Research Questions Flashcards
(34 cards)
How can research questions be generated?
- Identify gaps/weaknesses in the literature
- Expand study to a new population, set of materials, setting, etc.
- Apply different outcome measure
- Assess social validity of the research
- Use state-of-the-art testing and analytical approach
What is the role of theory in the development of research questions?
- Motivates/provides a framework for the current experiment
- Organizes, explains, and accounts for data
- Describes relationships among concepts to explain a phenomenon
How can theory and data relate to research?
- Data-driven research: reasoning from the data to the general theory via induction
- Theory-driven research: reasoning from a general theory to the data via deduction
How should theories be evaluated?
- Comprehensiveness
- Precision & Testability
- Parsimony
- Heuristic value
Describe comprehensiveness of theory.
- Broad enough to account for as much data as possible
- If there are data relevant to a theory that it can account for: 1) adapt the theory to account for the new data OR 2) develop a new theory that incorporates the new data
Describe precision & testability of theory.
- A good theory should have concepts that are clearly and explicitly defined
- Should contain rational, logically-related statements
- Should have an empirically testable hypothesis
- Beware of omnipotent theories
What are omnipotent theories?
- Theories that are so powerful, general, or flexible that they account for everything
- So, not testable/falsifiable
- Must tread carefully between explaining a lot of data and too much
What is a “Strawman argument?”
-Someone who sets up a counter argument to a hypothesis that cannot possibly be true
Describe parsimony of theory.
- All things being equal, the simplest explanation tends to be the best
- Occam’s razor: the explanation of any phenomena should make as few assumptions as possible
Describe heuristic value of theory.
- Makes predictions (basic or applied), generates new knowledge, stimulates future research
- Makes general predictions doesn’t describe EVERY data point
What should be taken into consideration when narrowing down a research question?
- Types of research that might be conducted
- Whether you can practically conduct them
- Crafting the alternatives and the logic (avoiding a “strawman” hypothesis)
What are 4 main classifications of research?
- Quantitative
- Qualitative
- Empirical
- Non-empirical
What is quantitative research?
- Numerical data to which statistics can be applied
- Quantify attitudes, opinions, behaviors, etc.
- Generalize results from a sample to a population
- Methods, including surveys, structured interviews/observations, systematic experiments
What is qualitative research?
- Primarily exploratory: aim to reveal underlying reasons, opinions, motives, trends
- Often used to generate hypotheses to be tested in subsequent quantitative research
- Methods include unstructured/semi-structured techniques, verbal measures
- EX: focus groups, individual interviews, observations, fieldwork
- Sample size may be small
What is empirical research?
- Gathering new information through observations/materials
- Can be experimental, quasi-experimental, or non-experimental/observational
What is non-empirical research?
-Uses existing information in a research study
What does PICO stand for?
- P = population
- I = intervention, issue
- C = comparison/alternative
- O = outcome
What is a research hypothesis?
- Formal statement of the predicted outcome, includes IV/DV/population
- Null Hypothesis (H0)
- Alternative Hypothesis (H1)
- Can be directional or non-directional
Describe characteristics of well-formed hypotheses.
- Constructs are operationalized
- Measures are valid
- Measures are reliable
Describe how constructs are operationalized.
- Precise, specific IV and DV
- General statement becomes specific
Describe measure validity.
- How accurately a measure represents the knowledge/skill/trait you aim to assess
- Can be empirical or non-empirical
What are the types of non-empirical validity?
- Face validity
- Content validity
What are the types of empirical validity?
- Construct validity: convergent validity, divergent/discriminant validity
- Criterion validity: concurrent validity, predictive validity
Describe measure reliability.
- Consistency of measurement
- Inter-observer reliability
- Test-retest reliability
- Parallel-forms reliability
- Split-half reliability