Preference segments: A deeper understanding of consumer acceptance or a serving order effect?

Cluster analysis is used widely to group consumers into homogenous “preference segments” from their patterns of overall liking ratings. In several case studies, cross tabulation of sample frequencies by serving orders reveals the assignment of consumers to segments may be due to the effect of serving order as opposed to genuine differences in acceptance patterns. This effect is defined as the “serving-order-by-segment effect”. To nullify this effect, four adjustment factors were developed and applied to the assessors’ data from six consumer tests. The adjustments differed in two ways: serving position and samples. Regarding serving position, two of the adjustments affected the first-serve responses only, while the other two adjustments affected the responses for all serving positions. Regarding samples, two of the adjustments used the same value to adjust the responses for all of the samples, while the other two adjustments used sample-specific values to adjust the responses for each sample. The number of samples in each of the six consumer tests varied from three to eleven. With the exception of the eleven sample study, cluster analysis on adjusted data yielded different conclusions than for non-adjusted data and results among the four adjustments varied. Serving-order-by-sample interaction plots may be useful in determining the most appropriate adjustment factor.     

The split plot with repeated randomised complete block design can reduce psychological biases in consumer acceptance testing

A randomised complete block design (RCBD) has been extensively used in consumer testing; however, it is ineffective in preventing sample presentation biases. This study demonstrated a more efficient experimental design [i.e. split plot with repeated randomised complete block design (SPRRCBD)] to help minimise positional and first serving‐order (FSO) biases in consumer tests. Results suggested that positional biases were strongly pronounced for the poor‐quality sample. Comparing between RCBD and RCBD without FSO experiments, there were significant differences in overall liking (OL) scores of the same sample served at the same position. Hence, omitting the first sample score from data analysis was not recommended. The mean‐squared error (MSE) of SPRRCBD was lower than RCBD and RCBD without FSO (2.28, 2.72 and 2.60, respectively), indicating a more powerful design to explain variations in mean hedonic differences. SPRRCBD extracted more explained variances, resulting in a decreased Type‐II error in the model.     

A tale of 3 scales: How do the 9-pt,Labeled Affective Magnitude,and unstructured Visual Analog scales differentiate real product sets of fresh berries?

Selection of scales for measuring liking remains an active area of discussion in sensory science, but there is still no real consensus on which scale a sensory scientist working in industry should use to measure actual hedonic responses to food products. Key criteria for a hedonic scale are whether it sensitively and efficiently discriminates among consumer responses to products. Therefore, the key question this study investigates is whether the use of different scales by the same subjects on the same products in a real-world situation would provide different results for the sensory analyst. To address this question, this study comprises the evaluation of 6 for-market varieties each of 4 berries—raspberry, strawberry, blackberry, and blueberry—by the same N = 147 untrained subjects using 3 popular scales—the 9-pt hedonic scale, the Labeled Affective Magnitude Scale (LAM), and the unstructured Visual Analog Scale (VAS). Data were analyzed by mixed-effects ANOVA with subsequent scale-performance and post-hoc measures, and by bootstrapping simulation studies to estimate the empirical power of each scale to detect differences. For each berry type, significant differences in liking were detected by at least one scale, but scale performance differed. The 9-pt scale was the only one of the three to detect differences among the blueberry samples, and examination of ANOVA and post-hoc results for all berries showed that the 9-pt scale consistently discriminated among samples as well or better than the other two scales. In simulation studies, the 9-pt scale showed reliable detection of differences at sample sizes smaller or equal to the other two scales. It is apparent that—in terms of real-world applications and expected detection of differences—the 9-pt scale discriminates consumer liking for different products as well or better than two continuous scales and can be retained in industry research programs.     

Recommendations for use of balanced presentation order of terms in CATA questions

Balancing presentation order of the terms of a CATA question between and within participants has been recommended to minimize the influence of primacy bias on consumer responses and maintain their attention throughout the task. Compared to balancing presentation order only between participants, this experimental factor may make the task more difficult/tedious for participants and shift their attention away from the sensory characterization task with possible detrimental effects for sample discrimination. The aim of the present work was to compare sensory product characterizations obtained using CATA questions in which presentation order of the terms was balanced between participants with those obtained using CATA questions in which presentation order of the terms was balanced both between and within participants. Nine studies with 1028 consumers involving different product categories were conducted. Between-subjects experimental designs were used in all studies. No major differences were found in the sensory product characterizations obtained using presentation orders balanced between participants (BB) and both between and within participants (BW).     

Comparison of pivot profile© to frequency of attribute citation: Analysis of complex products with trained assessors

Pivot© profile (PP), a method which compares samples to a reference (pivot), has shown profiling potential for complex matrices. However, various aspects require further investigation. This study’s aim was to compare PP to frequency of attribute citation (FC) considering individual judges’ data and sample set complexity. A trained panel analysed three wine sets with different within-set product similarity levels. The stability of the PP sensory space was tested by changing the pivot. PP and FC results were compared using RV coefficients. Confidence ellipses on correspondence analysis (CA) plots were constructed to consider individual judges’ data. CA plots constructed from different pivot PP data sets, were less similar to each other, than to CA plots of FC data, for the set with medium and the set with high within-set variation. The most profound differences were observed for the set with the high within-set variation. PP configurations of the set with low within-set variation, were more similar to each other than to FC configurations. Higher explained variance was obtained with PP than FC, but confidence ellipses overlapped more frequently indicating fewer significant differences between samples. PP and FC data were comparable for the set with medium within-set variation. From this study’s results PP is recommended for wine profiling if medium within-set variation between samples exist but not when sample sets with low or high within-set variation are profiled. PP is recommended over FC for comparative studies where a reference sample is required for example during benchmarking or for aging and shelf-life studies.     

Differences Between the Order Model and the Weibull Model in the Modeling of the Enzymatic Browning

This study was focused on a comparison between the first order model and the Weibull model regarding their suitability to fit the enzymatic browning kinetic data in samples of mushrooms, pears, apples, avocados, and bananas (based on the use of the L* luminosity, the ΔE*, and the BI as the browning indexes). In order to obtain statistically robust results, at least 900 points by curve kinetic were used to describe the enzymatic browning. A mean square error was used to measure the suitability of each model. In this study, 60 slices from each sample were analyzed in a computer vision system to calculate L* values. In the results, it was demonstrated that the Weibull model appears to be the best model for mushrooms, pears, apples, avocados, and bananas. According to that expected for a First order kinetic, the β values should be equal to one in the Weibull model. In this study, the β values observed were always less than one. Therefore, it is not possible to assert that the enzymatic browning always follows an empirical First order kinetic. According to the results of this investigation, it is recommended to use the Weibull model first to insure that enzymatic browning kinetic is of the first order.     

Effects of sampling artifacts on occupational samples of diesel particulate matter

Total carbon (TC) is sometimes used to measure or characterize diesel particulate matter (DPM) in occupational settings such as underground mines. DPM samples are collected on quartz fiber filters. When using quartz fiber filters, adsorption of gas phase organic carbon (OC) has been reported, causing a positive bias in the particulate TC results (adsorption artifact). Most of the data on the sampling artifacts and corrections applyto environmental air sampling, where samples are collected at a much higher filter face velocity and the OC concentrations are generally much lower relative to occupational sampling. In this study, we investigated the effects of adsorption artifact on samples from occupational settings. Samples were collected with and without denuders to determine the amount of gas phase OC collected and the accuracy of certain corrections. In underground stone mines, the adsorption artifact was found to positively bias the particulate TC by greater than 20% for filter loadings below 25 microg/cm2 TC (8-h time weighted average = 262 microg/m3). The tandem filter correction reduced the effect of the artifact, as high as 60% of the TC value, to less than 11% for laboratory data. It also significantly reduced the effect of the artifact obtained for field samples.     

Utilization of front-face fluorescence spectroscopy for analysis of process cheese functionality

The purpose of this study was to evaluate the feasibility of front-face fluorescence spectroscopy (FFFS) to predict the meltability of process cheese spreads or products. Twenty-seven commercial samples from 3 manufacturers were used in this study. Each sample was analyzed using dynamic stress rheometry, which was used to calculate the meltability index (temperature at tandelta = 1). Additionally, fluorescence spectra of tryptophan (excitation: 290 nm; emission: 305 to 400 nm) were collected on each sample at 20 degrees C using a front-face accessory. Fluorescence spectrum for each sample consisted of an average of 36 scans (6 scans performed on 6 replicates). The spectral data set consisted of normalized and mean-centered spectra from all the samples. Multivariate statistical analysis was used to correlate spectral data with cheese meltability index as measured by dynamic stress rheometry. A prediction model was developed using partial least square regression and was calibrated using a cross-validation method. A correlation coefficient of 0.93 was obtained between fluorescence spectra and cheese meltability. The regions 335 to 350 nm and 385 to 400 nm had the highest correlation to cheese meltability. A negative correlation between the peak height of tryptophan (335 to 350 nm) and cheese meltability index was observed. This correlation may be due to presence of tryptophan residues in a more hydrophobic environment in stronger emulsions as compared with a more polar environment in weak emulsions. These results indicate that the melt properties of process cheese spreads or products are related to molecular structure that can be measured using FFFS. Hence, FFFS can be used as an analysis technique to predict process cheese meltability.     

Stability of sample configurations from projective mapping: How many consumers are necessary?

Projective mapping for sensory characterisation with consumers has been used for a relatively short period of time, which suggests that the development of guidelines regarding best practices is strongly needed. The present work aims to provide an insight on the minimum number of consumers needed to reach stable sample configurations. Data sets from 21 different consumer studies, differing in product category, number of samples and degree of difference among them, were used to evaluate the influence of the number of consumers on the stability of sample configurations by means of a resampling approach. For each study, 1000 random subsets of different number of consumers were generated from the original data set. For each virtual panel, sample configurations were obtained using Multiple Factor Analysis. The agreement between them and the reference configurations (obtained with all the consumers) was evaluated through the RV coefficient, using the first two and the first four dimensions of the MFA. Results showed that the stability of sample configuration clearly depended on the degree of difference and type of differences among samples and the number of samples in the dataset. Across the 21 data sets analysed, results suggested that when working with widely different samples, 50 consumers seems as a safe recommendation of minimum consumer panel size to obtain reliable results with projective mapping. However, after any characterisation by projective mapping is completed, it is highly recommended to check, a posteriori, the reliability of the sample space configuration using a bootstrapping procedure.     

大波浪斜裙的裙摆下垂问题研究

大波浪斜裙在穿着过程中会产生裙摆下垂的问题，影响裙子的穿着效果。因此在打样时必须根据面料的性能及斜裙的造型来对样板的裙摆进行修正。文章选用常见面料为样本，对大波浪斜裙的裙摆下垂问题进行了研究，指出：对于同一种大波浪斜裙款型，不同的面料所制成的裙片样板是不同的；对于同一种面料，不同的斜裙款型所制成的裙片样板也是不同的。     

不同菌株对发酵大豆后熟期间品质特性的影响

为研究利用细菌和霉菌进行混合发酵的不同效果,开发新的混菌发酵工艺,本研究采用不同接种方式,得到3个单一菌株发酵样品,4个混合菌株发酵的样品,研究不同菌株对发酵大豆品质特性的影响。按照韩国清麴酱的方法,利用Bacillus subtilis菌株发酵制得样品B,按照豆豉的发酵方法利用Aspergillus oryzae和Mucor racemosus菌株制得样品A和M。混合菌株发酵样品中依据两种菌株接种的顺序不同,分别制得AB、BA、MB和BM。结果表明,后熟0~9 d期间,样品B的蛋白酶活性最高(257.32±3.04)U/g;样品M的还原糖在后熟第1 d达到1.17%,高于其他样品。样品A和MB氨基态氮含量较高,样品B中测得铵态氮含量最高。所有样品的pH呈下降趋势,而滴定酸度变化范围则相反,从0.02%上升到0.19%。在后熟9 d期间,各样品色度中的L*值呈持续下降趋势,样品MB褐变现象明显。在7个实验组中,样品MB通过混菌发酵工艺提高了发酵大豆的品质特性,验证了利用细菌和霉菌进行混菌发酵技术能够改善发酵大豆的品质特性。     

Hedonic rating of food: single or side-by-side sample presentation?

Taste panels made hedonic ratings of the sensory properties of four grades of sultanas, evaluated as two pairs (Experiment 1), three chicken casserole products (Experiment 2), and two cola drinks (Experiment 3). In each experiment, two modes of sample presentation were compared: single presentation, where panellists rated only one sample per session; and side-by-side presentation, where the samples were presented simultaneously. In Experiment 1, hedonic ratings from single presentation suggested there were no significant differences between grades, whereas the side-by-side presentation did indicate significant differences. This discrepancy between modes of presentation was not noted in Experiments 2 and 3. Selective operation of methodological bias in category rating might possibly account for the findings.     

Consumer sensory evaluation of plain and seasoned goat meat and beef products

Plain meat loaves and chili were prepared with ground goat meat or beef containing 15% fat. A consumer sensory panel first rated meat loaf samples for acceptability on a 9-point hedonic scale. Then, meat loaf samples, followed by chili samples, were each evaluated in triangle test for differentiation between goat meat and beef. Acceptability scores were similar for plain goat meat and beef loaves when the goat meat product was served before the beef product, but were lower for the goat meat with reverse serving order. In triangle tests, goat meat was differentiated from beef, whether plain or seasoned products.     

离子色谱法检测硝酸盐亚硝酸盐前处理方法的研究

目的 利用离子色谱法检测乳及乳制品中亚硝酸盐、硝酸盐,在实际检测中用该国标方法在前处理样品时,存在沉淀样品不完全进而影响检测准确性、重复性不理想的弊端,为提高检测的准确性,缩小检测偏差,需对国标方法中样品的前处理改进。方法 样品经乙腈沉淀脂肪蛋白质后,采用相应的方法提取和净化,以氢氧化钾溶液为淋洗液,阴离子交换柱分离,电导检测器检测。以保留时间定性,外标法定量。结果 该方法克服了乳及乳制品GB5009.33-2010第三法检测检出限高及检测用时长,精密度低的的弊端。结论本方法适合于乳及乳制品定量测定。     

基于非靶向代谢组学分析副干酪乳杆菌发酵枸杞汁各阶段代谢差异

为研究副干酪乳杆菌NXU-19004发酵对枸杞汁品质特性的影响,采用气相色谱-质谱非靶向代谢组学技术对4 种不同阶段枸杞发酵液差异代谢物进行鉴定分析,结合主成分分析和偏最小二乘判别分析对其进行精确区分。再通过京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes,KEGG）数据库注释差异代谢物参与的代谢途径,从中筛选出关键代谢物。结果表明,4 种发酵液共检测到63 种非挥发性差异代谢物,其中氨基酸类、有机酸类、酯类等成分在枸杞发酵液中平均相对丰度总体呈上升趋势。多元统计分析结果显示,各样品之间差异显著,均有明显样本聚集区,表明代谢产物可用于甄别发酵阶段。共注释到47 条代谢途径,富集代谢产物较多且显著性高的关键通路有9 条,且多与氨基酸代谢相关,可见发酵阶段对枸杞发酵液中氨基酸类物质的代谢影响最大。进一步确认了关键代谢物为丙氨酸、柠檬烯、苯甲醛、苯甲酸、烟酰胺、烟酸、角鲨烯、脯氨酸、肌氨酸、2-苯基乙酰胺。该研究从代谢组学角度初步揭示了枸杞及其乳酸菌发酵产品的代谢产物差异性以及特征化合物,为进一步研究影响乳酸菌发酵形成差异的因素提供理论依据。     

Cyclic presentation order designs for consumer research

Williams Latin square designs are used widely in consumer research to plan complete block experiments. It is demonstrated how these designs are a subset of a wider class of cyclic designs and how alternative designs, for odd numbers of samples p, are preferable both to obtain positional and carry-over balance at twice as many values of n the number of consumers. A consequence is that the new designs give better robustness to the variable number of consumers who turn up at a central location test. An algorithm is described that is used to sample from the wider class of designs and is sufficiently fast to be used for the interactive construction of designs within a computer package. Other specialist designs for situations where there may be factorial structure among the consumers are also described. Tables are given that list the first rows of cyclic designs generated by the algorithm, these give pairs, as well as triples of balanced Latin squares for between five and 19 products. Practical advice is given on how they may be used to construct suitable presentation order designs.     

Preliminary investigation on the balance of lead and cadmium content in milk and its by-products

Many studies have recently been devoted to the determination of lead and cadmium in milk, in view of the toxicity of these metals and the importance of milk in the diet of infants and children. Enrichment of lead and cadmium in fatty dairy products has been sometimes noted. A systematic study on milk and its primary products (skimmed milk, cream and cheese) obtained from the same batch of milk is considered in order to define a tentative balance of these two metals. Lead and cadmium were determined by atomic-absorption spectrophotometry with graphite furnace on a solution of the ashed sample. As a consequence of the appreciable interference due to the matrix effects, the method of standard additions was adopted to determine the metals in few samples with a statistically acceptable degree of accuracy and precision. The range of the lead and cadmium content found in these products is comparable with recent literature data concerning uncontaminated samples.     

Survey of undeclared soy allergen levels in the most frequently recalled food categories with or without precautionary labelling

A comprehensive study was designed to determine the frequency and levels of soy allergen in packaged bakery and snack food products. A representative sample of products with no soy allergen disclosed on the label was analysed using two widely used enzyme-linked immunosorbent assay (ELISA) methods. Samples were chosen that either had no soy identified on the product label or which had a soy precautionary statement. Among 558 bakery and snack products, soy protein was detected in 17% of the products using the Neogen (NE) kit and 11% of the products using the Elisa Systems (ES) kit. The disagreement rates between kits were 8.8% for bakery products and 3.3% for snack products. Overall soy protein was detected at higher frequency in bakery products than in snack foods. Among 284 bakery samples, soy protein was detected in 25% of the samples with no precautionary statement and 19% of the samples which had a precautionary statement. Among 274 snack samples, soy protein was detected in 11% of the samples with no precautionary statement and 9% of the samples which had a precautionary statement. The sample repeatability was at an acceptable level (< 9%) for each method and food commodity. The reproducibility between kits was 23% for bakery foods and 36% for snack foods. None of the bakery (21) and snack (6) products without precautionary labelling (measured level > 5 ppm) had a higher level of soy protein per serving compared with the eliciting dose₁₀ (ED₁₀) of 10.6 mg for soy allergic patients. But the level of soy protein per serving may be clinically relevant to a subpopulation of soy allergic patients if a more stringent eliciting dose is applied. These findings emphasise that suitable detection methodologies and references doses are crucial for labelling accuracy and the safety of soy allergic consumers.     

Evaluation of a conductance method for enumerating Escherichia coli on chicken, pork, fish, beef, and milk

Experiments were conducted to evaluate a rapid method for enumerating Escherichia coli on food products of animal origin. In study I, rinses from samples of chicken, ground beef, pork, and fish and samples of milk were inoculated with various levels of actively growing E. coli. Conductance assays were monitored at 44 degrees C on each sample using coliform medium supplemented with 2% dextrose. High correlations between E. coli concentrations and E. coli conductance detection times (ECDTs) were found (r = -0.97 to -0.99) for all foods tested in all replicates; however, in most cases, the concordance correlation coefficients (r(c)) were low, indicating a lack of predictive accuracy. In this study, low accuracy of the conductance method for estimating E. coli counts was attributed to use of concentrations of E. coli that exceed 10(6) CFU/ml, the detection threshold of the instrument. Slopes of the linear regression lines (E. coli concentration vs. ECDT) for each type of food tested were not significantly different (P < 0.0001), indicating that a single regression equation may be used to estimate E. coli counts for all of the types of food tested in 1 to 7.5 hours using ECDT. In study II, ECDTs for pork, fish, beef, and milk significantly (P < 0.05) decreased in a linear manner as time of temperature abuse increased. Although the ECDT for chicken decreased linearly, no significant differences were observed between 3 and 6 or between 9 and 12 h of abuse. These data demonstrate a strong relationship between increasing populations of E. coli due to temperature abuse and decreasing ECDT. Therefore, results from both studies indicate that this method could be useful for estimating naturally occurring populations of E. coli on foods of animal origin.     

Environmental consequences of water pumped from greater New Orleans following Hurricane Katrina: chemical, toxicological, and infaunal analysis

The U.S. Army Engineer Research and Development Center Environmental Laboratory, Vicksburg, MS, conducted a study to determine the extent to which Hurricane Katrina floodwaters in the New Orleans, Louisiana area may have had impacts on wildlife habitat and other biological resources in surrounding areas. These studies were conducted as part of the Interagency Performance Evaluation Taskforce, an investigation of environmental impacts originating from the failure of the hurricane protection system during Hurricane Katrina. This paper presents data regarding the effects of pumped floodwaters on sediment chemistry, toxicity, and benthic invertebrate assemblages near pumping stations that discharged floodwaters into marshes near Chalmette and Violet, Louisiana. Chemical contamination of sediments was observed and varied among sample locations (e.g., outfall locations, wastewater treatment plant, canals, and wetlands); however, trends in the chemistry data were not always consistent with bioassay results. A comparison of the sediment chemistry data from this study with three other studies reporting concentrations of chemicals in sediments within the city of New Orleans suggested that sediments and associated contaminants present within the levees were not pumped into the marsh in appreciable quantities.