The appropriateness of using statistical data to inform the design of any given service development or initiative often depends upon judgements regarding scale. Large-scale data sets, perhaps national in scope, whilst potentially important in informing the design, implementation and roll-out of experimental initiatives, will often remain unused…

Meta-analysis does not solve all the problems encountered in qualitative research reviews, but it holds out some hope for increasing the reliability and dependability of a reviewer's conclusions. Several developments in meta-analysis are cause for optimism. First, different meta-analysts are doing work in the same areas, leading to increased…

Discusses criticisms of null hypothesis significance testing (NHST), suggesting that historical use of NHST was reasonable, and current users should read Sir Ronald Fisher's applied work. Notes that modifications to NHST and interpretations of its outcomes might better suit the needs of modern science. Concludes that NHST is most often useful as…

Several developments in the use of the new method of meta-analysis give cause for optimism. First, different meta-analysts are doing work in the same areas, leading to increased confidence in meta-analytic results. Second, meta-analysts are beginning to include raw data in their reports, which helps readers pinpoint the exact studies that lead to…

Secondary analysis of datasets can extend previous work in the field, provide a greater quantity of material, and has the advantages of speed and lower cost. More use of secondary data could improve the low esteem of educational research. (Contains 15 references.) (SK)

This document reviews "The Collected Works of George E. P. Box," edited by G. C. Tiao. The two-volume collection is divided into five parts. Each part is prefaced with an introduction by a statistical researcher giving his view of the motivation and highlights of the papers presented. The 16 papers in Part 1 on statistical inference,…

Recommends that analysts (1) use conventional meta-analytic statistics when testing for moderator variables; (2) perform tests between mean effect sizes; and (3) continue to perform model tests in meta-analyses for which study outcomes are already consistent if they have theoretical expectations about moderators. (RS)

Criticizes the accuracy of the methodology currently used by the Australian Department of Education, Training and Youth Affairs for identifying socioeconomic status of higher education students. Proposes new individually based measures related to the occupation and education of parents when the student was in high school. (Author/DB)

Methods for selecting relevant and appropriate quasi-experimental studies for inclusion in research synthesis using the threats-to-validity approach are presented. Effects of including and excluding studies are evaluated. (BS)

Several problems in reporting and presentation of Higher Education General Information Survey (HEGIS) data that might affect data comparability are outlined and recommendations of a HEGIS study group are presented. Suggestions range from increased institutional involvement and increased feedback to institutions, to improvements in data storage and…

Argues that the application of Bradford's original formulation to Leimkuhler's choice of bibliographic citations more closely fits the statistical data than does his own. Operational research techniques are assailed as inadequate and wasteful to the needs and purposes of the social sciences. Six references are provided. (RAA)

Examines the importance of defining terms in the process of gathering statistical data. Cites the example of the debate over the number of children abducted by strangers. Concludes that statistics are part of the persuasive rhetoric of social problem promotion and have social consequences. (FMW)

In part stimulated by the information needed by the Private Industry Councils (PICs) under the Comprehensive Employment and Training Amendments (1978), a growing number of groups and agencies have conducted surveys of local employers to gather various types of occupational data. Such activities may be useful and productive, if the data collection…

The planning and management of educational systems has been examined widely in the last decades despite the lack of a standard research method and a unified underlying theory. Research based on quantitative analysis yields data that may be processed using the graphic capabilities of computers. The combination of numerical data and computer…

During the spring of 1980, 59,000 students from over 1,000 public and private high schools were surveyed to obtain base-year data for a projected longitudinal study entitled "High School and Beyond." Intended to provide background information for making policy decisions, the study focused on four specific policy concerns: discipline,…