An old problem in personnel psychology is to characterize distributions of test validity correlation coefficients. The proposed model views histograms of correlation coefficients as observations from a mixture distribution which, for a fixed sample size n, is a conditional mixture distribution h(r|n) = Σjλjh(r; ρj, n), where R is the correlation coefficient, ρj are population correlation coefficients and λj are the mixing weights. The associated marginal distribution of R is regarded as the parent distribution underlying histograms of empirical correlation coefficients. Maximum likelihood estimates of the parameters ρj and λj can be obtained with an EM algorithm solution and tests for the number of components t are achieved after the (one-component) density of R is replaced with a tractable modeling density h(r; ρj, n). Two illustrative examples are provided.

Corrections for restriction in range due to explicit selection assume the linearity of regression and homoscedastic array variances. This paper develops a theoretical framework in which the effects of some common forms of violation of these assumptions on the estimation of the unrestricted correlation can be investigated. Simple expressions are derived for both the restricted and corrected correlations in terms of the target (unrestricted) correlation in these situations.

Rosa x odorata (sect. Chinenses, Rosaceae) is an important species distributed only in Yunnan Province, China. There is an abundance of wild variation within the species. Using 22 germplasm resources collected from the wild, as well as R. chinensis var. spontanea, R. chinensis ‘Old Blush’ and R. lucidissima, this study involved morphological variation analysis, inter-trait correlation analysis, principal component analysis and clustering analysis based on 16 morphological traits. This study identified a high degree of morphological diversity in R. x odorata germplasm resources and the variation coefficients had a distribution range from 18.00 to 184.04%. The flower colour had the highest degree of variation, while leaflet length/width had the lowest degree of variation. Inter-trait correlation analysis revealed that there was an extremely significant positive correlation between leaflet length and leaflet width. There was also a significant positive correlation between the number of petals and duration of blooming, and the L* and a* values of flower colour were significantly negatively correlated. Principal component analysis screened five principal components with the highest cumulative contribution rate (81.679%) to population variance. Among the 16 morphological traits, style length, sepal width, flower diameter, flower colour, leaflet length and leaflet width were important indices that influenced the morphology of R. x odorata. This study offers guidance for the further development and utilization of R. x odorata germplasm resources.

We evaluated the performance of an FFQ in estimating phytosterol intake against multiple 24-h dietary recalls (24HDR) using data from 1011 participants of the calibration sub-study of the Adventist Health Study-2 (AHS-2) cohort. Dietary assessments of phytosterol intake included a self-administered FFQ and six 24HDR and plasma sterols. Plasma sterols were determined using the GLC flame ionisation method. Validation of energy-adjusted phytosterol intake from the FFQ with 24HDR was conducted by calculating crude, unadjusted, partial and de-attenuated correlation coefficients (r) and cross-classification by race. On average, total phytosterol intake from the FFQ was 439·6 mg/d in blacks and 417·9 mg/d in whites. From the 24HDR, these were 295·6 mg/d in blacks and 351·4 mg/d in whites. Intake estimates of β-sitosterol, stigmasterol, other plant sterols and total phytosterols from the FFQ had moderate to strong correlations with estimates from 24HDR (r 0·41–0·73). Correlations were slightly higher in whites (r 0·42–0·73) than in blacks (r 0·41–0·67). FFQ estimates were poorly correlated with plasma sterols as well as 24HDR v. plasma sterols. We conclude that the AHS-2 FFQ provided reasonable estimates of phytosterol intake and may be used in future studies relating phytosterol intake and disease outcomes.

Eighty-six chickpea (Cicer arietinum L.) genotypes, including 44 Kabuli type and 42 Desi type, were evaluated for their phenological, physicochemical and cooking quality traits. There were significant differences among the genotypes for days to 50% flowering (34–81 d), days to maturity (85–122 d), number of pods per plant (13–66), number of seeds per plant (15–85), 100-seed weight (10.5–58.6 g), seed yield (561–1852 kg/ha), hydration capacity (0.11–0.68 g water/seed), hydration index (0.80–1.21), swelling capacity (0.11–0.7 ml/seed), seed volume (0.1–0.52 ml/seed) and cooking time (38–125 min). The Desi and Kabuli types of chickpea differed significantly from each other for all the traits except for hydration index, swelling index and cooking time. High heritability coupled with high genetic advance was recorded for 100-seed weight, hydration capacity, swelling capacity and seed volume in both Desi and Kabuli genotypes. Seed size (100-seed weight and seed volume) showed significant positive correlations with hydration capacity and swelling capacity. Cooking time did not show any significant positive or negative correlation with any of the traits studied, including seed size, indicating that other additional factors may be involved in controlling cooking time. The results of this study indicate that it is possible to develop cultivars with faster cooking time in both Kabuli and Desi types and in all seed size categories.

The potential yield levels of sorghum, an important crop in the semi-arid regions of the world, are stagnating due to the narrow genetic base of the genotypes utilized in the sorghum improvement programmes worldwide. For utilization of genetic diversity in any crop, adequate genetic characterization of relevant germplasm is a prerequisite. Therefore, we attempted to identify the potential germplasm lines for improvement of yield in a set of 135 germplasm lines belonging to different races. The experiment was carried out in an augmented design during 2002, and in Randomized complete block design (RCBD) for selected lines during 2003. The correlation studies indicated that besides panicle weight, the number of grains/unit length of primary branch and the number of secondary branches/primary branch had significant positive relation with grain yield. IS nos 1259, 5447, 5725 and 2036 were promising for more than one yield component. The study indicated that besides caudatum race, the lines belonging to guinea, kafir and inter-races with them are good sources for different yield components. Genetic diversity and cluster analysis grouped 40 germplasm lines into eight clusters, with maximum inter-cluster distance between clusters II and III. Utilization of the germplasm lines belonging to different clusters in improving yield components is discussed.

This study investigates whether pelagic fish schools of different species or groupings (e.g. herring, "surface herring", "gadoids", mackerel, sprat and sandeels) were positively or negatively associated with each other in time and space. To do this, statistical models were fitted to pre-processed acoustic fisheries data to reveal how pelagic school prevalence varied with respect to spatial (latitude and longitude) and temporal (time of day) information. The model outputs, which take the form of probabilities fitted to the presence or absence of schools, were then used to calculate correlation coefficients, which are useful for measuring association between pairs of variables. Results depended upon the specific species pairs under investigation. Herring and "surface herring" were, for example, very generally sympatrically associated with each other in both space and time, while herring and gadoid schools, on the other hand, had allopatric distributions.