THE RELATIONSHIP BETWEEN CONSUMER PREFERENCES AND TRAINED PANEL RESPONSES

Eight samples of dried tomato soups were profiled by a trained panel, who also provided ratings of overall liking and liking for particular attributes on relative-to-ideal rating scales. A consumer panel tested a subset of four of the soups in home trials, assessing them for overall liking and for particular attributes on either hedonic or relative-to-ideal rating scales. The pattern of overall preferences differed between the trained and untrained panelists, demonstrating the inappropriateness of using trained panelists to provide measures of preference or acceptance. The conclusions regarding consumer preferences would differ depending on which rating scale was used; these differences disappeared when the relative-to-ideal ratings were converted into values of the same form as the hedonic ratings. The overall liking was best predicted by flavor rather than color or thickness. A principal components analysis (PCA) of the profile data compared well with a plot based on the trained panel preference data using MDPREF. The preference data were also fitted to the PCA dimensions using the PREMAP vector model, which gave a good fit for only six of the 15 trained panelists; the PREMAP ideal point model failed to show a better fit. In order to test these models adequately more than eight samples would need to be tested.     

On tree intensity estimation for forest inventories: Some statistical issues

The effect of the plot shape, number of subplots and their spatial arrangement on the sample variance for spatially explicit point populations is analysed for a simple intensity estimator. We derive the sample variance and covariance for sampling designs involving more than one subplot. Some numerical approximations are also presented. If a clustered point pattern has to be sampled, the best strategy to reduce the sample variance is to consider as many rectangular subplots as possible, for a prescribed total sample area, distributed over a grid. In contrast, if a regular point pattern is to be sampled, then a single circular subplot should be considered. If we assume that the point configuration is Poisson, then we can consider any subplot shape and spatial distribution ensuring no overlapping between the subplots. A case study in forestry is considered to assess the validity of our results.     

On the nature of population extremes

Summary Much ecology considers only the typical size of a population, yet extreme values may be of particular importance. Unusually low numbers may doom a population to extinction and unusually high numbers may pose an economic threat. Extreme values may also determine the evolutionary traits that predominate. Obviously, even for a fixed variance in annual numbers, the observed maximum and minimum population size will increase the more years that we count the population. Interestingly, over the time scales of available data (<100 years), most animal populations have an observed variance in annual numbers that increases the more consecutive years we use in its calculation. Consequently, populations will meet extreme values more quickly than if the variance were constant. We quantify the increases in variance for diatoms, insects, and vertebrates, first correcting the data for overall differences in variance. Short- and long-lived species are not consistently different. Species that cycle in density have relatively small increases relative to those that do not cycle. Species in marine ecosystems have larger increases than those in terrestrial and freshwater systems. All these results suggest that the system in which a species is embedded — rather the species' own characteristics — plays the crucial role in determining the nature of population extremes.     

Accounting for tree spatial distribution in a comparison of plot sizes and shapes in dense forest and woodland in Benin (West Africa)

The study examined simultaneously, the effect of tree spatial distribution, inventory plot size and shape on the estimation error of basal area in two contrasting environments. Twenty and fifteen square plots of 1 ha each (divided into 100 quadrats of 0.01 ha) were randomly set in dense forest and woodland, respectively. Thirteen subplots of various shapes and sizes were obtained from the association of adjacent quadrats. Estimation error was calculated using residual mean square of one‐way ANOVA, based on replications of subplot within 1 ha plots. Tree spatial distribution was measured using Green index. Weighted linear regression and mixed effect models were applied to Box & Cox transformed data. In general, the estimation error of basal area decreased with increase in subplot size. However, the effects of tree spatial distribution and plot shape varied with the vegetation type. Where trees tended to be aggregated, estimation error increased with degree of aggregation, and rectangular plots of 0.24 ha produced an acceptable precision. It was concluded that 0.24 ha rectangular plots can be used in tropical environments where the target parameters vary constantly according to one direction, while square plots of the same size are optimal for reliable analysis in case of randomness.     

Detecting spatial patterns in species composition with multiple plot similarity coefficients and singularity measures

Recently, several multiple plot similarity indices have been presented that cure some of the problems associated with the approaches for the calculation of compositional similarity for groups of plots by averaging pairwise similarities. These new indices calculate the similarity between more than two plots whilst considering the species composition on all compared plots. The resulting similarity value is true for the whole group of plots considered (called neighborhood in the following). Here, we review the possibilities for multiple plot similarity calculation and additionally explore coefficients that examine multiple plot similarity between a reference plot (named focal plot in the following) and any number of surrounding plots. The latter represent measures of singularity. Further, we establish a framework for applying these two kinds of multiple plot measures to gridded data including an algorithm for testing the significance of calculated values against random expectations. The capability of multiple plot measures for detecting species compositional gradients and local/regional hotspots within this framework is tested. For this purpose, several artificial data sets with known gradients in species composition (random, gradient, central hotspot, hotspot bottom right) are constructed on the basis of a real data set from a Tundra ecosystem in northern Sweden (Abisko). The coefficients that best reflect the positions of the plots on the realized gradients in species composition are considered as performing best with regard to pattern detection. The tested measures of multiple plot similarity and singularity produced considerably different results when applied to one real and 4 artificial data sets. The newly proposed symmetric singularity coefficient has the best overall performance which makes it suitable for local/regional hotspot detection and for incorporating local to regional similarity analyses in reserve selection procedures.     

MAGNITUDE ESTIMATION WITH AND WITHOUT RE-SCALING

Analysis of variance was applied to 3,107 magnitude estimations (ME) obtained upon 173 panelists (16 panels) evaluating the areas of 18 geometric figures. As Butler et al. (1987) pointed out, if a full ANOVA is applied to the values transformed to logarithms, there is no need to re-scale, i.e., normalize the values as is usually done. In fact, even with re-scaling, total variability is not accounted for adequately unless all the sources of variation are partitioned out in the analysis. Under some circumstances, re-scaling is needed: some laboratories re-scale to the arithmetic mean, others to the geometric one. The error was somewhat greater when the data were re-scaled to the arithmetic mean. The residuals of the ME values transformed to logarithms conformed essentially to a normal distribution. Kurtosis was greater, 5.30; otherwise the distribution was close to a log normal one. The average coefficient of variation was <10%.     

Determining a low disease activity threshold for decision to maintain disease-modifying antirheumatic drug treatment unchanged in rheumatoid arthritis patients

Introduction

The aim of this study was to determine a low disease activity threshold - a 28-joint disease activity score (DAS28) value - for the decision to maintain unchanged disease-modifying antirheumatic drug (DMARD) treatment in rheumatoid arthritis patients, based on expert opinion.

Methods

Nine hundred and sixty-seven case scenarios with various levels for each component of the DAS28 (resulting in a disease activity score between 2 and 3.2) were presented to 44 panelists. For each scenario, panelists had to decide whether or not DMARD treatment (excluding steroids) could be maintained unchanged. In each scenario, for decision, the participants were given the DAS28 parameters, without knowledge of the resultant DAS28. The relationship between panelists' decision, DAS28 value, and components of the score were analysed by multiple logistic regression analysis. Each panelist analysed 160 randomised scenarios. Intra-rater and inter-rater reproducibility were assessed.

Results

Forty-four panelists participated in the study. Inter-panelist agreement was good (κ = 0.63; 95% confidence interval = 0.61 to 0.65). Intra-panelist agreement was excellent (κ = 0.87; 95% confidence interval = 0.82 to 0.92). Quasi-perfect agreement was observed for DAS28 ≤ 2.4, less pronounced between 2.5 and 2.9, and almost no agreement for DAS28 > 3.0. For values below 2.5, panelists agreed to maintain unchanged DMARDs; for values above 2.5, discrepancies occurred more frequently as the DAS28 value increased. Multivariate analysis confirmed the relationship between panelist's decision, DAS28 value and components of the DAS28. Between DAS28 of 2.4 and 3.2, a major determinant for panelists' decision was swollen joint count. Female and public practice physicians decided more often to maintain treatment unchanged.

Conclusions

As a conclusion, panelists suggested that in clinical practice there is no need to change DMARD treatment in rheumatoid arthritis patients with DAS28 ≤ 2.4.     

A COMPARISON OF METHODOLOGY APPLIED TO THE SELECTION OF A PANEL FOR SENSORY EVALUATION OF INSTANT COFFEE

Twenty judges performed a variety of chemosensory tasks in order to select the best scores to form a panel for coffee evaluation. An average of correct responses (P%), one-way analysis of variance (ANOVA) and principal components analysis (PCA) were compared. The tests involved: ability to recognize the four basic tastes, identification and matching of odors, taste intensity evaluation and perception of small differences in taste. P% accounted for 71.17 ± 4.34% and 10 of the judges had scores greater than the final average. ANOVA and PCA resulted in 2 different panels consisting of 9 and 12 judges, respectively. The panel was composed by the nine panelists selected by the three methods. The other three panelists that were doubtful could improve to the point of acceptance with additional training. These methods should be used simultaneously to have more security in the acceptance or rejection of panelists.     

Spatial heterogeneity of ecosystem carbon fluxes in a broadleaved forest in Northern Germany

Carbon fluxes were investigated in a mature deciduous forest, located in Northern Germany (53°47′N–10°36′E), by means of eddy‐covariance technique, stand survey and models. This forest has been managed following a concept of nature‐oriented forestry since the 1980s. One of the goals of the study was to test whether changed management led to increased carbon sequestration. The forest contains several broadleaved tree species. Depending on wind direction, the fetch‐area of the eddy‐covariance data was dominated by different tree species. Three subplots dominated by Oak, Beech or Alder/Ash could be distinguished from the tower data. In each of these subplots, 30 × 30 m² areas were defined to analyse leaf area index, litterfall and the increase of the wood biomass. Eddy‐covariance analysis showed that the gross primary productivity (GPP′) was higher in the Oak subplot (?1794 g C m^?2 yr^?1) in comparison with the Beech plot and the Alder/Ash plot (?1470 and ?1595 g C m^?2 yr^?1, respectively). The total ecosystem respiration (TER) was the highest in the Alder/Ash‐dominated subplot (1401 g C m^?2 yr^?1) followed by the Oak plot and the Beech plot (1235 and 1174 g C m^?2 yr^?1, respectively). The resulting net ecosystem productivity (NEP) was ?559 g C m^?2 yr^?1 for the Oak‐dominated subplot, ?295 g C m^?2 yr^?1 for the Beech plot and ?193 g C m^?2 yr^?1 for the Alder/Ash plot. From Stand survey and modelling, the net primary productivity was estimated as 1103, 702 and 671 g C m^?2 yr^?1 in the Oak, Beech and Alder/Ash plot, respectively. Also carbon flux with litterfall was the highest in the Oak plot 343 g C m^?2 yr^?1 and lowest in Alder/Ash plot (197 g m^?2 yr^?1) with the Beech plot in between (228 g m^?2 yr^?1). The observations indicate an increase of the proportion of litterfall with increasing GPP′ and a different ability of carbon sequestration of the three stands in medium temporary scale. Only in the Oak stand that comprised the oldest trees and the most structured canopy the carbon sequestration was increased compared with conventionally managed forests.     

Heterogeneous genetic structure in a Fagus crenata population in an old-growth beech forest revealed by microsatellite markers

The within-population genetic structure of Fagus crenata in a 4-ha plot (200 x 200 m) of an old-growth beech forest was analysed using microsatellite markers. To assess the genetic structure, Moran's I spatial autocorrelation coefficient was calculated. Correlograms of Moran's I showed significant positive values less than 0.100 for short-distance classes, indicating weak genetic structure. The genetic structure within the population is created by limited seed dispersal, and is probably weakened by overlapping seed shadow, secondary seed dispersal, extensive pollen flow and the thinning process. Genetic structure was detected in a western subplot of 50 x 200 m with immature soils and almost no dwarf bamboos (Sasa spp.), where small and intermediate-sized individuals were distributed in aggregations with high density because of successful regeneration. By contrast, genetic structure was not found in an eastern subplot of the same size with mature soils and Sasa cover, where successful regeneration was prevented, and the density of the small and intermediate-sized individuals was low. Moreover, genetic structure of individuals in a small-size class (diameter at breast height < 12 cm) was more obvious than in a large-size class (diameter at breast height >/= 12 cm). The apparent genetic structure detected in the 4-ha plot was therefore probably the result of the structure in the western portion of the plot and in small and intermediate-sized individuals that successfully regenerated under the favourable environment. The heterogeneity in genetic structure presumably reflects variation in the density that should be affected by differences in regeneration dynamics associated with heterogeneity in environmental conditions.     

PERFORMANCE ASSESSMENT OF SENSORY PANELISTS

The replication variance of individual stimulus evaluations and scale utilization across a panelist's stimulus profile are simultaneously employed to develop statistics for assessing performance of panelists. The approach provides opportunities for comparison of panelists to each other, determination of attributes for which panelist confusion is observed, isolation of stimuli presenting unstable properties, and engagement of influence weights (based on relative precision) in subsequent analyses of data. Although the methodology has been developed for applications involving sensory panelists, the statistical concepts may be extended to other data collection scenarios involving replicated determinations from bounded quantitative measurement scales.     

Reversion of vegetation following the cessation of fertilizer application

Abstract. Tadham Moor in Somerset, England, is an exceptionally rich wetland site which has been mown for hay for many years, with stock grazing the aftermath, but with no history of any fertilizer use. A randomized blocks field experiment (1986–1989) was used to study the effects of five levels of nitrogen input treatments: 0 = control, 25, 50, 100 and 200 kg of N fertilizer per ha per yr. In Phase II of the experiment (1990–1993), each plot was split into two subplots. The allocated fertilizer treatment for the plot was continued in one, randomly selected, subplot but the treatment was discontinued in the other subplot. The experiment not only identified and quantified the changes occurring in the vegetation of hay meadows under different levels of N input, it also provided valuable insight into the dynamics of the sward upon the discontinuance of the treatments. The data for Phase II were used to estimate the time required by the changed vegetation (under different nitrogen treatments) to revert to a state comparable to that prevailing in the control plots. A method for estimating reversion times is described. The main difficulties in estimating the reversion times are identified, the choice of robust vegetation variables being critical. Reversion time estimation methods are presented and used to obtain working estimates for the four nitrogen treatments, applied for 5 yr. These estimates are 3, 5, 7 and 9 yr respectively. The validity of the estimates of 3 yr for the lowest nitrogen input treatment (25 kg /ha/yr) was checked using the available post cessation data.     

Fine‐scale persistence of boreal bog plants

Abstract. Persistence, or the tendency of a species to remain in its original position without colonizing new sites, is studied for 24 species on the ombrotrophic Northern Kisselbergmossen in SE Norway. Data sets comprise presence/absence in 436 sample plots (16 cm x 16 cm) and 6976 subplots (4 cm x 4 cm) analysed with a 5‐yr interval. Persistence was calculated for both spatial scales, and the observed values were compared with null models of completely random presence/absence of species. Species characteristics (plot occurrences and persistence) were related to depth to the water table and species optima along ecologically interpreted DCA ordination axes. The observed persistence was significantly higher than predicted from the random model for all vascular plants and cryptogams at both spatial scales. All taxonomic groups showed higher persistence at the sample plot scale than at the subplot scale. No general relationship between persistence and depth to the water table was found, but for the cryptogams there was somewhat higher persistence for the less peat‐producing species. The persistence of the vascular plants depended on ramet longevity, growth form and vegetative mobility. In general, the observed persistence of most cryptogams was high, probably because of their perenniality, low growth rates and high reproductive output. Differences in growth‐form and life history, as well as the higher number of occurrences, are the most likely reasons for somewhat higher mean persistence of hepatics and Sphagna than of vascular plants at the subplot scale.     

NOTE ON COMPUTERIZED DATA COLLECTION IN CONSUMER SENSORY EVALUATION

Panelists with computer experience (n = 24) and without (n = 26) rated three foods representing liked, neutral and disliked categories using a 9-point hedonic scale on computer and paper ballots. In paired t-test results between computer and ballot ratings, no differences were found in overall acceptance, flavor and appearance attributes. Regression analysis indicated a significant (p<0.01) linear relationship between computer and paper ballot ratings. Computer ballot ratings for overall acceptance were comparable to paper ballot ratings in a consumer sensory evaluation test. Slight differences in overall acceptance were found between computer and paper ballot responses from panelists with and without computer experience. GLM indicated the group without computer experience rated overall acceptance and flavor attributes for cereal, and appearance and flavor attributes for potato chips significantly higher (p<0.05) than panelists with computer experience.     

Multiple‐scale assessment of understory herb species diversity in pine plantations after long‐term restoration

Plantations are established for a variety of purposes including wood production, soil and water conservation, and carbon sequestration. However, their implications for species diversity are considerably debated. To assess restoration effect of species diversity in plantations after 50 years’ natural restoration, understory herb species diversity of Chinese pine (Pinus tabulaeformis) plantations were characterized and compared with secondary growth oak (Quercus wutaishanica) forests by additive partitioning across three different nested spatial scales (subplot, plot and site scales) in the middle of the Loess Plateau, northwestern China. Whether these two forest types demonstrate different structuring processes was also examined by quantifying the relative contributions of space and the environment on community composition. Overall, the two forests displayed similar accumulation of species diversity across spatial scales. The contribution of species diversity components increased with the spatial scale, and both forests displayed lower alpha diversity at the subplot scale but higher beta diversity at the plot and site scales. There was no significant difference in alpha or beta diversity between the two forests at any of the three scales, but in species compositions at the subplot and plot scales (p < 0.05). The diversity of both forest types were simultaneously governed by environmental and spatial processes, with the exception of a greater contribution of each component in the secondary growth oak forests, but both forests were dominated by environmental processes. Our study highlight the significant role of spatial scales in assessing the result of biodiversity restoration of plantations. These results suggest that pine plantations function in a similar manner as secondary growth oak forests for understory herb diversity, albeit with different community compositions.     

CHEESE HARDNESS ASSESSMENT BY EXPERTS AND UNTRAINED JUDGES

Although expert assessment of food characteristics is recognized as a key step in product development, the use of consumer based measurements is sometimes recommended as an equivalent to the experts. From cognitive psychology, support of the role of perceptual learning is found in some instances, although this could not be relevant in others. To address this point performance analysis of experts and untrained panelists in cheese texture evaluation was carried out. Neither the untrained panelists nor the experts were familiar with either the scales or the kind of cheese. The same Cheddar cheese was given to 44 untrained subjects in three trials to assess hardness. The results showed that their judgment has a 29% average random error variance; the interrater reliability being low. The same experiment gave a random error variance of 2% for three highly skilled judges (experts). The difference in variance was linked to training. Untrained panelists also showed an adaptation error. Nevertheless, there was no significant difference between the average ratings of both groups, whether untrained or experts.     

Relationships between temperature responses and bacterial community structure along seasonal and altitudinal gradients

In this study, soil bacterial communities and the temperature responses (Q10) of substrate-induced respiration were compared between an alpine dry meadow and a subalpine forest in the Colorado Rocky Mountains. Bacterial communities in three seasons from each environment were described with 16S rRNA gene clone libraries. The main goal of this comparison was to relate phylogenetic differences among bacterial communities with variation in soil respiratory temperature sensitivities along seasonal and altitudinal gradients. The warmer, lower elevation, subalpine forest soil exhibited large seasonal variations in Q10. Subalpine Q10 values were highest in summer, and were higher than alpine values in all seasons except winter. Q10 in alpine soils were consistently low throughout the year. Alpine and subalpine bacterial communities both varied seasonally, and were markedly distinct from each other. Based on Fst analysis, subalpine communities from colder times of year were more similar to the alpine communities than were subalpine summer communities. Principle component analysis of the pairwise genetic distances (Fst) between communities produced two factors that accounted for 69% and 22% of the total variance in the data set. These factors demonstrated a significant relationship between bacterial community structure and temperature response when regressed on log-transformed Q10 data.