| theme | _class | paginate | backgroundColor | footer | marp |
|---|---|---|---|---|---|
gaia |
lead |
true |
Computational Thinking and Social Science | :copyright: Matti Nelimarkka | 2023 | Sage Publishing |
true |
- Describe the various roles that computational methods serve in the research process.
- Evaluate different operationalisations.
- Elaborate on the benefits of having guiding concepts for research activities.
- Identify the challenges present in multidisciplinary collaboration, and propose solutions for at least reducing them.
- Discuss the impacts of computational social science on scholarship.
- exploratory or confirmatory
- experimental, causally oriented, or descriptive
- an effort to be objective versus focused on subjectivity
- an articulation of a qualitative, quantitative, or mixed-method approach
Programming can help to
- gather data
- analyse data
- interpret data
- Operationalisation: connecting the theoretical concept to a measurable variable.
- These examples demonstrate that computational approaches themselves do not indicate how our general tools should or should not be used in relation to specific research methods or theoretical concepts.
- It is the job of the scientists to anchor the code and its outputs in concepts and theories – the computer always runs what it is asked to run, even if the outcome has no clear conceptual relevance.
if your plans translate to shooting yourself in the foot, computational methods allow using a bazooka instead of a pellet gun.
Views about the role of theory in the research process differ with one’s approach to work in computational social science.
- Deductive reasoning: the hypothesis or explicit research questions are defined, and then the data are used to answer them.
- Inductive reasoning: evidence gathered through the research is utilised for developing proposals that explain the phenomena.
- Abductive reasoning: blending deductive and inductive styles.
Many ways of doing science
- Computational social science can be about translating established traditions and scholarship for computers.
- Computational social science can be about curiosity-driven enquiry with no established methods or problem formulations, thus pushing the boundaries.
- Instead of rigid operationalisation, work can focus around some conceptual insights, still giving freedom to exploration.
- In solo computational social science activities, the research process is rooted solely in computation and the research questions are answered only through computational methods.
- Mixed-methods approach, computational and other methods jointly.
- Group-based research requires development of
trading zones for conceptual work and active facilitation.
- Trading zone a site, physical or otherwise where collaboration takes place across disciplines’ boundaries.
- Facilitation focuses on ensuring participation and reification, producing concrete outcomes together.
- There is consensus at least that excelling in our research requires both fluency in social science and skills in computational domains.
- Multidisciplinary work can produce higher impact work.
- But: do we create artificial lines by dividing teams into ‘methodological experts’ and ‘domain experts'?
- Knowledge gaps and interest gaps limit abilities to collaborate in multidisciplinary manner.
- Should computational social science be in the heart of its own institute/research centre or, instead, encapsulated within established departments?
- Where does the locus of legitimate interpretation lay?
- How does computational social science impact the epistemologies of social science?
- Why is operationalisation an important stage in research?
- What is a guiding concept?
- What is the difference between solo and mixed computational social science activities?
- What challenges can emerge in collaborative academic work?
- What is a trading zone?
- What impact is computational social science already making on scholarship?