammar features were thoroughly researched using the robust corpora available through Cambridge University Press, including some specialized corpora created for this project.
Let’s begin by making sure we know what we mean by a corpus. A corpus is a principled collection of natural texts that are in electronic format and can be searched using various tools. For the Grammar and Beyond project we used several corpora: the Cambridge corpus of North American English (texts range from casual conversation to academic prose), the Cambridge learner corpus (made up of learner language and tagged for grammatical mistakes), and some specialized corpora created for this project (e.g., freshman writing used to inform Level 4). It is important to remember that corpora can be searched using very specific criteria, and that particular subsets of a corpus can be included or excluded from a particular search depending on the goals of the specific question being explored.
DATA FROM THE REAL WORLD
Now on to some specifics. The most noticeable areas where a corpus was used are the many “Data from the Real World” boxes and information that is highlighted with a globe icon. These sections include information that can only come from the analysis of a corpus. In many cases, these sections contrast spoken and written language, using graphs or specific concrete examples such as, “Say this… but write this….” In some cases, these sections provide information about common vocabulary that is associated with a particular grammatical structure (e.g., prepositions that commonly occur with certain adjectives, verbs that frequently occur in a particular aspect).
Vocabulary is a constant that is informed by corpus research. For example, Grammar and Beyond includes lists of commonly used count or noncount nouns, or expressions that are frequently used in compare-and-contrast essays. In contrast to some other ESL/EFL books, when we identify a feature or vocabulary item as common, we know that to be true, since it is based on a large amount of data and not casual observations that might not hold true. This principle of teaching the most common words informs every level in the series.
In addition to our own corpus research, we included words from the Academic Word List (Coxhead, 2000) throughout the books. Many of these words have multiple senses, and some of these are non-academic, and we were careful to take this into consideration. The words that appear in the Academic Word List and that are used in the academic sense in our units are highlighted along with their definition (see the appendices of each level).
Another aspect of Grammar and Beyond that comes directly from the corpus is the “Avoid Common Mistakes” section included in each unit. To provide the information in this section, the target feature for the unit was explored in a corpus of learner language to identify mistakes that were common across learners from many different first languages. This learner corpus is hand coded for errors, so that it can be searched using sophisticated programs that allow the user to search and sort for specific errors and levels of learner proficiency. Being able to group target errors by proficiency level was essential in providing level-appropriate errors for the different levels of Grammar and Beyond.
AMOUNT OF TIME SPENT TEACHING DIFFERENT FORMS
Some more subtle aspects that are informed by the corpus include the amount of space devoted to various features. Not all grammar forms are used equally. Some forms are somewhat rare and others are quite common.
Other factors in addition to frequency were also considered in the creation of Grammar and Beyond. The fact that a feature is common does not necessarily make it easy to master, so this was also considered when determining the amount of space to devote to various features. In some cases, relatively common but “simple” features, like modals, still need lots of practice since the use is quite tricky.
Corpus research also influenced the type of exercises in each unit. If a form is particularly common in spoken language, it is practiced in spoken language or in informal writing such as e-mail. If a particular structure has a strong association with a particular writing task (e.g., problem solution, cause-effect, narrative), it is presented and practiced in the context of that task. This provides learners with deliberate and meaningful input, which helps them master appropriate use of target forms.
RESEARCH ON THE FEATURES OF COLLEGE WRITING
Another subtle aspect that was informed through the corpus is the use of the freshman corpus, which was collected specifically for this project. This collection of writing from freshmen across the US was used not only to identify grammar features used by writers, but also to help identify the tasks that entering college students are often asked to perform (e.g., compare and contrast, argue a particular position).
Extensive corpus-based writing research also helped to identify the linguistic features that are strongly associated with particular writing tasks. For example, expressions commonly used when making comparisons are taught in units on compare and contrast essays. In addition to considering the types of text, Grammar and Beyond also considers the function of the task. A clear example of this is the section on narrative texts. In Grammar and Beyond, the narrative texts, especially in Level 4, reflect the use of narratives in informational writing, rather than the often-presented personal narratives found in most ESL/EFL textbooks. Grammar and Beyond also includes features not found in most ESL/EFL textbooks, such as the use of appositives (e.g., “Mr. Smith, Board Chairman, claims…”) and noun~noun sequences, which are frequent features of academic prose but are missing from almost all ESL/EFL writing and grammar books.
As the information and examples above demonstrate, extensive research, especially corpus research, is carefully and sometimes subtly woven into the very fabric of Grammar and Beyond, providing a robust and carefully crafted series that is solidly based on research, but that has a friendly and familiar appearance. This series is a tremendous resource for students and teachers alike.
Coxhead, A. (2000). A new academic word list. TESOL Quarterly, 34, 213-238