What are five concepts connected to Econometrics?
- Linear regressions
- Moderating effects
- Mediating effects
- Time series
- Panels
What concepts are connected to machine learning?
- Supervised learning
- Unsupervised learning
(- Reinforced learning)
What are five concepts connected to research design?
- Data analytics process
- Research question
- Hypotheses
- Data collection
- BSc thesis
What is solving problems in DA like?
Like solving them through a lens: how we frame the situation affects the results.
What is a good quantitative research question?
- Clear
- Focused
- Allows the writer to take an arguable position but does not leave room for ambiguity.
- Avoid the “all-about” questions.
- Empirically addressable.
- Something that you are interested in or care about.
What are six questions to ask for evaluating a research question?
- Is the research question clear?
- Is it too broad or too narrow?
- Is it something I/others care about?
- Is it a new spin on an old idea, or does it solve a problem?
- Is it researchable within the given time frame and location?
- Is it measurable and will the process produce data that can be supported or contradicted?
Why is a research questions (RQ) important?
- Clarifies what you are looking for (e.g. core concepts).
- Determines your research method.
- Helps you avoid asking something that cannot be answered empirically.
- Narrows down your theory and data collection.
- Enables you to communicate to others involved (co-researchers, participants, audiences, etc).
What is a quantitative research question?
Used to answer questions on relationships with an intention to explain, predict and control a phenomena.
What is a qualitative research question?
Used to answer relatively open-ended questions on experiences, interpretations and meanings.
What two overarching research methods are there?
- Quantitative methods
- Qualitative methods
What are quantitative research methods?
- Numbers and anything that is measurable in a systematic way of investigation of phenomena and their relationships.
- Used to answer questions on relationships, with an intention to explain, predict, and control a phenomena.
- Functions: identify a cause, quantify the effect of a cause, predict outcomes, explain, test, and refine theory or hypothesis.
What are qualitative research methods?
- Data like words and observations, which may be unstructured, and which often focus on meanings.
- Used to answer relatively open-ended questions on experiences, interpretations, or complex processes.
- Functions: Understand, process, explore, interpret, create foundation for theory.
What is a theory?
A statement about HOW a set of constructs are related to each other explaining a broad phenomenon.
- Answers questions of “how”, “when” and “why”, as opposed to only “what”.
- An important purpose of a theory section is to ground hypotheses.
What is a hypothesis?
A proposed explanation of a fairly narrow phenomena or set of observations.
- Not a guess: a theory-driven and informed attempt to describe what has been initially observed.
(statement that can be true or false)
What is the role of hypotheses?
- Identify and organize different issues that we need to consider in order to ANSWER THE RESEARCH QUESTION.
- Provide statements of how different constructs in the model are linked together.
A hypothesis is true or false, we decide to…
… accept (with uncertainty) or reject it
What is a Type I error?
The null hypothesis is true, but we reject it.
What is a Type II error?
The null hypothesis is false, but we accept it
What is the connection between constructs and variables?
A construct/concept is an abstract category in your theory which cannot be measured directly. In empirical research, you assess constructs using variables, which can be measured.
What is research design about?
HOW data are collected.
- Experimental
- Descriptive
- Correlational
Research question guides research design.
Research design is all the necessary steps around the data analytics for it to make any sense.
What are some strengths of quantitative methods?
- Scientific objectivity: data can be interpreted with statistical analysis, and since statistics are based on the principles of mathematics, the approach is viewed as scientifically objective and rational.
- Most useful for testing, validating and refining already constructed theories.
- Fast analysis
- Replication
What are some weaknesses of quantitative methods?
- Variability of data quantity
- Fairly narrow settings
- Be aware of three sources of biases: you, the respondent, and the method.
Correlation is not causation. Which three conditions must be met for causality?
- The cause must precede the effect.
- The cause and effect should correlate.
- All other explanations of the cause-effect relationship must be ruled out. An effect should be present when the cause is present, and when the cause is absent the effect should also be absent.
What are the four main types types of quantitative research designs?
- Descriptive
- Correlational
- Quasi-experimental
- Experimental
