An error of reasoning brought about by arguing that the relationship between properties of groups of people must apply to the people themselves (or vice versa). The analogous situation with categorical data is described as Simpson's paradox.
Ecological fallacy. Considering only the subpopulation means (black dots), we would conclude that the correlation between the random variables X and Y was positive. However, the data for each subpopulation (the groups of open circles) show that X and Y are in fact negatively correlated.
A wrong assumption about an individual based on aggregate data for a group; the presence of a relationship between two variables at an aggregated level that is due simply to aggregation, rather than to any real association (Openshaw (1984) Env. & Plan. A 16). See Dark and Bram (2007) PPG 31, 5 on the modifiable areal unit problem and the ecological fallacy; see G. King (1997) on a solution.
An erroneous interpretation of observed association between two variables at the aggregate level as an existence of such association at the individual level. High sample correlation at the aggregate level between two variables weakly correlated at the individual level, referred to as the ecological correlation, is caused by the removal of individual variation by the aggregation (the aggregation problem).
- anti-static precautions
- Antisthenes (c.445–c.360)
- anti-Stokes radiation
- antisuffocation valve
- antisymmetric
- antisymmetric matrix
- antisymmetric relation
- antitail
- antithesis
- antithetic fault
- antithetic variable
- antithixotropic fluids
- antitoxin
- antitrade
- antitrades
- antitrust
- antiviral
- anti-virus program
- anti-virus software
- anti-Zeno effect
- Anti‐Comintern Pact (25 November 1936)
- Anti‐Corn Law League
- Anti‐Masonic Party
- anti‐Semitism
- Anti‐Trust laws