Prior Empirical Findings Research Articles

Three important determinants of user performance for database retrieval

Three important factors that determine user performance during database retrieval are representation realism, expressive ease, and task complexity. Representation realism is the level of abstraction used when formulating queries. Expressive ease is the syntactic flexibility permitted when formulating queries. Task complexity is the level of difficulty of queries. A controlled laboratory experiment was conducted to assess the effects of these three factors on user productivity during database retrieval. The independent variables were representation realism (high versus low), expressive ease (high versus low), and query complexity (simple versus complex). The dependent variables were query accuracy and query time. Results show that all these three factors significantly affected user performance during database retrieval. However, their relative impact on query accuracy and query time differed. Moreover, these factors interacted in unique ways to moderate query accuracy and query time. Besides verifying prior empirical findings, these results offer several suggestions for future research and development work in the area of database retrieval.

Post-Dispersal Predation on Isolated Seeds: A Comparative Study of 40 Tree Species in a Southeast Asian Rainforest

Many studies of post-dispersal seed predation have focused on density and distance dependent mortality, while relatively few have examined the fates of isolated seeds. Yet, scatter-dispersed seeds (sensu Howe) are commonly deposited singly or in small groups. Furthermore, even in species with highly aggregated seed distributions, the fates of the most widely dispersed individuals may be critical for recruitment. We compared predation rates on single, isolated seeds, among 40 species of trees in lowland tropical rain forest at Gunung Palung, West Kalimantan, Indonesia. Seeds were placed along four replicate transects and monitored for damage by predators, removal and germination in four trials, each lasting at least 30 days. Tethering of seeds did not affect removal rates, indicating that removals were attributable to seed predators and not merely to physical disturbance by animals or abiotic factors. After 30 days, mortality due to seed predation, averaged over species, was more than 50%; among species, predation losses ranged from 0 to 100%. Over the range of seed sizes we examined (0.1 g to 11.6 g fresh weight) predation rates were negatively associated with seed size and with the thickness and hardness of the seed coat. Lower predation on larger seeds is contrary to theoretical predictions and some prior empirical findings, and may be partially explained by the scarcity of predators capable of penetrating the physical defenses of large seeds with hard seed coats. Large, soft seeds with low predation rates may have poor nutrition or may be protected by chemical defenses. Species differed greatly in 30-day germination rates, ranging from 0 to 47%. Some species with low predation rates also had low germination rates; the implications for the overall risk of predation during the seed stage are discussed. Predation rates were not associated with species' natural dispersal mode (clumped vs scatter-dispersed), contrary to theoretical predictions. Spiny rats (Maxomys spp.) were the most abundant seed eating rodent. Caged spiny rats avoided large, hard seeds and preferred soft, medium sized seeds. The substantial rates of post dispersal predation on isolated seeds that we measured may be sufficient to influence strongly the population dynamics and life history evolution of trees in this rain forest community.

Relation between Order of Completion and Performance on Timed Examinations

The relation between order of completion and test performance was examined. Subjects were 116 students in two upper-division psychology courses. Tests were timed and contained both objective and nonobjective items. Contrary to previous suggestions but consistent with prior empirical findings, there was neither a consistent linear nor curvilinear relation between order of completion and performance. In one class, middle finishers tended to score lower and to be more variable in performance than early and late finishers. Results are discussed in the context of previous research on the relation between time (or order) of completion and test performance.

Action embedded in social structure: A systematic theory Part II, explanations

This research develops a systematic theory, which premises a responsible and reasonable actor, of how social structure can affect rational social action, applies this general theory to explain 29 specific empirical findings about voting, and illustrates how deductive theories can synthesize prior empirical findings and provide a focus for subsequent empirical research.

Action embedded in social structure: A systematic theory Part I, theory and evidence

This research develops a systematic theory, which premises a responsible and reasonable actor, of how social structure can affect rational social action, applies this general theory to explain 29 specific empirical findings about voting, and illustrates how deductive theories can synthesize prior empirical findings and provide a focus for subsequent empirical research.

Training and Contextual Interference Effects on Memory and Transfer

An experimental intervention program to train subjects on coincidence anticipation and prediction skills was administered to novice females. An attempt was undertaken to study the effects of this sports training and training on laboratory tasks. According to Battig's predictions on the structure of practice sessions, both random and blocked contexts were presented. These acquisition contexts were viewed as orienting tasks that preceded sports training. The effects of high-contextual interference and sports training on acquisition, retention, and transfer were investigated. Practice conditions in the intervention were selected because of processing demands similar to random acquisition. Other research has shown that acquisition in random is influenced by sport-skill expertise and further indicates that processing of events occurring outside the laboratory influences performance of laboratory tasks. Observed results supported prior theoretical predictions and empirical findings on contextual interference. Predicted hypotheses were supported, and the effects of sports training were most profound during retention and transfer if the acquisition context for the trained subjects was random. Supported by prior research, activities performed outside the laboratory influenced performance on laboratory tasks. These results shift more emphasis on the learner as an active processor of information, which relates to Lee's (1988) views on transfer-appropriate processing.

Responding to Older Offenders: Age Selectivity in the Processing of Shoplifters

This article examines the effect of the offender's age upon private and public responses to shoplifting offenses at a number of key decision points in the criminaljustice process. After briefly reviewing prior empirical research findings about the effect of suspect 's age upon witness and victim decisions to pursue redressive action against shoplifters, the article examines age selectivity at the prosecutorial stage in the criminal justice process. Using crosstabular analysis with simple elaboration techniques, the authors examine the decision to dismiss or prosecute charged shoplifting offenses across a trichotomous age classification. The findings demonstrate empirically that prosecutors selectively enforce legal norms against shoplifting, and that offender's age is a consideration in such decisions. Older shoplifters are found to be prosecuted less often than adults in the middle of the age distribution, but no less often than younger shoplifters. The implications ofthesefindingsfor the exercise ofprosecutorial discretion, public policy and program development, and common perceptions about the criminality of older citizens are highlighted in the conclusion.

The Measurement of Identification

Various attempts to measure parental identification have been seriously criticized. The present paper describes a new psychometric technique for measuring identification and discusses a large number of studies in which the technique has been employed. The nature of the relationships obtained between the identification scale and several relevant variables (level of adjustment, sex identity, conscience development, and ordinal position of the child; model attributes, nurturance, and social power of the parents; maternal approval of the father as a model) were consistent with theory and with prior empirical findings. It was concluded that the evidence supports the construct validity of the identification measure.