Dominance Curve

Dominance Curve

Purpose

TDS is a multi-attribute temporal profiling method that enables dominant attributes to be selected and recorded in real time. For example, during the consumption of a bite of chocolate or over multiple sips of a drink.

Data Format

  1. TDS EyeQuestion format.xlsx
  2. TDS Binary Format.xlsx
Two different data formats are possible. The data type is specified as TDS in both cases.
The Binary Format has a column per time point and each attribute is marked as 1 if the attribute has been selected as dominant at that time point, otherwise 0.
The other format has a Start and Stop column which indicate the times between which the attribute has been selected as dominant.
Attributes can be grouped into different time points using the Multiple column – this provides a text category which specifies which time point the attribute belongs to e.g. before, after.
Intensity is not recorded; the data is binary as in an attribute is either on (1) or off (0) at each time point.

Background

bIn most applications judges are presented with 8-12 attributes on a screen and ask to click the attribute that is dominant as the consumption experience progresses. In some applications, intensity of the dominant attribute will be captured, but many practitioners feel that this detracts from the intuitive nature of capturing dominant attributes, and intensity is not captured for this analysis.
The dominance curve uses the proportion of evaluations for each attribute and product that are recorded as dominant at each time point. The result is a curve for each attribute and each product showing how the proportion varies with time.

Options

  1. Smoothing: Either select ‘no’ for no smoothing or choose one of the other 4 options to apply a weighted moving average algorithm to smooth out the curves of attribute averages vs time. The “lower” the amount of smoothing chosen the closer to the original proportion data the curve will represent, however the appearance maybe hard to read.
  2. Significance Threshold: The alpha level for the 1-sample proportion test that decides at which times an attribute is significant for each product. Default 5%.
  3. Level filter: The alpha level for the 1-sample proportion test that decides at which times the pairwise difference between curves is significant for each product.
  4. Scale by judge: Whether to standardize each product evaluation in the time domain, so that they all finish at the same time. If the food samples under test need to be chewed and swallowed, then the time taken to the swallow point can vary considerably between judges – use judge scaling to correct for this.
  5. Pair Comparison of Curves: Should a pairwise comparison of the curves be created.
  6. Automatic Time Interval: Curves are drawn with time intervals as automatically selected by the software based on the length of the total time interval.
  7. Manual Time Interval (seconds): Curves are drawn with time intervals as selected.
  8. Export binary dataset: If ‘Yes’ then an additional tab is created in the export that contains the data in the Binary format (i.e. one column per time point).
  9. Number of Decimals for Values: Choose preferred number of decimal places in subsequent output (default=2)

Results and Interpretation

  1. Dominance Curves per Product: One plot per product (for each attribute group as specified in the Multiple column) is produced. The proportion of evaluations that have selected each attribute as dominant at each point is plotted. Attributes can be added or removed from the plot by clicking on the attribute in the plot’s legend. In addition, the significance line and the chance line are indicated on the plot.
    1. The chance level is simply 1/p where p is the total number of attributes, since a judge who is randomly moving between p attributes, has a probability of selection of 1/p for each one.
    2. The significance line is determined by performing a 1-sample proportion test (Z-test) to decide where the height of the curve exceeds 1/p with an alpha level set in the Significance Threshold option.
  2. Significance: The significance value is determined by performing a 1-sample proportion test (Z-test) to decide where the height of the curve exceeds 1/p (where p is the number of attributes) with an alpha level set in the Significance Threshold option.
  3. Chance Level: Simply 1/p where p is the total number of attributes in each group, since a judge who is randomly moving between p attributes, has a probability of selection of 1/p for each one.
  4. Comparison: Differences between each pair of products are calculated and the part of the curve where the differences are significant are displayed.

Technical Information

  1. The R functions used for analysis were all developed in house at Qi Statistics.

References

  1. Galmarini, M. V., Visalli, M., & Schlich, P. (2017) “Advances in representation and analysis of mono and multi-intake Temporal Dominance of Sensations data” Food Quality and Preference, v56, pp247-255.

    • Related Articles

    • Temporal Dominance of Sensation (TDS)

      Introduction The Temporal Dominance of Sensation (TDS) involves presenting the panellist with a comprehensive list of attributes. Subsequently, the panellist is tasked with identifying the dominant attribute, representing the most prominent ...
    • Band Plot

      Purpose Creates a graphical summary of temporal dominance of sensations (TDS) data. Data Format EyeOpenR has a specific temporal data format that should be adhered to TDS EyeQuestion format.xlsx. The ‘product’ and ‘judge’ tabs of the Excel file are ...
    • Time Intensity Analysis

      Purpose Time Intensity tests are designed to measure the temporal evolution of a single attribute. Results are plotted as time-intensity curves and key parameters of the curves are calculated. Results can be averaged per panellist or per product. ...
    • R-Index

      Introduction The R-index is a tool in sensory science that helps quantify the difference between two samples. The R-index is designed to quantify the area under an empirical Receiver Operating Characteristics (ROC) curve within the framework of ...
    • R-index Analysis

      Purpose To analyse the results of an R-index test. Data Format R_index_rank_withMD.xlsx Background The R-index applies signal detection theory as an alternative approach to discrimination testing. It is often used where there are a large number of ...