- 2.0 Table of Contents
- 2.1 Learning Objectives
- 2.2 Summary
- 2.3 Academic Strengths and Developmental Areas
- 2.4 Literature Searching
- 2.5 Writing a Literature Review
- 2.6 Reference Management Tools
- MLO3 - Undertake a literature search
- MLO3 - Identify relevant texts from a literature search
- MLO3 - Critique and summarise relevant texts
- MLO3 - Utilise reference management tools
- MLO3 - Write a draft proposal title
Research projects require thorough literature surveys and critical evaluation. The process begins with a literature search and review to justify the project's worth and establish its relevance. Various resources, such as university skills guides and online databases, aid in finding credible sources.
Critical reading skills are essential to evaluate evidence, identify gaps, and understand limitations. The SQ3R model is a common approach to effectively summarise, absorb, and recall information.
Writing a literature review involves presenting a coherent argument, referencing past and current literature, and justifying the proposed project. Proper citation and referencing are crucial. Finally, managing information and using reference management tools streamline the research process.
- Required: Chapter 5, Projects In Computing And Information Systems: A Student's Guide, Dawson
- Extension: none
The initial investigative work is crucial in any project and can be the determining factor between a good project and failure.
Good projects are built on a literature survey, which consists of two primary components:
- The literature search represents the mechanics of looking for, sorting, managing, and digesting the available research material
- The literature review represents written understanding, critical evaluation, and conceptualisation of the research material
Performing this survey justifies that the project is worth doing, establishes wider context for the project, and helps future researchers understand how the project evolved.
It also helps establish the relevance of the topic by answering these questions:
- Has it been researched before?
- Is the topic important, and if so, to which fields of research?
- What methodologies have been used to study it before?
As it is rare to find a topic that has never been researched previously, there should be a vast amount of published academic research that can be used to frame research projects.
Insights gained may illustrate suitable methodologies and methods for the topic, but could also result in replicating previous methods in favour of something new.
It is possible to study the same topic or research object by altering the unit of analysis.
The university provides support for this proceess:
The literature survey is made up of concurrent searches and reviews. It is:
- An ongoing, iterative process
- Constantly refined and redefined depending on material found
The process starts with the intended literature search being defined: identifying the boundaries of the search and the topics of interest.
- This could start as a personal and subjective understanding of the field
- A more focused start could be to limit searches to key authors or publications
- Or conceptual maps, such as relevance trees, could be used to help identify a starting point
With the boundaries defined, a literature search can be undertaken, according to the definition previously established.
Any material found requires critical evaluation. This provides the basis for the literature review.
Having completed a cycle of the spiral, more questions or influencing factors may have been uncovered. Similarly, it may be apparent that the focus was too broad or too narrow and requires adjustment. The cycle should, therefore, start again.
It is difficult to know when to stop the literature review. While it is impossible to read everything published in an area, the validity of the resulting study will be enchanced by undertaking a systematic process.
- Required: Chapter 5, Projects In Computing And Information Systems: A Student's Guide, Dawson
- Required: Chapter 3, Writing For Computer Science, Zobel
- Extension: none
With the topic identified, a systematic search for relevant and credible literature can begin. Activities, such as relevance trees, will provide various key areas, phrases, and words with which to start.
The purpose of literature searching is to be well informed about existing research on a topic, identify gaps in current knowledge, and avoid duplicating previous work.
It is important to develop effective reading skills in order to identify the contributions, value, and flaws of research papers.
As papers are read, the reesarcher should attempt to establish:
- Novelty
- Theories
- Debates
- New directons
- Alternative perspectives
Note that it is recommended to seperate the searching and reading of papers: gather material uncritically, and analyse subsequently.
The types of literature suitable for examination are:
- Primary - original research reports
- Secondary - reviews, surverys, and evaluations of primary literature
- Tertiary - encyclopedias and textbooks that summarise secondary literature
It would not be expected to use sources like lab notebooks or survey responses in the search.
Generally, credible sources will have been published and peer reviewed.
The literature ultimately relied upon should primarily consist of refereed and published papers, theses examined at reputable institutions, and books based on refereed sources.
A possible starting point if the university library webpages:
Abstracts appear at the beginning of academic papers and provide a concise summary of the research conducted. Searching abstract databases can be an efficient approach. However, it is important to read the actual papers when relevant.
Similarly, specialised internet search engines, such as Google Scholar, can be used to locate resources. The university provides some guidance on using search engines.
Reference lists at the end of academic papers contain lists of sources used by the authors in their research. These can be used to direct a literature search.
Wikipedia can be useful for general overviews, but caution is needed due to the lack of academic refereeing.
Resources specific to Computer Science include:
- ACM - digital library of all material published by the ACM
- IEEE Computer Society - online biblographic database
- IngentaConnect - collection of academic and professional research articles
- HCI Bibliography - focus on human computer interaction research
- Journal of Digital Information - focus on digital information management
Several tips have been suggested to help the search process:
- Note interesting quotes and references as the search progresses
- Use review articles and books to help the search
- Reference correctly from the start
- Know when to stop based on the plan and research boundaries
- Organise and catalogue the read material effectively
- Read recognised leaders and original theorists in the field
- Start with a broad search before focusing
- Be honest if the research idea is not as original as first thought
Most academic publications come in three forms: articles, edited chapters, and books.
Critical reading goes beyond understanding a text. It involves:
- Evaluating evidence and arguments
- Understanding research limitations
- Identifying explicit or implicit influences
- Personal agreement or disagreement
- Identifying emerging themes across various texts
An important first step is the source: ideally peer-reviewed journal articles.
Evaluating sources is a key part of critical reading. Factors to consider include:
- Authority of the author
- Intended audience
- Purpose of the work
- Quality of the argument
- Relevance of the content
- Accuracy of the content
Questions that can aid a critical reading are presented below:
However, critical thinking should not be confused with general criticism: the former is about evaluation strength and validity of arguments, not finding fault in a derogatory way.
Several models exist to help read and review literature. The SQ3R model is one such example, which can be used for effective reading and information absorption:
- Survey
- Question
- Read
- Recite
- Review
The first step is to survey the document by looking at the title, authors' names, abstract, table of contents, and headings.
Next, questions should be created based on these areas. These should be specific queries.
The reading process involves reading the document selectively, section by section, and using annotations to summarise key information. This should allow answerable questions from the previous stage to be identified.
Reciting or rewording the information is essential for understanding and paraphrasing the key points.
After some time has passed, perhaps a day, the notes can be reviewed for clarity.
Once this process is complete, revisiting questions and reviewing answers aids both retention and comprehension.
Further supporting resources are available through the university.
- Required: Chapters 2, 5, Projects In Computing And Information Systems: A Student's Guide, Dawson
- Chapters 9, 10, Dissertations and Project Reports: A Step by Step Guide, Cottrell
- Extension: none
The purpose of a literature review is to refine research questions, explore gaps in current knowledge, avoid repeating past work, and identify useful research methods.
Generally, it should provide:
- A coherent argument leading to a description of the proposed study
- References to past and current literature in the field
- Indicate areas where previous researchers agree
- A discussion of current omissions and identified biases
The field should be presented as a series of inter-related subtopics, with references supporting arguments and explaining the topics covered. However, references should not be used to pad out the report or simply prove key texts have been read.
The research field can provide a framework for the literature review, as illustrated below:
Three common ways to present literature reviews are:
- Presented as a single cohesive chapter
- Divided into a series of smaller chapters
- Raised within the report as relevant issues arise
It is important that the review is selective, not just a report on the content of each article.
Every literature review should implicitly incorporate these points:
- Arrange relevant literature in the field
- Critically evaluate past and current research
- Identify the proposed project within a wider context
- Justify the proposed project by showing how it will fill a gap in the field
Proper citation and referencing is critical to avoid plagiarism and properly acknowledge the work of others.
The review should be structured logically. This is usually thematic or chronological, as in the example structure below:
- An introduction justifying the content
- Demonstrating how the subject area has evolved
- Expanding on the area of study
- Discussing recent developments
- Highlighting current limitations in the field
- Justifying the relevance of the paper
Critical evaluation involves comprehensive analysis of research articles.
This goes deeper than that of the SQR3 model in the literature search, providing the required depth for the writing of the literature review.
Suggested steps, collated from various sources, include:
- Considering the type of article: review paper, theory paper etc...
- Identify useful ideas, techniques, or quotes from the article
- Evaluate the author's credibility and recognition in the field
- Determine the contribution of the article to the field
- Check if the conclusions flow logically from the work presented
- Check if the arguments are logical and supported
- Differentiate fact from unsubstantiated opinion
- Assess if there are counter arguments
- Consider if the article contradicts or supports other viewpoints
- Check the references are appropriate, relevant, and up to date
- Evaluate the methodology, design, measures, and generalisations
It is important to note that critical evaluation goes beyond understanding and repeating content: it requires awareness of boundaries, developing areas, dead-ends, limitations, and contradictions.
- Required: Chapter 5, Projects In Computing And Information Systems: A Student's Guide, Dawson
- Required: Chapter 8, Dissertations and Project Reports: A Step by Step Guide, Cottrell
- Extension: Reference Management: A Practical Guide, University of York
Research will involve a large numbers of articles, periodicals, and books being read and it is important to have a system to organise and manage the resulting information.
A conceptual model, like a relevance tree or spider diagram, can help identify and link topics of interest.
Articles could be oranised into plastic wallets or folders, with appropriate labels for quick information retrieval.
During reading, highlighting key phrases, sentences, and paragraphs is a useful approach. A colour coding scheme can be adopted, such as using separate colours for quotes, topic explanations, ideas, and contraditions.
Making summary notes on the front pages of articles and using post-it-notes to identify important pages are both potentially useful.
Index systems can be created digitially, allowing for information on articles and books read to be maintained and easily updated.
Reference management software can help manage references, simplifying managing citations. Among other benefits, this allows references to be in the correct format from the beginning.
The university supports four reference management software: EndNote, Mendeley, Zotero, and Paperpile. It provides a practical guide to using these.