Observational Methods 2: Coding and Reliability

Coding schemes: Preliminary Considerations 

Coding schemes applied to observations to produce data. 
In other words, data are what you get out of a coding scheme.

Courtship Behaviour -> Coding Scheme -> Data -> Analysis
Coding Scheme = Measuring Instrument 

Coding Schemes: Principles of Measurement

• There's no such thing as perfect measurement 
• Measurements are more or less accurate. 
  Accuracy is about correct or valid measurement; the degree to which measurements actually capture the phenomena of interest. 
• Measurements are more or less precise. 
  Precision is about exactitude; the degree to which measurements are reliable or replicable.

Accuracy vs Precision 

• Key point 1: It is not effective to use a coding scheme that is more or less accurate than the research question for which the data are collected.
• Key point 2: It is not effective to use a coding scheme that is more or less precise than the research question for which the data are collected.

Principles of Design

• Codes should be mutually exclusive and exhaustive. 
• Mutually exclusive means that each behaviour can only be put into one category
• Exhaustive means all possible behaviours have categories

Intra- and Inter-observer Reliability:
Preliminary Considerations

• Reliability is about the precision of the coding scheme, NOT necessrily about the precision of the observer. If an observer is sloppy, it will show…