02.3 - Research Design - IES

Main elements

configuration of 4 key design elements:

• Research problem: what motivates your research
• Theory
• Method:
  • How to collect data
  • How to analise data to produce results
• Data (interviews, surveys, existing statistics, participant observation)

Concepts and variables

Dimensions in research design

What is the purpose of the study?

• Test a theory VS build a theory.
  • The first would use deductive approach, the second would use inductive approach

What type of data are you going to use?

• Quantitative
• Qualitative
• Mixed?

For deductive studies usually quantitative data is used.

For qualitative data, we usually use a more inductive approach

Survey design VS case design

Cross sectional vs lognitudinal time frame
For the latter, a longer time frame is needed

Logic of reasoning

Conceptualisation:
Process of formulating and clarifying concepts

Operationalisation
Research operation that will specify the value or category of a variable.

Deductive reasoning

Rely on theory to define the study. From there you define the method and then collect data.

Inductive reasoning

Start from the data, then craft a method, analysie the data and consult a theory. This can be iterative process.

Generic research designs

• [[#Quantitative research design]]
• [[#Qualitative research design]]
• [[#Case research]]
• [[#Design research]]

Quantitative research design

It's any form of research where any evidence is measured or translated into numbers.

It'w usually a cross-sectional study that tests some hypothesis.

It's usually related to statistical testing. Based on hypothesis development. Analysis is based on theory and statistical tools. Usually deductive. The researcher is isolated from the studied subjects

You consult theory to prepare the hypothesis.

Qualitative research design

Features and process is very different from [[#Quantitative research design]]. Collecting and analysing qualitative data is extremely costly and time consuming.

You try to focus on creating a very good and representative sample.

The phases are fundamentally interconnected.

• collect some material
• begin analysis
• construct an idea of a theory
• collect more material
• keep going...

In qualitative research you usually don't start with hypothesis. Researcher has to try to remain neutral.

Starting with hypothesis would highly bias results.

The role of researcher is active and maybe participatory. It's important to avoid influencing the interviewed people.

Case research

Quantitative and qualitative surveys tend to have more informants, and you try to analyse them independently of the context. In case research, everything is analysed closely to the context.

In case research, when working with multiple cases, discoveries may imply that we need to redefine the sampling strategy.

It aims at logical generalisation, but this process is quite limited. More options in comparative research.

In single case, it's hard to generalise since concepts may not apply.

Design research

Try to form a solution to a problem applying research method.

How to research question and often technical in nature.

The process of design science is different from the other forms of research.

It's more pragmatic. Formulate topic, then problem.