u	How much variation is there in the data, and to what extent can variation resulting from measurement or observational error be distinguished from true phenomenal variation?
u	The better the quality of data collected (and especially in conjunction with larger data sets), the better the chances of ascertaining a clearer picture of the phenomenon’s behaviour;
u	What is the relative base rate information?
u	What is the validity of the information source(s) – to what degree, if at all, does the information meaningfully support ‘predictiveness’?
u	What is the reliability of information source(s), what is the margin of error for the information? – extreme predictions based upon extreme observations from data sources of poor reliability are likely to have lower predictive validity (see regression);
u	Redundancy and validity should be discriminated in assessing the likely predictive value of the information – consistency between data sources is only a good cue to confidence in prediction to the degree to which the sources of information are not redundant;
u	To what extent might alternative explanations account for the data?
u	Is it possible for any predictions made to be rigourously tested?