An improved quantitative science would emphasize the use of confidence intervals (CIs), and especially CIs for effect sizes. This article reviews some definitions and issues related to developing these intervals. Confidence intervals for effect sizes are especially valuable because they facilitate meta-analytic thinking and the interpretation of intervals via comparison with the effect intervals from related prior studies. Several recommendations for the thoughtful use of such CIs are presented.

/pdf/what-future-quantitative-social-science-research-could-look-49rajut1qj.pdf

What Future Quantitative Social Science Research Could Look Like: Confidence Intervals for Effect Sizes

Reform of statistical practice in the social and behavioral sciences requires wider use of confidence intervals (CIs), effect size measures, and meta-analysis. The authors discuss four reasons for promoting use of CIs: They (a) are readily interpretable, (b) are linked to familiar statistical significance tests, (c) can encourage meta-analytic thinking, and (d) give information about precision. The authors discuss calculation of CIs for a basic standardized effect size measure, Cohen’s δ (also known as Cohen’s d), and contrast these with the familiar CIs for original score means. CIs for δ require use of noncentral t distributions, which the authors apply also to statistical power and simple meta-analysis of standardized effect sizes. They provide the ESCI graphical software, which runs under Microsoft Excel, to illustrate the discussion. Wider use of CIs for δ and other effect size measures should help promote highly desirable reform of statistical practice in the social sciences.

A primer on the understanding, use, and calculation of confidence intervals that are based on central and noncentral distributions

Aspects of the reproductive performance over the last decade of Black-Browed, Grey-Headed and Wandering Albatrosses, Gentoo and Macaroni Penguins and Antarctic Fur Seals, at Bird Island, South Georgia and for Adelie and Chinstrap Penguins at Signy Island, South Orkney Islands, are summarized and reviewed. Breeding success of the Wandering Albatross, which breeds in winter and eats fish and squid, has remained constant, while population size has declined gradually but significantly. The other species at South Georgia, which breed in summer and feed extensively on krill, have shown major fluctuations in some or all of: breeding population size, breeding success, foraging trip duration and offspring growth rate. 1977–78 and 1983–84 were summers of particularly poor reproductive performance by almost all species; circumstantial evidence relating this to reduced availability of krill is discussed. The fluctuations in reproductive performance of the krill-eating, summer-breeding penguins at Signy Island are not synchronized with those at South Georgia; they correlate best (especially for Chinstraps, which suffered badly in 1980–81 and 1982–83) with the date of ice break-out in late spring. Numerous parameters of albatross, penguin and fur seal biology are reviewed in terms of their sensitivity and suitability for detecting changes in the marine environment.

Reproductive Performance of Seabirds and Seals at South Georgia and Signy Island, South Orkney Islands, 1976–1987: Implications for Southern Ocean Monitoring Studies

The advantages that confidence intervals have over null-hypothesis significance testing have been presented on many occasions to researchers in psychology. This article provides a practical introduction to methods of constructing confidence intervals for multiple and partial R2 and related parameters in multiple regression models based on “noncentral”F and χ2 distributions. Until recently, these techniques have not been widely available due to their neglect in popular statistical textbooks and software. These difficulties are addressed here via freely available SPSS scripts and software and illustrations of their use. The article concludes with discussions of implications for the interpretation of findings in terms of noncentral confidence intervals, alternative measures of effect size, the relationship between noncentral confidence intervals and power analysis, and the design of studies.

Correct Confidence Intervals for Various Regression Effect Sizes and Parameters: The Importance of Noncentral Distributions in Computing Intervals.

The origins of agriculture in the Near East has been associated with a 'core area', located in south-eastern Turkey, in which all major crops were brought into domestication within the same local domestication system operated by a single cultural group. Such an origin leads to a scenario of rapid invention of agriculture by a select cultural group and typically monophyletic origins for most crops. Surprisingly, support for a core area has never been directly tested with archaeological evidence. Over the past decade a large amount of new archaeological and genetic evidence has been discovered which brings new light on the origins of agriculture. In this review, this new evidence was brought together in order to evaluate whether a core region of origin is supported. Evidence shows that origins began earlier than previously assumed, and included 'false starts' and dead ends that involved many more species than the typical eight founder crops associated with the core area. The rates at which domestication syndrome traits became fixed were generally slow, rather than rapid, and occurred over a geographically wide range that included the North and South Levant as well as the core area. Finally, a survey of the estimated ages of archaeological sites and the onset of domestication indicates that the domestication process was ongoing in parallel outside of the core area earlier than within it. Overall, evidence suggests a scenario in which crops were domesticated slowly in different locations around the Near East rather than emanating from a core area.

M. G. Kendall

Papers

Biometrika Tables for Statisticians, Vol. 2