Campus Units

Food Science and Human Nutrition

Document Type


Publication Version

Accepted Manuscript

Publication Date


Journal or Book Title

Journal of Dairy Science




The objective of this study was to survey consumers' milk purchasing behavior and investigate the effect of sensory experiences and an educational message on the perceived value of fluid milk at the beginning and near the end of code. Eleven auction sessions were carried out (n = 100 consumers), which included an explanation of the experiment, a survey about demographics and milk purchasing and consumption, sensory evaluation, an educational message, and 3 rounds of nth price auctions. Consumers were blindly served 2 pairs of milk samples from white-pigmented high-density polyethylene [2% and skim milk within 2 to 3 d of production (fresh) and 2% and skim milk with 2 to 3 d until the end of code (near-end)], and asked to indicate their preference and the level of acceptability for each sample using a 7-point hedonic scale. All samples were simultaneously evaluated by a panel of 8 judges who were trained to evaluate milk quality attributes on a 15 cm unstructured line scale. Results from the consumer panel acceptability rating session, trained panel descriptive analysis, and consumer auction bids were analyzed using multivariate factor analysis of variance. Subjecting pre- and post-survey responses to k means cluster analysis revealed 4 bidding populations in each round. Most participants (82%) indicated that they check the code date on milk every time they shop; 77% said they reached for the code date that was farthest out every time. However, on blind tasting, consumers did not prefer fresh over near-end milk. These findings were in agreement with their acceptability scores for fresh 2% (5.0/7.0), near-end 2% (5.2/7.0), fresh skim (4.5/7.0), and near-end skim (4.6/7.0) milks. Trained panelists did not detect a difference in “lacks freshness” flavor in fresh skim (1.9 cm/15.0 cm) or near-end skim milk (1.3 cm). Surprisingly, trained panelists did detect higher “lacks freshness” flavor in 1 lot of fresh 2% (2.3 cm) compared with near-end 2% milk (0.3 cm). When consumers bid on half gallons of milk with visible code dates, fresh skim was valued $0.27 higher than near-end skim, and fresh 2% was valued $0.29 higher than near-end 2%. After blind sensory evaluation, the margin between the fresh and near-end bids decreased to almost zero (fresh skim was valued only $0.03 more than near-end skim; near-end 2% was valued $0.01 more than fresh 2%). After the educational message about the meaning of code dates, consumer bids for near-end ($0.63) and fresh milk ($0.81) decreased compared with the first round of bidding ($0.74 and $1.01, respectively). Additionally, the margin in bids for fresh milk remained numerically higher than those for near-end milk ($0.15 for 2% and $0.21 for skim). The educational message about code date did not have the intended result of increasing consumer value for milk. Although consumers go out of their way to buy the freshest milk, they cannot necessarily distinguish fresh milk from milk at the end of code; consumers appear to value a code date that was farther out more than superior taste, even after a sensory experience and educational message.


This accepted article is published as Paterson, M., Clark, S., Use of auctions to assess consumer value for fresh and end-of-code milk. Journal of Dairy Science, February 26, 2020. doi: 10.3168/jds.2019-17418

Copyright Owner

Elsevier Ltd.



File Format


Available for download on Friday, February 26, 2021

Published Version