A manual for writers of research papers, theses, and dissertations, 7th edition - Kate L. Turabian 2007
Draft your final introduction - Establish a brief context of prior research
Writing your final introduction and conclusion
Part I. Research and writing: from planning to production
Once you have a final draft and can see what you have actually written, you can write your final introduction and conclusion. These two framing parts of your report crucially influence how readers will read and remember the rest of it, so it's worth your time to make them as clear and compelling as you can.
Your introduction has three aims. It should do the following:
put your research in the context of other research
make readers understand why they should read your report
give them a framework for understanding it
Most introductions run about 10 percent of the whole (in the sciences they are often shorter).
Your conclusion also has three aims. It should do the following:
leave readers with a clear idea of your claim
make readers understand its importance
suggest further research
Your conclusion should usually be shorter than your introduction. (In theses and dissertations, the introduction and conclusion are usually separate chapters.)
10.1 Draft your final introduction
Different fields seem to introduce reports in different ways, but behind most of them is a pattern with the four parts described in 6.2.2:
1. Opening context or background. When this summarizes relevant research, it's called a literature review that puts your project in the context of other research and sets up the next step. Keep it short.
2. A statement of your research question. This is typically a statement of what isn't known or understood or of what is flawed about the research you cited in step 1. It often begins with a but, however, or other word signaling a qualification.
3. The significance of your question. This answers So what? It is key to motivating your readers.
4. Your claim. This answers your research question expressed in step 2. Here is an abbreviated example (each sentence could be expanded to a paragraph or more):
For centuries, risk analysts have studied risk as a problem in statistics and the rational uses of probability theory.context But risk communicators have discovered that ordinary people think about risk in ways that seem unrelated to statistically based probabilities.question Until we understand how nonexperts think about risk, an important aspect of human cognition will remain a puzzle.significance It appears that nonexperts judge risk by visualizing worst-case scenarios, then assessing how frightening the image is.claim
10.1.1 Establish a brief context of prior research
Not every report opens with a survey of research. Some begin directly with a research question stated as something not known or understood, followed by a review of the relevant literature. This is a common strategy when the gap in knowledge or understanding is well known:
The relationship between secondhand smoke and heart disease is still contested.
But if that gap isn't well known, such an opening can feel abrupt, like this one:
Researchers do not understand how ordinary people think about risk.
As a rule, writers prepare readers by describing the prior research that their research will extend, modify, or correct. If the report is intended for general readers, the context can be brief:
We all take risks every day—when we cross the street, eat high-fat food, even when we take a bath. The study of risk began with games of chance, so it has long been treated mathematically. By the twentieth century, researchers used mathematical tools to study risk in many areas: investments, commercial products, even war. As a result, most researchers think that risk is a statistically quantifiable problem and that decisions about it should be rationally based.
In a report intended for other researchers, this opening context typically describes the specific research that the report will extend or modify. It is important to represent this prior research fairly, so describe it as those researchers would.
Ever since Girolamo Cardano thought about games of chance in quantitative terms in the sixteenth century (Cardano 1545), risk has been treated as a purely mathematical problem. Analyses of risk significantly improved in the seventeenth century when Pascal, Leibniz, and others developed the calculus (Bernstein 1996). In the twentieth century, researchers widened their focus to study risk in all areas of life: investments, consumer products, the environment, even war (Stimson 1990, 1998). These problems, too, have been addressed almost exclusively from a mathematical perspective. [Detailed discussion of contemporary research follows.]
Some reports, especially theses and dissertations, go on like that for pages, citing scores of books and articles only marginally relevant to the topic, usually to show how widely the researcher has read. That kind of survey can provide helpful bibliography to other researchers, especially new ones, but busy readers want to know about only the specific research that the researcher intends to extend, modify, or correct.
It is important to represent this prior research fairly and fully: describe it as the researcher you're citing would, even quoting, not selectively or out of context, but as she would represent her own work.
Early in your career, you might not be able to write this review of prior research with much confidence, because you're unlikely to know much of it. If so, imagine your reader as someone like yourself before you started your research. What did you then not know? What did you then get wrong that your research has corrected? How has it improved your own flawed understanding? This is where you can use a working hypothesis that you rejected: It might seem that X is so, but. . . . (see also 4.1.2).