Publication Type: Conference Paper/Unpublished Manuscript
Abstract: This paper examines what variables make victory more likely for defendants in criminal trials. Private as vs. public counsel, youth, and in-county status make victory more likely. Minority race, sex, prior record, and others make it less so.

Publication Type: Abstract
Abstract: To help conserve water for essential needs, many municipalities in the western U.S. are imposing restrictions on the amount of water that can be used for irrigating urban landscapes. These restrictions may adversely affect the growth of exotic plants in existing landscapes and will limit the selection of plant species that can be used as their replacements or that can be practically maintained in new landscapes.

Water savings are not obtainable through plant selection alone. Irrigation efficiencies must be maximized through appropriate system design and maintenance and by scheduling irrigations based on minimum plant requirements for acceptable quality. To help identify suitable species and their water requirements, a plant demonstration garden, exhibiting about 100 xeric-adapted species irrigated at four irrigation levels, was established at New Mexico State University’s Agricultural Science Center in northwestern New Mexico.

The garden was planted in 2002 and irrigation treatments (0, 20%, 40%, and 60% of reference evapotranspiration [ETr]) were imposed in late 2003. Landscape coefficients (irrigation/ETr) for climate-based irrigation scheduling were suggested for each species based on subjective qualifications of growth and quality at the different irrigation treatments. Suggested landscape coefficients (KL) for most species ranged from 0.2 to 0.4 (20% to 40% of ETr) which are considerably lower than the KL values of 0.7 and 0.9 formulated at the same site for warm and cool season turfgrasses, respectively.

This presentation will briefly summarize this study and will discuss climate-based irrigation scheduling with special emphasis on water management in western urban landscapes.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: Since McDonald and Moffitt (1980) introduced the decomposition of tobit coefficients into two different mechanisms (i.e. changes in the probability of having nonlimit values and actual changes of the nonlimit values), sociological literature has applied the decomposition method to substantial interpretations of tobit coefficients, particularly for the second mechanism. McDonald and Moffitt, however, did not highlight two limitations of the decomposition, and following sociological practices have been partly misleading. First, a decomposition ratio between the two mechanisms depends on the choice of values of respective predictors in a model. Sociological literature has performed the decomposition only at the mean point of sample and wrongly generalized the ratio to the whole sample. Second, once the values of predictors are fixed, a decomposition ratio is the same across all coefficients or across all predictors. Namely, two mechanisms are not substantially different in tobit model, but wrongly assumed to be different in research. Specific errors and confusions in literature are reviewed, and alternative utilities of the decomposition are discussed.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: What do Catalan Numbers and Bertrand's Postulate have in common? Other than the fact that they are misnamed and should really be known as Euler Numbers and Chebyshev's Theorem, respectively, they are both tied in some way to the Central Binomial Coefficient, C(2n,n). Other things that can be linked to the Central Binomial Coefficient include the Harmonic Series and related series, truncated, of course; Bernoulli Numbers, and some conjectures due to Erdös. This talk will expose all connections and present the ideas behind the proofs of most results that are discussed.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: A systematic review of 459 published and unpublished research studies reporting rater agreement for the Psychopathy Checklist family of measures revealed that few studies (n = 17) provide information about the reliability of scores calculated for real-world practice. ICC values for these studies (Mean ICC = .81) tended to be lower than those from studies in which trained researchers or clinicians scored the PCL for a research study (Mean ICC = .88). However, few studies (n = 16) reported the rater agreement most relevant to forensic practice (absolute agreement ICC for a single evaluator).