0.105252027511601128739NRSNEPublished to Web1Publication2Informally Refereed (Peer-Reviewed)3General Technical Report (GTR)<![CDATA[Statistical tests and measures for the presence and influence of digit preference]]> 1998-1Digit preference which is really showing preference for certain numbers has often described as the heaping or rounding of responses to numbers ending in zero or five. Number preference, NP, has been a topic in the social science literature for some years. However, until recently concepts were not adequately rigorously specified to allow, for example, the estimation of number of responses for a given stated value that are a consequence of NP. This paper reviews developments, identifies current research directions and then builds on methods of estimating numbers exhibiting NP to provide measures and statistical tests. A chi-square test of significance of NP; a test for the significance of particular heaps and confidence intervals for the estimated proportion of a population exhibiting NP are presented. The interpretation of the tests and measure is illustrated. Of particular practical value is a measure of the potential for NP to bias the value of the mean and total of a stated response variable.https://www.nrs.fs.fed.us/pubs/gtr/gtr_ne241/gtr_ne241_044.pdf 672 KBhttps://www.fs.usda.gov/treesearch/pubs/17075170753817100In: Vogelsong, Hans G., comp, ed. Proceedings of the 1997 Northeastern Recreation Research Symposium; 1997 April 6 - 9; Bolton Landing, NY. Gen. Tech. Rep. NE-241. Radnor, PA: U.S. Department of Agriculture, Forest Service, Northeastern Forest Experiment Station. 44-50.024144-50T0Beaman, Jay; Michel, Grenier; 07-MAY-2009 00:00:0007-MAY-2009 00:00:00AYBeaman, Jay001Michel, Grenier002Beaman, Jay001Michel, Grenier002This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.

]]>