Academic Writer

academic writer: evidence: secondary sources: literature review

Literature Review
If you are working on an academic paper, it is likely that you will want to extend or challenge current knowledge, theories or assumptions in the field. You will therefore need to demonstrate your awareness of other published work in the area. This can be done by reviewing the existing literature in the field and by using citations and references. Such a review will probably be descriptive and neutral in tone. It may also identify a research gap which you will then go on to exploit with your own work. A shorter form of review is a survey.
EXAMPLE

Learning strategies in ESP classroom concordancing: an initial investigation into data-driven learning
This interest in small-scale corpora concordancing began in the mid-80s, most notably with the work of Higgins & Johns (1984), and Johns (1986, 1988), which stirred up great interest in this field. The result is that computerised text analysis has been brought much closer to teachers, course designers, materials developers and learners alike, and small-corpora concordancing as a tool for text analysis or as a pedagogic activity is increasingly brought to test and experimentation in places where microcomputers are available.
Tribble and Jones (1990:7) describe concordancing as "locating all the occurrences of a particular word and listing the contexts", while Levy (1990:178) defines a concordance as "a collection of all the occurrences of a word, each in its own textual environment, together with references and word frequencies". What the computer does in concordancing is to display all the contexts in which a certain word or string appears in a text or collection of texts, called a corpus.
As small-corpora concordancing penetrates the ESL classroom, it is developing into an ESL methodology and is strongly supported by a number of researchers and applied linguists, notably Stevens (1988, 1990, 1991a, 1991b), Johns (1986, 1988, 1991a, 1991b), and Tribble & Jones (1990). The rationale for this methodology, called classroom concordancing (CC), is one of authenticity and discovery. Johns (1986) names this concordance-based approach 'data-driven learning' (DDL). As the name suggests, this approach is characterised by language data taking on a primary role in language learning. In Johns' view, concordances provide `intake', (after Corder, 1967) i.e. the part of input that is actually helpful, to the language learner, which strikes a healthy balance between the "highly-organized, graded and idealized language of the typical coursebook" and the "potentially confusing but far richer and more revealing `full flood' of authentic communication" (Johns 1986:159).
Learning strategies in ESP classroom concordancing: an initial investigation into data-driven learning, Bruce K C Ma

EXAMPLE 2

The relevance of environmental air pollution studies
Before we discuss the relevance of environmental air pollution studies, the curious reader may ask, ‘Why are such studies needed in the first place?’ In a commentary about the state of medical knowledge on respiratory illnesses, Florey and Leeder (1982) said that animal studies and laboratory studies on human volunteers have provided important pointers on possible health effects from environmental exposures, but these results have limited generalizability to human populations.
McMichael (1989) and Lam et al (1992) examined the environmentalist’s role in setting exposure standards and pointed out that at high levels of ambient air pollutant exposure, the empirical evidence of a causal association with ill-health is compelling. For example, in the 1930s through 1950s, exposures to high peaks of air pollutant concentrations in many cities in the West were consistently found to be strongly associated with excess mortality rates. Goldsmith (1990-91) called this the ‘era of air pollution disasters’ and cited incidents such as the infamous 1952 London smog and the pollution episodes in the Meuse Valley of Belgium in 1931 and Donora in Pennsylvania in 1948. Stricter regulations and monitoring in the aftermath of such tragedies had decreased deaths attributable to air pollutant exposures by substantial proportions.
A study by Lave Seskin (1977) is a case in point. Their investigation was concerned with mortality rates in the 1960s, an era when air pollution levels in the United States were beginning to drop considerably, after the first air quality standards were implemented (Goldsmith 1990-91). They concluded from their models that a 50% decrease in air pollution levels in the United States would result in a 4.7% decrease in mortality. However, as pointed out by Tietenberg (1992), when an independent group of researchers analysed the same set of data, the estimated decline in mortality was only 0.43%.

The above example highlights the problem of assessing mortality effects of chronic exposure to relatively low (compared to the 1930s to 1950s) but still potentially harmful levels of air pollution. For this reason, many epidemiologic investigations in the last 20 years have focused on morbidity effects (e.g. WHO 1978, Ferris et al 1979, Ware et al 1986, Dockery et al 1989, Forastiere et al 1992, Peters et al 1996). Adverse health effects from long term exposure to lower levels of air pollution are smaller and harder to measure (McMichael 1989, Lam et al 1992).

Liu (1994) conducted a survey of the environmental air pollution literature, in particular focusing on the effects of poor outdoor air quality on the respiratory health of children from 1970 to 1993. This relationship is important because there is some evidence that chronic respiratory conditions in childhood increases the risk of lung cancer in later life (Samet et al 1983, Cooreman et at 1990). For nine out of the thirteen studies reviewed, chronic exposure to ambient air pollution was associated with an increased risk of respiratory illness.

EXAMPLE 3

In 1990 Theodore S. Glickman and Michael Gough assembled a number of papers, "Readings in Risk", as an introduction to risk definition, assessment, evaluation and management.1 Three papers in the series offer definitions of what we mean by risk, as it relates to the engineering profession.2 These three papers pose the classic question about technology-induced risk: "How safe is safe enough?"
In his "Social Benefits Versus Technological Risk,"3 Chauncey Starr concludes that we are generally willing to take voluntary risks that are 1000 times (three orders of magnitude) as uncertain as involuntary risks. In Starr's studies, risk is generally proportional to the cube of the incremental wages involved in recognition for taking that risk. For example, doubling wages would tend to convince a worker to take eight times the risk.
To assist in solving this problem, Norman C. Rasmussen, in "The Application of Probabilistic Risk Assessment Techniques to Energy Technologies,"4 suggests a more basic definition for "risk," namely, "consequences/unit time." Rasmussen argues that risk is the product of frequency (events/unit time) and magnitude (consequences/event).

 

Academic Writer 2000