Research Basics Flashcards
What are the different types of knowledge?
Explicit:
- know what
- codified
- formalised
Tacit (Polanyi 1966):
- know how
- context dependent and personal
- includes cultural beliefs, mental models, skills etc.
Embedded:
-locked in processes, structures etc.
What are knowledge sources?
-Perception (empiricism) \+intuition, common sense, tradition \+personal experience (introspection, memory) \+second-hand experience (testimony) \+trial and error -inductive/deductive reasoning (rationalism) -empirical research \+scientific sources \+structured observation \+experimenting
Why is research superior?
-less likely to be biased
-because of:
+structured way of collecting, analysing and interpreting data
+pre-determined procedures
+repicability
+clear quality criteria
-emipircal rather than normative
-objective rather than subjective
-systematic rather than random
-rigorous rather than haphazard
What is described by the term ‘theory’?
- much more than pursuing ‘facts’, research is aimed at developing theory
- definition: a broad, natural explanation for a wide range of phenomena. Theories are concise, coherent, systematic, predictive, and broadly applicable, often integrating and generalizing many hypotheses.
- facts can be observed or measure; a theroy is based on interpretation of facts
- a theory is always provisional/tentative
The scientific method behind QRM
- observations
- question
- preliminary background research (inductive)
- hypothesis
- deductive research to test the hypothesis
- more integrated theroy or falsification
- leading to a new hypothesis
- adjustment of the theory when findings do not support it
- no cherry-picking!
What is measurement?
Measurement is the assignment of a number to a characteristic of an object or event, which can be compared with other objects or events
What is a construct?
A construct is a focused, abstract idea on something inferred from an observable phenomenon. Most constructs are (part of) theories and may be complex.
Variables and values
characteristic = variable
assigned number = value
comparison = quantitative analysis
Constructs, variables and items
Contructs are broad concepts or topics we want to study. These need ‘operational definitions’, in order to make them measurable. Operational definitions specify how a variable can be calculated, and what values it can have.
Variables are used to measure constructs.
Items are single questions in a questionnaire.
These single items are variables, but variables may also be computed from the values of more than 1 item.
What are the different measurement levels?
Nominal: no quantitative value, names
Ordinal: ordered
Interval: ordered, we know the exact difference between values, no real zero
ratio: continous, with a real zero
What are independent and dependent variables?
When you compare groups, the variable which defines the group is the independent variable. What you want to compare on them is a dependent variable. In case of causality, the cause is independent and the effect the dependent variable.
What is reliability?
Reliability is the consistency of a measure. A construct is considered to be more or less stable. A measure for that construct should therefore give the same value, regardless of the circumstances.
Test-retest reliability: consistency over time
Internal consistency: consistency across items
interrate reliability: consistency across different researcher
What is validity?
Validity is the extent to which a measure actually measures the construct it is supposed to measure.
face validity: the measure appears ‘on its face’ to measure the construct
content valdity: whether the measure covers the construct in its entirety
criterion validity: whether the measure correlates with other measures that shoulg be correlated or does not correlate with measure of unrelated constructs
predictive validity: if the criterion is measured in the future
What are internal and external validity?
Internal:
- whether an experimental condition or treatment makes a difference to the outcome or not
- and whether there is enough evidence to support the claim
External:
-generalisbability of the condition or treatment across settings