Corpus Analysis

Background
Typically, the research papers in your corpus will contain abstracts. Abstracts are thumbnail summaries of the research paper and should give potential readers enough information to decide whether or not the article is of further interest.
For this task, we will consider two broad approaches to writing abstracts:
• a results-driven abstract that concentrates on the research findings and what might be concluded from them;
• a summary abstract that provides short synopses of each section in the paper, often aligning with the IMRD structure.
Abstracts can further be described as either
• indicative (describe what was done), or
• informative (include the main findings).
This leads to an attribute matrix as follows:
Results-Driven Summary
Indicative
Informative
Four combinations result from the intersection of each column and row.

Analysis Task
1. Analyze the abstracts in your corpus according to the attribute matrix above. Provide counts for each combination. Offer your interpretation of the results.
2. How long are the abstracts in your corpus? Document the length (word count) of each and provide summary statistics.
3. What tenses are used throughout the abstracts in your corpus? Document the occurrence (frequency) of each and provide summary statistics. Do you observe patterns in tense usages, e.g. in opening statements or descriptions of methods and results? Offer your interpretation of the results.
4. Do you find first person pronouns (we, our, I, my) in the abstracts in your corpus? If so, how common are they?
5. What strategies do the authors of the articles in your corpus use to “sell” readers on their article? Sentence 8 of the abstract on page 385 of your text is an example of how two authors (Fowler and Aksnes) attempt to sell readers on the value of their article.
6. Do any of the articles in your corpus use structured abstracts? If so, what headings do they use?