Improved interpretation of TDS and TCATA curves using reference lines

Sara King/ May 11, 2015/ Oral Presentation/ 0 comments

It is common practice to incorporate chance and significance lines in Temporal Dominance of Sensations (TDS) curves. These lines draw attention to time intervals at which an attribute is cited (significantly) more often than would be expected by chance.

For Temporal Check-All-That-Apply (TCATA) data, in which attributes can be cited concurrently, the chance of an attribute being selected by an assessor cannot be determined a priori in the same manner. To guide interpretation, reference curves are proposed. TCATA data related to several product categories are used to illustrate interpretability benefits of reference curves. Noting these advantages, reference curves are then proposed to a supplement to TDS chance and significance lines, again using data sets from several previous TDS studies to illustrate the benefits.

TDS (TCATA) plots show dominance (citation) proportions per attribute over time for one product. After construction of these plots, reference lines are superimposed for each attribute based on the dominance (citation) proportion over time for the other products, based on pooled data. At time intervals where the dominance (citation) proportion for the product and the dominance (citation) proportion the other products are significantly different from zero, the reference lines are displayed, and suppressed otherwise.

Reference lines provide an objective approach for contrasting dominance (citation) proportions for the product against other products included in the study. Whereas chance and significance lines identify attributes that occur (significantly) above a chance level determined a priori, the reference lines are used to create a contrast of the product with the other products. Reference lines clarify whether a high dominance (citation) proportion is particular to a product, or generic for the product category, and help to indicate where a product has unusually high or low citation frequencies compared to other products in the study. Each approach provides different information but it is suggested that interpretability of TDS and TCATA curves are improved, often substantially, using this approach.

Castura, J. C. (2015). Improved interpretation of TDS and TCATA curves using reference lines. IV E3S Symposium & XVI Nordic Workshop in Sensory Science. 11-12 May. Oslo, Norway.


Share this Post

Leave a Comment

Your email address will not be published. Required fields are marked *