Subroutines to calculate exact chi square and Fisher's exact probability tests are presented for 3 by 2 cross-classification tables. A nondirectional probability value for each test is computed recursively. (Author/GDC)

Discusses current themes in communication research. Claims that this research offers an increasingly sophisticated view of journalism as a reflection not of an objective reality, but of political-economic power and sociocultural conflict. Includes an example of an imaginary symposium and a strategy for reframing the journalistic quest for truth.…

Contends that theoretical and conceptual problems relating to the validation of assessment centers must be resolved before quantitative validation methods may be used. An open design is suggested to facilitate conceptual clarification. (CH)

Argues for the application of behavioral methods including natural observations over reification in parapsychological research. Research methodology stressing the importance of observation, replication, and continuity in applied science is discussed. (Author/BL)

Outlines costs and benefits associated with different cross-validation strategies; in particular the way in which the study design affects the cost and benefits of different types of cross-validation. Suggests that the choice between empirical estimation methods and formula estimates involves a trade-off between accuracy and simplicity. (JAC)

Over the last decade emphasis has shifted from theoretically based models of practice to empirically based models whose elements are derived from clinical research. These models are defined and a developing model of practice through the use of single-case methodology is examined. Potential impediments to this new role are identified. (Author/BL)

Describes problems in implementing placebo controls in psychotherapy including the difficulty of ensuring that therapists and their patients will view both the experimental and placebo treatments as comparably credible. Considers six research issues stemming from the definitional requirement that the placebo control for the critical and specific…

Considers multiple design issues and strategies raised by comparative outcome studies including the conceptualization, implementation, and evaluation of alternative treatments; assessment of treatment-specific processes and outcomes; and evaluation of the results. (Author/BL)

Author responds to critique by Daniel Griffiths (Educational Administration Quarterly v. 19, p. 201-21) by defending his work in three areas: derivation of hypotheses, testing of hypotheses, and erroneous attribution of assumptions. (JW)

Suggests a framework for structured observation using a psychosocial approach to describe work styles, differences in performance, and outcomes of behavior patterns. (JW)

The need for replication studies is discussed. A statistical procedure is explained for pooling the probabilities of a series of independent studies testing the same hypothesis. (DT)

This paper examines the differences between multilevel modeling and weighted ordinary least squares (OLS) regression for analyzing data from the National Educational Longitudinal Study 1988 (NELS:88). The final sample consisted of 718 students in 298 schools. Eighteen variables from the NELS:88 dataset were used, with the dependent variable being…

Creating a survey that asks the right questions at a level appropriate for the intended audience is a difficult task. This guide is designed to support educators who want to be confident that the data they gather will be useful. The guide is organized according to the developmental steps in creating a survey. Individual chapters correspond to the…

Suggests a possible research strategy for ensuring successful follow-up of respondents in longitudinal surveys. (TO)

Response to article PS 502 099. (MB)