Relating Objective and Subjective Ratings of Snap Bean Pod Color to Likelihood of Purchase

Snap bean (Phaseolus vulgaris L.) cultivars with pods representing a range of greenness were grown in Oklahoma field trials in 2001 and 2004. Objective color evaluations (L* value and hue angle) performed on raw and cooked pod samples from 10 (2001) or 12 (2004) of these cultivars indicated that color testing of raw snap bean pods may not be sufficient to determine the color after cooking. Although L* values may be expected to decrease after cooking, the magnitude of the changes may not be predictable. Changes in hue angle values after cooking appear to be even more variable among cultivars. Therefore, if the color of the cooked beans is expected to be a deciding factor in cultivar selection, we would recommend conducting color tests on cooked bean pods as well as the raw product. Twenty relatively straight, unblemished pods per cultivar were harvested on 20 June 2001 and on 30 June 2004 from plants of ‘Blue Lake 274’, ‘Brio’, ‘Charon’, ‘Jade’, and ‘Seville’. The five most uniform pods per cultivar (all sieve Size 4) were presented as raw samples that same day to an untrained panel consisting of seven males and 18 females (2001) or nine males and 18 females (2004). This was an affective test; panelists were asked to evaluate intensity of color and likeliness to buy using a 5point semantic differential scale. Correlation coefficients for the two attributes were calculated. Sensory panelists were able to make subjective distinctions among the cultivars based on color. However, these differences did not necessarily correlate with either objective color measures or likelihood of purchase. Snap bean pod color is not an overriding selection criterion, but only one of many criteria considered by consumers. Color and appearance factors are the primary quality attributes that can be appraised by consumers of fresh produce at the time of purchase (Gamble et al., 2006; Kramer, 1951). Historically, the snap beans purchased by U.S. consumers were a relatively light shade of green. However, in the 1990s, a major seed company introduced dark green beans to the U.S. fresh market (Brooker and Eastwood, 1992). The company promoted the dark green beans as being more attractive to buyers, but little objective information was available to back these claims. Moreover, it was not known how color might change after the dark green beans were cooked. The human eye can perceive visual differences in greenness among fresh samples of snap bean pods (Hoffman and Kanapaux, 1955). It is less well understood how these subjective distinctions relate to both objective measures of greenness and subjective ratings for likelihood of purchase. Hoffman and Kanapaux (1955) found a significant correlation (r = 0.906, P # 0.01) between visual color ratings and chlorophyll content of fresh snap bean pods, but chlorophyll is not the only pigment responsible for color in snap beans (Lopez-Hernandez et al., 1993). Brooker and Eastwood (1992) surveyed shoppers for selection preference when comparing fresh pods of the light green ‘Strike’ and the dark green ‘Labrador’ snap beans. They found a slight preference (52.2%) for the light green cultivar, but they did not attempt to relate these preferences to any objective measures of greenness. Lee (1958) noted that when green vegetables are heated in water, the chloroplasts become swollen and may burst, resulting in a diffusion of green pigment throughout the cells and a more intense green appearance to the surface of the vegetable. Muftugil (1986) found that bean samples blanched with water or steam were more green than raw beans, but cultivars were not compared for possible differential responses. The objectives of this study were: 1) to perform objective color evaluations on raw and cooked pod samples from several snap bean cultivars with pods representing a range of greenness; and 2) to determine whether consumers could distinguish differences in greenness among raw pod samples from several of these cultivars and, if so, whether these subjective distinctions would relate to both objective measures of greenness and subjective ratings for likelihood of purchase. Materials and Methods Field production trials. Studies were conducted at the Oklahoma Vegetable Research Station, Bixby, in 2001 and 2004. The soil was a Severn very fine sandy loam [coarsesilty, mixed (calcareous), thermic Typic Udifluvent]. Weeds were controlled with a broadcast preplant-incorporated application of trifluralin at 560 g ha supplemented by hand and machine cultivation in both years. Plant water requirements were met with rainfall supplemented by overhead sprinkler irrigation. Standard foliar insecticides were applied as needed. In 2001, a broadcast preplant-incorporated application of 48N–21P–39K (kg ha) was made on 3 Apr. Plants were topdressed with urea to supply 34 kg ha nitrogen (N) on 17 May. In 2004, a broadcast preplant-incorporated application of 29N–13P–24K (kg ha) was made on 2 Apr. Plants were topdressed with urea to supply 34 kg ha N per application on 20 May and 11 June. Thirteen snap bean cultivars were used in 2001 (Table 1) and 12 cultivars were used in 2004 (Table 2). Those represented by the seed companies as ‘‘dark green’’ included ‘Charon’, ‘Dusky’, ‘Jade’, and ‘Savannah’. The seed companies discontinued ‘Castano’, ‘Hialeah’, and ‘Mercury’ between 2001 and 2004. The experimental design was a randomized complete block with three replications. Seeds were sown by machine in rows that were 90 cm apart on 27 Apr. 2001 and on 29 Apr. 2004. No inoculants were used. Stands were thinned to 60 plants in 6 m on 10 May 2001 and on 19 May 2004. Each cultivar was harvested by hand one time at prime maturity by pulling up and depodding all plants in 3 m of each plot (dates of harvest are shown in Tables 1 and 2). Pods were weighed and a random sample (200 g or greater) was taken. Sampled pods were sorted into sieve size classes using a commercial gauge and weighed to determine percentages in each class. Sieve sizes are based on pod diameter with sieve Size 1 ranging from 4.8 to 5.7 mm; sieve Size 2 ranging from 5.8 to 7.3 mm; sieve Size 3 ranging from 7.4 to 8.3 mm; sieve Size 4 ranging from 8.4 to 9.5 mm; and sieve Size 5 ranging from 9.6 to 10.7 mm (Rutledge, 1996). Twenty-five total pods per plot (sieve Sizes 3 and 4) also were measured for pod length. Samples were taken to the laboratory on each date of harvest, and laboratory analyses typically occurred the next day after overnight storage at 5 C. Cooking. Sample pods of 10 cultivars (2001) or 12 cultivars (2004) from the field production trials were prepared to simulate fresh cooked beans by immersing raw beans Received for publication 2 Dec. 2008. Accepted for publication 10 Feb. 2009. Approved for publication by the Director, Oklahoma Agricultural Experiment Station. This research was supported in part under project H-2026. We gratefully acknowledge the technical assistance of Lynda Carrier. The information given in this publication is for educational purposes only. Mention of a trademark, proprietary product, or vendor does not constitute a guarantee or warranty of the product nor does it imply approval or disapproval to the exclusion of other products or vendors that may also be suitable. Professor. Associate Professor. To whom reprint requests should be addressed; e-mail [email protected]. HORTSCIENCE VOL. 44(3) JUNE 2009 737 POSTHARVEST BIOLOGY AND TECHNOLOGY