Citation

Michael RP, Zumpe D. Soc. Biol. 1983; 30(3): 263-278.

Copyright

(Copyright © 1983, Society for the Study of Social Biology)

DOI

unavailable

PMID

6680801

Abstract

This study was designed to assess the significance of seasonal changes in assaults, rapes, robberies, and murders committed during 2 to 3 consecutive years in 16 different locations in the US and to relate them both to each other and to seasonal and geographic differences in temperature and photoperiod. The cosinor method of analysis was used to detect annual rhythms in the 4 types of violent crime. Data generated by the cosine functions showed that the annual rhythms in assaults were statistically significant in 12 of 16 locations with maxima between July and September, while annual rhythms in rapes were statistically significant in 14 of 16 locations with maxima also mainly between July and September. Annual changes in robberies were significant in only 5 locations, with maxima between Noverber and December. The annual rhythm for murder was significant in only 1 location. The close temporal relationships between the maxima in assaults and rapes and in the magnitude of the annual fluctuations, together with the fact that rapes are invariably less numerous than assaults, suggest that rapes comprise a subpopulation of assaults. A close statistical association was established between annual assaults and rape rhythms and annual temperature fluctations in the different locations, suggesting that some aspects of human violence may be influenced by exteroceptive factors. There were no significant associations between the mean latitudes of the different locations and the amplitudes of any of the 4 types of violence; the results do not eliminate a role for the photoperiod but fail to demonstrate it.

Language: en