01_Introduction Flashcards
What categories are used to evaluate the quality of empirical research?
- Objectivity
- Reliability
- Validity
What is Objectivity?
The measurements and results of the analysis are independent from researcher
–>Same result, if conducted by someone else
What is reliability?
= Absence of measurement errors
–>possible to reproduce results under constant condition
E(e) = 0
Reliability: how to improve reliability and what are ppotential causes of error?
- repeated measuring improves accurarcy (potentially)
Potential causes of error:
- lack of features consistency
- precision of measure
What is validity?
A measures measures what it is suppose to measure
Expected systematic error: E(s)= 0
Can repeated measuring improve results if validity is not given?
No:
because if validity is not given, that repeated measuring does not improve results
Is reliability a sufficient condition for validity?
No:
Reliability is a necessary,, but not sufficient condition for validity
Different forms of validity?
- Validity of statsitical interference:
- Internal validity
- Construct validity
- External validity
What is validity of statistical interference? (example)
Validity of statsitical interference: about the correlation (covariance) between treatment and outcome
—> the degree to which inferences and conclusions are meaningful and accurate
Example: if group sizes between control and treatment group differ ->low validity of statistical interfences
What is internal validity?
Internal validity: The validity of inferences about whether observed covariation between A (treatment) and B (outcome) reflects a causal relationship from A to B as thos variables were manipulated or measured
Ex: Is the intro of sales traning really the reason for increase in customer satisfaction?
What is Construct validity?
Construct validity: THe validity of inferences about the higher order consturcts that represent sampling particulars
Ex: Assesses the extent to which a measure or test truly represents the underlying theoretical construct or concept it is intended to measure
What is external validity?
External validity: the validity of inferences about whether the cause-effect relationship holds over variation in persons, settings, treatment variables, and measurement variables
Ex: extent to which the results are generalizable for other circumstances outside the study
Under which assumptions can a Causal relation be inferred?
Key element: A change in one variable causes a change in another varible
“If X is a cause of Y, we have in mind that a change in X produces a change in Y and not merely that a change in X is followed by or associated with a change in Y.”
Explanation of variables according ot their causal function?
(Independent, dependent, mediator, moderator)
independent variable: causes (x1,x2) (sales training)
dependent variable: efffect (CS)
Mediator: variable z by means of which x influences y –>explains the relationship
Moderator: variable m that effects the strength and direction between x and y –>influences the strength between a relationship (high, low moderator)
What is mediation?
Mediation openening the black box and explain the relationship
What is partial mediation?
Partial mediation: we have the indirect relationship (CS and WTP) and the direct relationship (CS to Loyal, Profits)
—> so the mediator explains some part of the relationship between the IV and the DV
What is full mediation?
Full mediation only indirect effect (CS to Mediator to Profits)
—> mediator explains the relationship between the IV and the DV completely, so that the direct effect of IV on DV becomes irrelevant
What does the moderator do?
Moderator does not open the black box, but establishes boundary conditions
–>High moderator or low moderator level
What are the 4 conditions for making causal inferences?
- Distinct entities: Cause (IV) and effect (DP) need to be distinct
- Association: Cause and effect need to covary (No deterministic correlation necessary, just higher liklihood of effect´s appearance)
- Temporal precedence: Cause has to occur before the effect (IV measured before DV)
- No rival causal explainations: Alternative explaination for the presumed relationship between the cause and the effect have to be ruled out
What is Quantitative research?
Quantitative research: that emphasizes quantification in the collection and analysis of data
What is qualitative research?
Qualiatative research: Research strategy that usually emphasizes words rather than quantification in the collection and analysis of data
What is the difference between quantitative research and qualiatative research?
Quantitative research:
- Numbers (quantification)
- Deductive (testing of theory)
- Realistic
- Static
- structured
- hard, reliable data
- Focus on behavior
Qualitative research:
- words
- inductive (generation of theory)
- Relativistic
- Process (dynamic)
- unstructured
- Rich, deep data
- Contextual understanding
- focus on meaning
