Color, Flavor, and Texture: Which Blackberry Sensory. Attribute is the Most Important to Consumers?

Similar documents
Primocane Fruiting Blackberry Trial Results

Introduction to Workshop (Eric Stafne, Assistant Professor and Fruit Crops Specialist)

Gasoline Empirical Analysis: Competition Bureau March 2005

COMPARISON OF THREE METHODOLOGIES TO IDENTIFY DRIVERS OF LIKING OF MILK DESSERTS

Development of Value Added Products From Home-Grown Lychee

The premium for organic wines

Gail E. Potter, Timo Smieszek, and Kerstin Sailer. April 24, 2015

MBA 503 Final Project Guidelines and Rubric

A Comparison of X, Y, and Boomer Generation Wine Consumers in California

AJAE Appendix: Testing Household-Specific Explanations for the Inverse Productivity Relationship

Tofu is a high protein food made from soybeans that are usually sold as a block of

Ex-Ante Analysis of the Demand for new value added pulse products: A

An application of cumulative prospect theory to travel time variability

Citrus Attributes: Do Consumers Really Care Only About Seeds? Lisa A. House 1 and Zhifeng Gao

Results from the First North Carolina Wine Industry Tracker Survey

2016 STATUS SUMMARY VINEYARDS AND WINERIES OF MINNESOTA

Varietal Specific Barrel Profiles

Measuring economic value of whale conservation

The University of Georgia

Investment Wines. - Risk Analysis. Prepared by: Michael Shortell & Adiam Woldetensae Date: 06/09/2015

Genotype influence on sensory quality of roast sweet pepper (Capsicum annuum L.)

Forestry, Leduc, AB, T9E 7C5, Canada. Agriculture/Forestry Centre, Edmonton, AB T6G 2P5, Canada. *

Labor Supply of Married Couples in the Formal and Informal Sectors in Thailand

wine 1 wine 2 wine 3 person person person person person

Comparative Analysis of Fresh and Dried Fish Consumption in Ondo State, Nigeria

Emerging Local Food Systems in the Caribbean and Southern USA July 6, 2014

Flexible Working Arrangements, Collaboration, ICT and Innovation

COMPARISON OF CORE AND PEEL SAMPLING METHODS FOR DRY MATTER MEASUREMENT IN HASS AVOCADO FRUIT

2011 Regional Wine Grape Marketing and Price Outlook

RESEARCH UPDATE from Texas Wine Marketing Research Institute by Natalia Kolyesnikova, PhD Tim Dodd, PhD THANK YOU SPONSORS

Session 4: Managing seasonal production challenges. Relationships between harvest time and wine composition in Cabernet Sauvignon.

Sensory Quality Measurements

Blackberry Variety Development and Crop Growing Systems. John R. Clark University Professor of Horticulture

The Effects of Dried Beer Extract in the Making of Bread. Josh Beedle and Tanya Racke FN 453

The Role of Calorie Content, Menu Items, and Health Beliefs on the School Lunch Perceived Health Rating

AWRI Refrigeration Demand Calculator

Wine-Tasting by Numbers: Using Binary Logistic Regression to Reveal the Preferences of Experts

Multiple Imputation for Missing Data in KLoSA

Previous analysis of Syrah

2016 China Dry Bean Historical production And Estimated planting intentions Analysis

DEVELOPMENT AND SENSORY EVALUATION OF READY-TO- COOK IDLI MIX FROM BROWNTOP MILLET (Panicum ramosa)

Relationships Among Wine Prices, Ratings, Advertising, and Production: Examining a Giffen Good

Survival of the Fittest: The Impact of Eco-certification on the Performance of German Wineries. Patrizia Fanasch University of Paderborn, Germany

A Hedonic Analysis of Retail Italian Vinegars. Summary. The Model. Vinegar. Methodology. Survey. Results. Concluding remarks.

Missing Data Treatments

ORGANOLEPTIC EVALUATION OF RECIPES BASED ON DIFFERENT VARIETIES OF MAIZE

Appendix A. Table A.1: Logit Estimates for Elasticities

Transportation demand management in a deprived territory: A case study in the North of France

Academic Year 2014/2015 Assessment Report. Bachelor of Science in Viticulture, Department of Viticulture and Enology

The Effect of Blackstrap Molasses on Cookies. 11/21/2011 FN 453 Written Report Hannah Abels, Shane Clingenpeel and Jennifer Smith

BLUEBERRY MUFFIN APPLICATION RESEARCH COMPARING THE FUNCTIONALITY OF EGGS TO EGG REPLACERS IN BLUEBERRY MUFFIN FORMULATIONS RESEARCH SUMMARY

THE EFFECT OF DIFFERENT APPLICATIONS ON FRUIT YIELD CHARACTERISTICS OF STRAWBERRIES CULTIVATED UNDER VAN ECOLOGICAL CONDITION ABSTRACT

Pasta Market in Italy to Market Size, Development, and Forecasts

Statistics & Agric.Economics Deptt., Tocklai Experimental Station, Tea Research Association, Jorhat , Assam. ABSTRACT

Online Appendix to The Effect of Liquidity on Governance

Effects of Acai Berry on Oatmeal Cookies

2. Materials and methods. 1. Introduction. Abstract

Is Fair Trade Fair? ARKANSAS C3 TEACHERS HUB. 9-12th Grade Economics Inquiry. Supporting Questions

JCAST. Department of Viticulture and Enology, B.S. in Viticulture

Internet Appendix for CEO Personal Risk-taking and Corporate Policies TABLE IA.1 Pilot CEOs and Firm Risk (Controlling for High Performance Pay)

SPONGE CAKE APPLICATION RESEARCH COMPARING THE FUNCTIONALITY OF EGGS TO EGG REPLACERS IN SPONGE CAKE FORMULATIONS RESEARCH SUMMARY

Predicting Wine Quality

Dietary Diversity in Urban and Rural China: An Endogenous Variety Approach

Valuing Health Risk Reductions from Air Quality Improvement: Evidence from a New Discrete Choice Experiment (DCE) in China

The Roles of Social Media and Expert Reviews in the Market for High-End Goods: An Example Using Bordeaux and California Wines

Economic Contributions of the Florida Citrus Industry in and for Reduced Production

To make wine, to sell the grapes or to deliver them to a cooperative: determinants of the allocation of the grapes

1. Continuing the development and validation of mobile sensors. 3. Identifying and establishing variable rate management field trials

Supply & Demand for Lake County Wine Grapes. Christian Miller Lake County MOMENTUM April 13, 2015

Consumer preferences for organic and welfare labelled meat A natural field experiment conducted in a high class restaurant

Non-Allergenic Egg Substitutes in Muffins

CONTEMPORARY RESEARCH IN INDIA (ISSN ): VOL. 7: ISSUE: 2 (2017)

Zeitschrift für Soziologie, Jg., Heft 5, 2015, Online- Anhang

Mobility tools and use: Accessibility s role in Switzerland

Awareness, Attitude & Usage Study Executive Summary

(A report prepared for Milk SA)

Effect of Breed on Palatability of Dry-Cured Ham. S.J. Wells, S.J. Moeller, H.N. Zerby, K.M. Irvin

BORDEAUX WINE VINTAGE QUALITY AND THE WEATHER ECONOMETRIC ANALYSIS

Gender and Firm-size: Evidence from Africa

This appendix tabulates results summarized in Section IV of our paper, and also reports the results of additional tests.

Peach festival consumer insights of white peaches. Dr. Amy Bowen

F&N 453 Project Written Report. TITLE: Effect of wheat germ substituted for 10%, 20%, and 30% of all purpose flour by

Volume 30, Issue 1. Gender and firm-size: Evidence from Africa

You know what you like, but what about everyone else? A Case study on Incomplete Block Segmentation of white-bread consumers.

National Pork Board Report on Pork Cut Nomenclature. National Pork Producers Council 9/4/2009 1

Sensory Approaches and New Methods for Developing Grain-Based Products. Symposia Oglethorpe CC Monday 26 October :40 a.m.

Final Report to Delaware Soybean Board January 11, Delaware Soybean Board

Audrey Page. Brooke Sacksteder. Kelsi Buckley. Title: The Effects of Black Beans as a Flour Replacer in Brownies. Abstract:

Title: Report, High Tunnel Fresh Market Slicer Tomato Variety Trial 2010

Internet Appendix to. The Price of Street Friends: Social Networks, Informed Trading, and Shareholder Costs. Jie Cai Ralph A.

Feasibility Study: The Best Chewy Chocolate Brand Name Granola Bar Available at the Denton Wal-Mart.

Materials and Methods

*p <.05. **p <.01. ***p <.001.

SWEET DOUGH APPLICATION RESEARCH COMPARING THE FUNCTIONALITY OF EGGS TO EGG REPLACERS IN SWEET DOUGH FORMULATIONS RESEARCH SUMMARY

An Advanced Tool to Optimize Product Characteristics and to Study Population Segmentation

Relation between Grape Wine Quality and Related Physicochemical Indexes

Final Exam Financial Data Analysis (6 Credit points/imp Students) March 2, 2006

Imputation of multivariate continuous data with non-ignorable missingness

DEVELOPMENT OF MILK AND CEREAL BASED EXTRUDED PRODUCTS

DERIVED DEMAND FOR FRESH CHEESE PRODUCTS IMPORTED INTO JAPAN

Transcription:

農林學報 63(4): 225-233 (2014) - 225 - Journal of Agriculture and Forestry, 63(4): 225-233 (2014) Color, Flavor, and Texture: Which Blackberry Sensory Attribute is the Most Important to Consumers? Shang-Ho Yang [1] ABSTRACT Blackberries have received more recognition recently, compared to other berries, because of the high nutritional value and source of antioxidants. Nevertheless, regarding the important sensory attributes, such as color, flavor, and texture, consumers do not evaluate these attributes equally. This study focuses on the sensory attributes: color, flavor, texture, and overall appeal of six different blackberry cultivars, i.e., Apache, Chesapeake, Chickasaw, Kiowa, OAL-W6, and Ouachita. The attributes were measured and analyzed by a sensory examination, which was implemented by the Food Systems Innovation Center at the University of Kentucky. Results indicate that the taste and appearance of selected blackberry cultivars are different and can be distinguished based on color, flavor, and texture. Especially, flavor is considered the most important attribute in sensory examination compared to color and texture. Along with prior experiences of consuming blackberries, cooking experiences also enhance overall appeal of the taste of a blackberry. Key Words: Blackberry, food sensory, consumption. I. Introduction Worldwide production of commercial blackberries, Rubus spp., has increased more than 45% since 1995, and North America accounted for 42% of the production, the largest contribution of worldwide production [11,16]. With growing consumer demand for blackberries in the U.S., research scientists [2,7,13] have attempted to improve blackberry cultivars and tackle the potential constraints like: first, higher labor costs in production [6,16] ; second, more than 70 cultivars/species create inconsistent taste and sweetness of blackberries, is spreading around North America according to the question is how likely consumers prefer one blackberry cultivar over another based on the specific attributes of color, flavor, texture, and overall appeal in the consumer market. Blackberry production in the U.S. has been concentrated in Oregon (72%) and California (7%) in 2005 [16]. The blackberry production in Arkansas (4%) is increasing because of an interest for the value-added crop and one of the largest blackberry breeding programs is located at the University of Arkansas [11]. Similar to other southeastern states, Kentucky farmers look for new alternative crop opportunities after the announcement of the suspended tobacco-quota Plants Profile (USDA/Natural Resources Conservation support system [5]. Further, high quality blackberries Service); third, the blackberry is the most perishable are readily marketable in Kentucky since blackberries berry [12], so it enhances marketing challenges of continue to be popular in the market [3]. Kentucky shipment and shelf life. These constraints would consumers generally perceive that the taste of thorny require more time to overcome, but another important blackberries is better than thornless cultivars [19]. [1] Assistant Professor at Graduate Institute of Bio-Industry Management, National Chung Hsing University, Taichung 40227, Taiwan. The author is grateful to Dr. Timothy A. Woods, FSIC research team and two anonymous referees for valuable comments. All remaining errors are the author's. * Corresponding Author. E-mail: bruce.yang@nchu.edu.tw

- 226 - Shang-Ho Yang: Color, Flavor, and Texture: Which Blackberry Sensory Attribute is the Most Important to Consumers? The objectives of this study were: (1) to find out how consumers evaluate these attributes differently in terms of their overall appeal of the given blackberry cultivars, (2) to interpret what attribute is the most important to blackberry consumers, and (3) to identify how prior experience in blackberry consumption and cooking with blackberries influences consumer perception for each cultivar. II. Materials and Methods 1. Study Area Blackberries have become more popular as an Though breeding advances may enhance quality and flavor for thornless blackberries, changing consumer perception for the quality and flavor between thorny and thornless blackberries would require further examination. The market popularity of blackberries has received more focus in the U.S. because of the recognition of high nutritional values and as a source of antioxidants [15]. Blackberries have been the top choice for the juice blend market. Nutraceutical marketers, a combination meaning of nutrition and pharmaceuticals, have focused on health-related fruit juice that usually blends blackberry, blueberry, Concord grape, etc. The healthrelated fruit juice has a high consumer acceptance in the alternative crop in Kentucky [3]. With the growing market [8,10]. Lawless et al. [8] found a positive consumer willingness-to-pay for a nutraceutical-rich juice blend. demand of locally produced fruits, blackberry growers are interested in expending their scale and offering high Consumers do have a higher purchase intent if they were quality blackberries in Kentucky. The growing informed of the potential health properties, like conditions and climate in Kentucky are very suitable for anthocyanin content in the juice [9]. Therefore, the many different blackberry cultivars, at least 30 different blackberry market, when associated with nutrition and health properties, would receive consumer attention. Nevertheless, consumer preferences on blackberry taste and flavor among different cultivars would require further study. Each blackberry cultivar has its own unique aroma and flavor which may vary if grown in different climates, cultivars according to the Plants Profile (USDA/Natural Resources Conservation Service). Nevertheless, growers may have difficulty deciding which cultivars will receive the most acceptance from consumers. Regardless of the differences between costs and profits for each cultivar, this study solely focuses on the taste and flavor of blackberry cultivars in the market based on regional environments, and soil types [18]. This study a consumer sensory examination in Kentucky. focuses blackberry cultivars in Kentucky. Research scientists have investigated the attributes of color, flavor, and texture profiling individually among different 2. Blackberry Samples Based on the growth habit of blackberry, it can be blackberry cultivars [1,4,10,14,18]. However, with all these grouped as: erect (thorny and thornless), semi-erect attributes together, do consumers evaluate each (thornless), and trailing, which is not suitable for attribute equally or differently? If equally, it is commercial production in Kentucky [3], so the trailing reasonable to focus on each attribute individually type is not included in this study. Since semi-erect regarding any specific study; if differently, the biggest concern is finding which attribute is the most important. Particularly, prior consumer experience in consumption and cooking using blackberries would have potential cultivars, which were not ripe at the time of evaluation, were not included, this study focuses fresh blackberries on the thorny erect (including Chesapeake, OAL-W6, Kiowa, and Chickasaw) and thornless erect (including different effects on their preference and should be Apache and Ouachita) cultivars. Four thorny erect further investigated as well. cultivars, i.e., Chesapeake, OAL-W6, Kiowa, and

農林學報 63(4): 225-233 (2014) - 227 - Chickasaw, were harvested from the University of blackberry samples were given randomly to each Kentucky Research and Education Center, Princeton, panelist. Panelists were compensated for their time Kentucky. Two thornless erect cultivars, i.e., Apache after completing the blackberry taste evaluation. and Ouachita were harvested and purchased from Caludi's Fields at Lexington, Kentucky. In order to preserve as fresh as possible for best evaluation of consumer preferences, the four thorny erect cultivars were refrigerated at 40 F while the two thornless erect cultivars were refrigerated at 30-32 F, and were transported in coolers to the sensory experiment lab at the Food Systems Innovation Center (FSIC) at the University of Kentucky in Lexington. 3. Sensory Analysis 4. Statistical Analysis The collected data in this study were analyzed using STATA (version 12.1 for Windows; StataCorp LP, College Station, Texas). In order to understand whether these six blackberry cultivars are different with respect to each attribute, i.e., color, flavor, texture, and overall appeal, an analysis of variance (ANOVA) with Tukey's HSD (Honestly Significant Difference) test for multiple means comparisons was used. Panelists were asked to indicate their overall appeal in each trial among giving Blackberry sensory evaluations were completed in three different cultivars of fresh blackberries. Two June, 2012. Partitioned white booths accommodated regressions, ordinary least square (OLS) and up to 8 panelists at a time. Panelists (N = 59) were Multinomial Logit, were conducted to determine the recruited by the FSIC. Each panelist was instructed to begin with how to proceed their evaluation. In order to optimize the sensory taste outcomes and reduce bias from tasting fatigue, panelists were only given four trials: A, B, C, and D. Panelists were instructed to rinse with water between each tasting. impact of attributes, i.e., color, flavor, texture, and prior experiences, on the overall appeal of fresh blackberries. The overall appeal was explained with a linear function of color, flavor, texture, consumption, and cooking: In each trial, three different cultivar samples of fresh 1 blackberries were provided. In each tasting within each trial, panelists were asked to evaluate how they liked the where Overall_Appeal (including Color, Flavor, and Texture) was a sensory rating by panelists, Consumption following attribute by using a hedonic scale. The (including Cooking) was a dummy variable indicating attributes evaluated were color, flavor, and texture. The hedonic scale ranged from 1 (dislike extremely) to 8 (like extremely) for the attributes of color, flavor, and texture. At the end of each trial, panelists were then asked to indicate their overall appeal from their fresh blackberry tastings by using another hedonic scale, which is ranged that panelists had blackberry consumption or cooking experience during the past 12 months, were the coefficient estimates, and was assumed to be a normally distributed error term. An OLS model was included to provide a baseline estimate for relevant parameters. The values of Overall from 1 (Unappealing) to 12 (Most appealing). After Appeal in OLS are just in levels from 1 to 12. The trial D, each panelist was asked to recall their prior application of OLS would not provide any specific blackberry consuming and cooking experiences. A information in respect to high or low appeal differences data set of 688 (59 panelists 12 tasting=708, but 20 observations were drop because of unclear writing and missing) total observations was compiled and readily for for fresh blackberries. In order to determine the overall appeal of sensory tasting that can reveal potential differences among low, middle, and high appeals, this analysis. In order to reduce order bias, the fresh study segmented the overall appeal into three

- 228 - Shang-Ho Yang: Color, Flavor, and Texture: Which Blackberry Sensory Attribute is the Most Important to Consumers? categories: Low Appeal (1-4), Middle Appeal (5-8), and High Appeal (9-12). A Multinomial Logit model 2 would be suitable to explain these attributes among the overall appeal segments. The summary and descriptive for each variable is presented in Table 1. III. Results and Discussion This study focuses on three research questions: first, whether these six blackberry cultivars are different to panelists regarding the attributes of color, flavor, texture, and overall appeal; second, whether the attributes of color, flavor, texture, and prior experiences have any potential impact on overall appeal; third, if the prior experiences, i.e., consumption and cooking, have any significant difference, and how consumption and cooking are associated with each attribute in respect to Table 1 Definitions and Sample Statistics of Variables (N=688) Variables Description of Variables Mean Std. Dev. Min. Max. Overall Appeal Color Flavor Texture Consumption Cooking Discrete variable; panelists were asked to indicate the overall appeal for the taste comparison in each trial, with 1 being least appealing and 12 being most appealing. Discrete variable; panelists were asked to indicate the color score for each observing, with 1 being least preferred and 8 being most preferred. Discrete variable; panelists were asked to indicate the flavor score for each tasting, with 1 being least preferred and 8 being most preferred. Discrete variable; panelists were asked to indicate the texture score for each tasting, with 1 being least preferred and 8 being most preferred. Binary variable=1 if panelist had blackberry consumption during the past 12 months, 0 otherwise. Binary variable=1 if panelist did use blackberry for any purpose of cooking during the past 12 months, 0 otherwise. 7.30 3.35 1 12 6.06 1.48 1 8 5.15 2.04 1 8 5.53 1.73 1 8 0.93 0.25 0 1 0.45 0.49 0 1 Table 2 Blackberry Evaluation on Color, Flavor, Texture, and Overall Appeal Cultivars Color 1 Flavor 1 Texture 1 Overall Appeal 2 Apache 4 6.523b 5 5.608a 5.841a 8.016b Ouachita 6.578b 5.260ab 5.669a 7.894ab Chesapeake 6.116ab 5.520a 5.733a 7.635ab OAL-W6 5.821ac 5.285ab 5.660a 7.363abc Kiowa 5.948a 4.730bc 5.315ab 6.710ac Chickasaw 5.398c 4.452c 4.966b 6.155c ANOVA F-test 11.47*** 3 6.10*** 4.36*** 5.66*** Note: 1 Blackberry color, flavor, texture: 1 = Least preferred; 8 = Most preferred. 2 Rating blackberry for overall appeal: 1 = Unappealing; 12 = Most appealing. 3 Asterisks indicate levels of significance: *** = 0.01. 4 The order of varieties followed by Overall rankings' mean. 5 Means within a column followed by the same letter are not significantly different (Tukey's HSD, P < 0.05). 2 The Ordered Logit methodology are also suitable, and the results are ready based on request.

農林學報 63(4): 225-233 (2014) - 229 - each blackberry cultivar. The results of ANOVA with Tukey's HSD tests are appeal in compared to other cultivars. As a result, the thornless erect cultivars Apache and Ouachita were presented in Table 2. Before the multiple means rated first and second as most appealing to panelists, comparison procedure, the ANOVA F-test was respectively. In general, the assessed outcomes performed and highly significant, which indicates that each blackberry cultivars are different in respect to each significantly show that the thorny erect cultivars rated in this study do not have a better flavor and texture than attribute. The values of each cell show an average the thornless erect cultivars Apache and Ouachita. score of each cultivar in respect of each attribute, i.e., The outcomes of OLS and Multinomial Logit Models color, flavor, texture, and overall appeal. The are presented in Table 3 and demonstrate that the indications of Tukey's HSD tests for the multiple means comparison show that each blackberry cultivar is somewhat different according each attribute. Among the attributes, texture and overall appeal ratings were not significantly different in the cultivars of Apache, ratings of the overall appeal of blackberries can be explained by color, flavor, texture, and prior experiences. The F-test and Wald test show that there are strong differences overall across these attributes for OLS and Multinomial Logit Models, respectively. The results of Ouachita, Chesapeake, OAL-W6, and Kiowa. The AIC (Akaike Information Criterion) and BIC (Bayesian cultivar Apache scored highest in flavor, texture, and Information Criterion) indicate that the Multinomial overall appeal. The cultivar Ouachita scored highest Logit Model has better goodness of fit over than OLS, so and Apache second in color. It is an interesting result we utilize the results of the Multinomial Logit Model as that the popular thorny erect cultivar, Chickasaw, in our final interpretation. However, the coefficient Kentucky scored last in color, flavor, texture, and overall estimates of OLS and Multinomial Logit Models are Table 3 The Outcomes of OLS and Multinomial Logit Model Estimators OLS Multinomial Logit Dependent Variable Overall Appeal (1-12) Middle 1 Appeal (5-8) High 1 Appeal (9-12) Independent Variables Color 0.267*** (0.055) 0.133 (0.085) 0.424*** (0.112) Flavor 1.078*** (0.047) 0.803*** (0.091) 1.544*** (0.123) Texture 0.296*** (0.058) 0.164* (0.087) 0.479*** (0.115) Consumption -0.214 (0.346) -0.782 (0.618) -1.092 (0.716) Cooking 0.370** (0.155) 0.334 (0.260) 0.770** (0.313) Constant -1.502*** (0.503) -3.793*** (0.931) -11.520*** (1.255) Number of observations 688 688 F-test 257.710*** Adjusted R 2 0.651 Log likelihood -1443.093 Wald χ 2 159.240*** Pseudo R 2 0.367 Log Pseudolikelihood -467.584 AIC 4.212 1.394 BIC 2925.390 1013.575 Notes: 1 The base category "Low Appeal (1-4)" is compared. Asterisks indicate levels of significance: *=0.10, **=0.05, and ***=0.01. Standard errors are reported in parentheses.

- 230 - Shang-Ho Yang: Color, Flavor, and Texture: Which Blackberry Sensory Attribute is the Most Important to Consumers? consistent. The category "Low Appeal (1-4)" is use blackberries for any purpose of cooking during the normalized to zero as a base category. In comparison past 12 months or not. Figure 1 represents how to low appeal, higher scored color, flavor, and texture are associated with a higher likelihood of high appeal (9-12) cooking experiences can be associated with overall appeal ratings for each blackberry cultivar. A pairwise for fresh blackberries. Individuals with blackberry comparison test was performed in each figure as well. cooking experience are more likely to have a high appeal for fresh blackberries compared to those who have not had cooking experience. Hence, the sensory attributes, i.e., color, flavor, and texture, plus cooking experience have a positive impact on the overall appeal for fresh blackberries. The average marginal effects (in Table 4) of the Each dot along the curve presents an average rating of overall appeal regarding their cooking experience for the specific blackberry cultivar. In Figure 1, individuals with some levels of cooking experience scored Kiowa, OAL-W6, and Ouachita higher in overall appeal than those who do not have cooking experience. The results in Figure 1 correspond to our regression findings in Multinomial Logit Model sum up to zero in each Tables 3 and 4. Particularly, these three cultivars, i.e., attribute across different categories of overall appeals. The outcomes of average marginal effects represent a relative impact among categories. It is clear to observe how each attribute and prior experience are corresponding to each other among the overall appeal categories. We see that high appealing individuals are more positively affected by color, flavor, texture, and cooking experience than those who are middle and low Kiowa, OAL-W6, and Ouachita, are recognized as the major drivers for the overall appeal rating. Marketers may utilize this information for potential marketing strategies to target the household meal planner with specific blackberry cultivars for the purpose of cooking. IV. Conclusions The sensory properties, i.e., color, flavor, texture, and appeals. Particularly, the flavor (11%) is the major overall appeal, are significantly different among these driver (at least two times larger) for high appealing six blackberry cultivars. Although the mean score of individuals in comparison to color (4%), texture (4%), and cooking experience (6%). This reveals that flavor is the most important attribute to consumers. flavor is lower than the mean scores of color and texture, the influence of flavor scoring is much stronger (at least two times larger) on overall appeal than on color and There did appear to be some evidence that texture scorings. This result suggests that the future individuals' experience associated with cooking also has study for fresh blackberry fruits should focus on a positive impact on overall appeal. The cooking enhancing the quality and flavor-oriented component experience simply indicates that whether individuals did for an increasing demand of blackberries in Kentucky. Table 4 The Outcomes of Average Marginal Effects in OLS and Multinomial Logit Models Estimators OLS Multinomial Logit Dependent Variable Overall Appeal (1-2) Low Appeal (1-4) Middle Appeal (5-8) High Appeal (9-12) Color 0.267*** (0.055) -0.018** (0.007) -0.022* (0.012) 0.040*** (0.010) Independent Variables Flavor 1.078*** (0.047) -0.090*** (0.005) -0.022** (0.008) 0.112*** (0.007) Texture 0.296*** (0.058) -0.021*** (0.008) -0.022* (0.012) 0.044*** (0.011) Consumption -0.214 (0.346) 0.080 (0.057) -0.024 (0.076) -0.055 (0.061) Cooking 0.370** (0.155) -0.040* (0.024) -0.023 (0.034) 0.063** (0.028) Notes: Asterisks indicate levels of significance: * = 0.10, ** = 0.05, and *** = 0.01. Standard errors are reported in parentheses.

農林學報 63(4): 225-233 (2014) - 231 - Especially, thornless erect cultivars Apache and Ouachita highly interested in new alternative crop opportunities receive higher overall appeal than other thorny erect whenever the market is ready for it. With the cultivars, which are more popular in Kentucky. The sensory attributes and cooking experience are highly associated with the overall appeal of blackberries. Particularly, consumers with high blackberry appeal are positively affected by color, flavor, texture, and cooking experience, while consumers with middle blackberry appeal are positively affected by flavor and texture in compared to those with low blackberry appeal. Consumers with high blackberry appeal are also influenced from their cooking experience with the aforementioned findings, blackberry farmers and marketers will have better information to decide which blackberry cultivars they can focus on growing and selling for their potential consumers. Although only six major blackberry cultivars are used in this study, the blackberry attributes among different cultivars still present somewhat different results. For future studies, it would be necessary to pull in different cultivars that are also available in Kentucky, because consumers do have diversified experiences that may associate how utilization of blackberries. Three cultivars, i.e., Kiowa, they consume blackberries. OAL-W6, and Ouachita, turn out to be the major drivers among these six cultivars for blackberry overall appeal. Blackberries have received more attention not only Acknowledgments The authors are grateful to FSIC research team and two anonymous referees for valuable comments. All in fresh berry markets but also in the nutritionaloriented juice blend markets. Kentucky farmers are remaining errors are the authors'. Adjusted Predictions of cultivars#cooking_experience with 95% CIs Linear Prediction 5 6 7 8 9 Apache Chesapeake Chickasaw Kiowa OAL-W6 Ouachita cultivars Have cooked before None Figure 1 A Pairwise Comparison for Six Blackberry Cultivars between Overall Appeal and Cooking

- 232 - Shang-Ho Yang: Color, Flavor, and Texture: Which Blackberry Sensory Attribute is the Most Important to Consumers? References 450. [10] Lawless, L. J. R., Threlfall, R. T., Howard, L. R. and [1] Bett Garber, K. and Lea, J. (2013). "Development of Meullenet, J. F. (2012c). "Sensory, Compositional, Flavor Lexicon for Freshly Pressed and Processed and Color Properties of Nutraceutical-Rich Juice Blueberry Juice." Journal of Sensory Studies. Blends." American Journal of Enology and [2] Clark, J. R. (2005). "Changing times for eastern United States blackberries." HortTechnology 15: Viticulture 63: 529-537. [11] Rodríguez, H. G., Popp, J., Thomsen, M., Friedrich, H. 491-494. and Rom, C. R. (2012). "Economic Analysis of [3] Crop Diversification & Biofuel Research Education Center (2010). "Blackberry. Reviewed by Strang" Investing in Open-field or High Tunnel Primocanefruiting Blackberry Production in Northwestern J., College of Agriculture, University of Kentucky Cooperative Extensive Service. [4] Du, X. F., Kurnianta, A., McDaniel, M., Finn, C. E. and Arkansas." HortTechnology 22: 245-251. [12] Safley, C. D. "Market outlook for blackberry production in the Southeast." Paper presented at Qian, M. C. (2010). "Flavour profiling of 'Marion' the annual meeting for the SE Regional Fruit & and thornless blackberries by instrumental and sensory analysis." Food Chemistry 121: 1080-1088. [5] Duffrin, M. W. and Pomper, K. W. (2006). "Development of flavor descriptors for pawpaw fruit puree: A step toward the establishment of a native Vegetable Conference, Savannah, GA, January 9-11, 2009. [13] Safley, C. D., Boldea, O. and Fernandez, G. E. (2006). "Estimated costs of producing, harvesting, and marketing blackberries in the southeastern United States." HortTechnology 16: 109-117. tree fruit industry." Family and Consumer Sciences [14] Sousa, M. B., Canet, W., Alvarez, M. D. and Research Journal 35: 118-130. [6] Ernst, M., Woods, T., Strang, J. G. and Jones, T. (2009). Fernandez, C. (2007). "Effect of processing on the texture and sensory attributes of raspberry (cv. "2008 Kentucky blackberry cost and return estimates: For thorny, thornless erect, and thornless Heritage) and blackberry (cv. Thornfree)." of Food Engineering 78: 9-21. Journal semi-erect blackberries." College of Agriculture, [15] Stafne, R., McGlynn, W. G., Stafne, E. T., Lucas, E. A. University of Kentucky Cooperative Extensive Service ID-149. [7] Jones, R. T. and Strang, J. G. (2005). "Growing blackberries & raspberries in Kentucky." College of Agriculture, University of Kentucky Cooperative Extensive Service HO-15. [8] Lawless, L. J., Nayga, R. M., Akaichi, F., Meullenet, J. F., Threlfall, R. T. and Howard, L. R. (2012a). and Clark, J. R. (2009). "Sensory, Health and Quality Evaluation of Two Blackberry (Rubus subgenus Rubus) Cultivars from Arkansas and Oklahoma." Hortscience 44: 1121-1121. [16] Strik, B. C., Clark, J. R., Finn, C. E. and Bañados, M. P. (2007). "Worldwide blackberry production." HortTechnology 17: 205-213. [17] USDA/Natural Resources Conservation Service, "Willingness to Pay for a Nutraceutical Rich Juice "Plants Profile: Rubus L. Blackberry." Accessed Blend." Journal of Sensory Studies 27: 375-383. [9] Lawless, L. J., Threlfall, R. T., Meullenet, J. F. and Howard, L. R. (2012b). "Consumer Based Optimization of Blackberry, Blueberry and Concord Juice Blends." Journal of Sensory Studies 27: 439- April 7, 2013. http://plants.usda.gov/java/profile? symbol=rubus. [18] Wang, Y. Y., Finn, C. and Qian, M. C. (2005). "Impact of growing environment on Chickasaw blackberry (Rubus L.) aroma evaluated by gas chromatography

農林學報 63(4): 225-233 (2014) - 233 - olfactometry dilution analysis." Journal of Agricultural and Food Chemistry 53: 3563-3571. [19] Woods, T. A. Yang, S. H., Anandappa, A., Wolfe, D. and Strang, J. (2013). "Consumer evaluation of thornless erect and thorny blackberry varieties based on flavor, color, texture, and overall taste." 2012 Fruit and Vegetable Research Report. College of Agriculture, University of Kentucky Agricultural Experiment Station Publication PR-656: 23-24. 2014 年 08 月 20 日收稿 2014 年 09 月 18 日修正 2014 年 11 月 28 日接受

- 234 - Shang-Ho Yang: Color, Flavor, and Texture: Which Blackberry Sensory Attribute is the Most Important to Consumers?