Primary and Secondary Research in Computer Science EE
Your research should involve both primary and secondary methods and sources.
Let’s explore what they are and how they can shape your research paper.
Primary research
A primary method refers to the direct collection of original data or information through methods such as:
experiments and laboratory research
observations
case studies
fieldwork.
This method involves generating new data that did not exist before, often with the researcher directly interacting with subjects or objects of study.
A primary source is an original, direct source of information. It provides firsthand accounts or direct evidence of the subject being studied. Examples include:
Original research articles reporting new findings (e.g., clinical trials, lab experiments)
Source code and software implementations are the actual code written for a new software system, algorithm, or application.
Experimental data: Raw data collected from experiments, such as results of algorithm tests, performance metrics, or user study results.
Documentation and technical reports detailing a new system or application's design, development, and implementation.
Technical reports and white papers
Secondary research
A secondary method refers to analysing or using data already collected by others. This method involves reviewing, interpreting, and synthesising existing information rather than generating new data.
It includes:
reviewing existing literature (journals, books, articles)
meta-analysis (combining results of multiple studies)
statistical analysis of existing datasets.
Secondary sources provide context, analysis, and commentary but do not present firsthand evidence. Examples of secondary sources in computer science include:
Books, articles, and surveys that analyse primary research, algorithms, or systems.
Review papers summarising the state of the art in a specific computer science field (e.g., advancements in neural networks or cloud computing).
Conference papers that review the latest trends and provide comparative analyses of existing technologies or methodologies.
White papers and technical reports produced by research institutions, universities, or tech companies that synthesise research findings.
Case studies from industry applications or software systems that analyze the effectiveness of specific technologies or frameworks.
Balancing Primary and Secondary Research
Implementing primary research in writing EE in DP Computer Science is not mandatory. However, there is a generic relationship between the quality of a research question and the use of primary and secondary methods and sources:
Balanced Use: An EE that effectively uses both primary and secondary methods and sources will likely have a strong RQ, offering room for literature-supported arguments from secondary research and a clear sense of discovery deriving from primary research.
Primary-Heavy: If an EE relies too heavily on primary data, it may lack academic context or the support of established sources, weakening its academic grounding.
Secondary-Heavy: If an EE leans too heavily on secondary sources, it risks being overly descriptive, with a weaker sense of discovery and limited academic justification.
No Use of Methods/Sources: An EE that does not employ either primary or secondary methods and sources is unsuitable, as it lacks a foundation in evidence and exploration.
Fig. 1 Primary and Secondary Research and Research Question Quality