====== Ecological fallacy ====== Describes a //fallacious// inference in which a statistical characteristic is inadmissibly transferred to a level of [[glossary:aggregation_statistics|aggregation]] lower than that which it actually describes. This may, for example, take the form of a (faulty) [[glossary:statistical_syllogism|statistical syllogism]], as in this example: > Most people from Scandinavia are blond. > Björn comes from Scandinavia. > Therefore: Björn is blond. Of course, we don’t know whether Björn isn’t one of the few none-blond Scandinavians, so we can not draw such a conclusion. ===== Other names ===== * Ecological (inference) fallacy * Population fallacy ===== Description ===== Statistical indicators are always //simplifications// ([[glossary:abstraction|abstractions]]) of the underlying data, which reflect what may be a complex situation. This simplification reduces complexity – for example, a simple //average// is easier to track than a long series of individual data points – but it also results in a loss of information, not least the individual data points themselves, or information about possible sub-groups within the data. For example, a city’s crime rate is a useful measure for comparing the development over a period time or against other cities. However, it is not helpful for identifying subsets (e.g. neighbourhoods) within that city, let alone for drawing conclusions about //individuals//. Even in the city with the highest crime rate, the vast majority of residents are not criminals. The fallacy described here lies in //projecting// such an indicator – which refers to a rather high level of aggregation – to a lower level. For example, by assuming that a city’s crime rate (measured at a high level of aggregation) allows information to be derived e.g. about a //neighbourhood// (a lower level of aggregation) or even about //individuals// (the lowest possible level of aggregation). ==== Classification and Delimitation ==== A distinctive feature of this fallacy is that it relates to statistical indicators. It can therefore also be regarded as a fallacy of mis-interpretation of statistics. In fact, this article was initially drafted in this context. However, this can also be interpreted to mean that the underlying fallacy seems to be more a case of confusing [[glossary:induction|inductive]] and [[glossary:deduction|deductive]] logic. For this reason, it can also be understood as a //fallacy of induction.// The //ecological fallacy// is reminiscent of the [[logic:emergence:division|fallacy of division]], although the latter relates specifically to the phenomenon of [[glossary:emergence|emergence]]. In this case, an (invalid) inference is drawn from the properties of a //whole// to those of its //parts//. In a similar vein, there are also overlaps with the category of [[generalization:index|fallacies of generalization]], in particular the [[generalization:mereological_fallacy|mereological fallacy]]. This refers to the mistake of confusing the properties of a //system as a whole// with those of its (functional) //components//. Last but not least, the //ecological fallacy// can also be regarded as a [[abstraction:index|fallacy of abstraction]], namely when it is understood as a mistake in distinguishing between different levels of abstraction. In the broadest sense, however, the real error here lies in attempting to derive new knowledge in an illicit way. As this is the most general interpretation, which also covers all other cases, this fallacy is classified here under as a [[index|fallacy of knowledge]] or //epistemological// fallacy. ==== Valid use ==== In situations where, simply put, no better information is available, extrapolations from higher aggregation level data //can// be useful as a basis for [[glossary:heuristic|heuristics]] – that is, as a simplified approach which, whilst not producing "perfect" results, does at least yield results that are often "good enough". However, one should, of course, always be aware of the limitations inherent in heuristic methods to avoid over-interpretation. ===== Atomistic fallacy ===== Whilst the //ecological fallacy// refers to the extrapolation of statistical measures from higher to lower levels of aggregation, the reverse also has the potential to lead to errors. This is the case when knowledge about //lower levels// of aggregation (from the local community or a smaller group, right down to the individual) is extrapolated to //higher// levels (e.g. the population of a country) without checking whether such [[generalization:index|generalisations]] are even plausible in the specific context. This is one aspect of what is known as the [[knowledge:atomistic_fallacy|atomistic fallacy]]. It consists of reducing the //whole// to its parts, whilst disregarding structural aspects – such as the relationships between these parts, which can give rise to [[glossary:emergence|emergent]] properties. This is similar to the [[logic:emergence:composition|fallacy of composition]], which specifically concerns the illicit transfer of properties from //parts// to the //whole//. The //atomistic fallacy//, on the other hand, refers to the (oversimplifying) reduction of a //whole// to its //parts//. Both of these fallacies of course overlap considerably. ===== Examples ===== ==== Economic metrics ==== One key statistical indicator that is frequently misinterpreted is the [[wp>Inflation|rate of inflation]]. It describes the average change in consumer prices, usually compared with the same period in the previous year, and in this context it can be //misunderstood// in two different ways: Firstly, the published inflation rate does not mean that all goods are becoming equally more expensive by this percentage. Some products or product groups may see significantly //higher// price rises, whilst others may see significantly //lower// ones – some products may even become //cheaper//, bucking the trend. Consequently, selective comparisons along the lines of "the price of product X has risen by more than the official inflation rate since last year" are of little significance. Even less so, of course, if the intention is to suggest that the official statistics are inaccurate or even manipulated. Moreover, an economy’s general inflation rate does not necessarily reflect the //individual// inflation rate experienced by a specific person, group of people or even a particular economic sector. Ultimately, every individual’s consumption profile differs – including their varying ability to switch to substitute goods – which in turn leads to very different //individual inflation rates//. Thus, e.g. a pay rise based on the national inflation rate may lead to a //real// loss of purchasing power for some individuals, whilst others may even //gain// purchasing power. In both cases, an //ecological fallacy// occurs when the //general// rate of inflation is extrapolated to a different level – be it to individuals or population groups, or to specific products or product groups. A similar situation can also be observed with other economic measures. For example, so-called "[[wp>Purchasing power parity|purchasing power parity]]" (PPP) is a factor by which buying power differs across different economies, or across different periods within a single economy. However, this is always an //average// figure for the economy as a whole. It is easy to see that the purchasing power calculated for, say, locally produced agricultural goods differs significantly from that for e.g. high-tech defence equipment. Therefore, this figure should not be applied without qualification to individual products or even just product groups. ==== IQ differences between population groups ==== There are a whole heap of //misconceptions// regarding the differences between [[wp>Normal distribution|normally distributed]] properties. This will be illustrated here using the so-called "[[wp>Intelligence quotient|intelligence quotient]]" (IQ) as an example: Comparisons of IQs between different population groups (e.g. broken down by occupational groups, countries of origin, political party preferences, media consumption, etc.) are published frequently. Such comparisons often contain, at the very least, certain elements of [[wp>rhetoric/appeal_to_emotion/index|appeal to emption]] or //self-aggrandizement//, which is why they are so readily shared on social media. After all, who wouldn’t like to believe that a group to which they feel they belong is more intelligent than others … It would take us too far out of the scope of this web site if we tried to list all the problems that such comparisons have. However, drawing a conclusion about differences in the intelligence of //group members// based on differences in //average// scores – a scenario specific to the topic of this article – can be seen as a //perfect// example of the //ecological fallacy//. In short: even in the "most intelligent" group, there are always individuals of varying degrees of intelligence. Since the IQ is, by definition, //normally distributed,// it would in principle even be possible to calculate the overlap between the distributions of this trait in the two groups with great precision. However, it is not really worth the effort, as the figures are still rather imprecise and, in most cases, a distinction can only be made if one states them with a [[mathematics:statistics:misrepresentations:spurious_accuracy|misleading degree of precision]]. In most cases, however, the area of overlap between the two populations is so large (see figure) that it is hardly possible to draw any meaningful conclusions. Under no circumstances can any conclusions be drawn from such //group averages// with regard to the IQ of //individuals//. Even less so, by the way, with regard to their //intelligence//, which is not necessarily the same thing ([[abstraction/semiotic_fallacy|Semiotic fallacy]]). ===== See also ===== * [[generalization:accident_fallacy|Accident fallacy]] * [[generalization:mereological_fallacy|Mereological fallacy]] * [[rhetorics:red_herrings:ad_hominem:guilt_by_association:nutpicking|Nutpicking]] * [[logic:emergence:division|Fallacy of Division]] ===== More information ===== * [[wp>Ecological fallacy]] on //Wikipedia// * [[https://www.britannica.com/science/ecological-fallacy|Ecological fallacy]] on //Encyclopedia Britannica// * [[https://www.logicallyfallacious.com/logicalfallacies/Ecological-Fallacy|Ecological Fallacy]] on //Logically Fallacious// {{page>templates:banner#Short-BG-Article&noheader&nofooter}}