When writing a report within the domain of Computer Science, there are several research methods you could use as a way to provide emperical data and evidence. This blog post showcases a few of them that I think is worth using.
Computer Science can be merged into different domains, and can be characterized as an empirical discipline; the programs can be seen as experiments, whose structure & behaviors can be studied.
When we talk about "research", we refer to the activity of systematic enquiry or investigation in a particular area, with the goal of discovering new knowledge.
We can perform this investigation using several methods.
## Experimental Method
The experiments show what occurs from "real world experiments and implementation." This method is used in many different fields in Computer Science like artificial neural networks, automatic theorem proving, natural languages, analysing performances & behaviors. The results should be reproducible.
## Simulation Method
This method allows for the investigation of systems that can be outside of the experimental domain or systems that are under construction. Complex phenomena cannot be reproduced in a lab. The domains that can adopt computer simulation methodologies are sciences such astronomy, physics, and economics. More specialised domains such as virtual reality or artificial life work well with the Sim method.
## Theoretical Method
The more classical approach as it's related more to math and logic. The ideas could be formed around the conceptual and formal models (like data models and algorithms). Theoretical Computer Science "inherits its base" from logic and math, and thus deals more with problems such as iteration, recursion, and induction.
It is an important method, as it dedicates itself to design and algorithm analysis in order to find better solutions within issues such as optimization, performance, and perhaps space complexities. It is a difficult method for many, as it deals with the limits of computation as well as finding new theories.
## Quantitative & Qualitative Data
The research data can be primary. secondary, or a mixture of the two. It's gathered through surveys, experiments, interviews, observations, documents, websites, etc. As the header suggests, the data could be quantitative, qualitative, or mixed.
- The meaning behind the numbers.
- The results from a collection of numerical/standardised data.
- Analysis conducted through use of diagrams & statistics.
- The meaning behind the words.
- The results from a collection of non-standardised data, which requires classification into categories.
- Analysis conducted through use of conceptualisation.
Once we have the data, we must analyse it. We can use data analysis techniques to organise, categorise, and code data to find results and draw our conclusions. Quantitative and Qualitative data are obviously different, not only in terms of numbers versus text, but also epistemologically1 and methodologically2.
It's important to think about what data needs to be collected (and its analysis) within the early planning of the research proposal.