Temporal Choice Workbook: TCATA
Overview
The Temporal Check All That Apply (TCATA) data can be analyzed directly in Compusense by generating the Temporal Choice
Workbook report. On this page you will learn how to generate the report
and what various options exist to help you meet your analysis needs.
TCATA Analysis
The Temporal Choice Workbook provides data, analyses, and graphs on the
results for all Temporal Choice questions, including the TCATA.
The
significance is calculated at
95%
.
Fisher's Exact Test
is used for the Difference Curves.
To generate the report:
-
Go to the
Results
. If you need to exclude specific samples and/or sample sets from the analysis, click Filters.
Click on the
Reports
menu and select
Create report
.
From the list of reports, select Temporal choice workbook.
Click the
2. Select options
menu and specify whether you would like to
include any of the curve smoothing options, time standardizations, set alpha value, and whether you would like to include the band plots in the report:
-
Curve smoothing.
Images in below examples represent the same data when different smoothing options are applied.
-
No smoothing. When this option is selected, no smoothing is applied to the curves.
-
Less smooth. When this option is selected, smoothing is applied to the curves. The below image shows the curves being smoother than when no smoothing is applied on the same data, as seen in the previous image.
-
Smooth. This is the default selection. When this option is selected, smoothing level applied to the curves is medium.
-
More smooth. When this option is selected, the highest smoothing level is applied to the curves.
- Standardize time.
By default, this group of options is deselected. This means that the time standardization is not applied; the original data is being graphed according to the Maximum time set in the Question options.
When any of the three options mentioned below are selected, the time is standardized to a 0-100 scale, regardless of what the Maximum time is set to in the Question options.
-
No trimming. When this option is selected:
- No data trimming of any kind is applied.
-
Other trimming options become disabled.
-
Left trim (Start to first attribute). Select this option to remove from the analysis the time before the first attribute is selected.
-
Right trim (last attribute to Stop). Select this option to remove from the analysis the time after the last attribute is selected.
-
Events.If your temporal question had
Events set up in the
Question options, then in the reporting area
you will get an option to split data on those events, if you choose to do so. Split by event(s). Place a checkmark to split the data by events. Selecting this enables all the other options pertaining to the data splitting by events. Band plots will not be affected by event splits.
- Split by event. Select an event to split by. If only one event was set up in the test, it will be selected by default. If multiple events exist, the first one will be selected by default. Only one event can be selected. You need to re-run the report separately for different events making the appropriate selections.
- Event options. Data splits by events will reflect in the TCATA Curves and TCATA Difference Curves sheets.
Choose from:
- Split data by first event selection. The data splits will be performed on the event chosen in the Split by event step based on the first time the event was selected by a panelist.
- Split data by last event selection. The data splits will be performed on the event chosen in the Split by event step based on the last time the event was selected by a panelist. If your test had only one event, the results will be the same if you select this option or Split data by first event selection option.
- Split data by max number of event selections. The data splits will be performed on the event chosen in the Split by event step based on the maximum number of times the event was selected by a panelist. This is not a cumulative number across the panel.
Please allow time for the report to generate when you select this option.
In this example image we can see that the data is graphed from the Start to the first time event was selected ("Swish #1"), then from the first time the even was selected to the second time it was selected ("Swish #2"), and so on until it reaches the last time the event was selected to Stop.
If a sample A (as an example) had a smaller max number of event selections than the sample B, the graphs will plot for the sample A only up to the maximum number of times that sample had the event selected.
- Discard evaluations where no event was selected. Select this option to automatically filter out results from panelists that did not select the particular event at all.
Alpha. Specify the alpha value to reflect in the TCATA Difference Curves and the Band Plot sheets.
-
Band plots.
Band plots are a visual representation of the duration significant descriptors/attributes were selected by the panel for a sample. The descriptors/attributes that have a value higher than the significance value will be graphed. Other descriptors/attributes will not be included on the graph. In the example image below we can see that 5 attributes were evaluated, but only one (the one indicating significant difference) is graphed for the specific sample.
Band Plots are not affected by the event splits while other selections outlined in the previous steps are reflected in the band plot sheets.
-
Include band plots. Select to include the BandPlots and BandPlotData sheets in your report.
-
Use selected descriptors only. When this option is deselected, the critical value calculation uses the total number of descriptors/attributes in the question. When selected, the critical value calculation uses the subset of descriptors/
attributes that panelists used for at least one sample during the evaluation. In other words, if a descriptor/attribute was not used by any of the panelists for any of the samples, then when this option is selected, that attribute will not be taken into critical value calculation.
Example when all descriptors/attributes are included in the calculation:
Example when only descriptors/attributes used by at least one panelist for at least one sample are included in the calculation:
From these two examples we can see that the Chance limit value, the significance value, and the graph can be affected when this option is selected vs. when it is not selected.
All the compatible questions are by default selected to be included in the report. Click the
3. Select questions
to review and make necessary changes.
The report can be generated in .xlsx file format only. Click
Create my report
, download the file and open it.
Further Analysis Details:
The analysis workbook for TCATA data can include the following sheets:
-
TCATA Curves. Displays the average of all complete responses for all included samples. There is one graph per sample.
-
TCATA Difference Curves.
Graphs displaying whether there is a significant difference between pairs of samples. Every sample is compared to another that is included in the analysis. Sample 1 in the pair is represented above the zero line and is also noted as the first sample in the graph title. Attributes displayed above the zero line indicate that the sample 1 in the pair is significantly different from the other sample in those attributes. This means that these attributes had significantly more selections for the given time period for sample 1 in the pair.
Sample 2 in the pair is represented below the zero line and is noted as the second sample in the graph title. Attributes displayed below the zero line indicate that the sample 2 in the pair is significantly different from the other sample in those attributes. This means that these attributes had significantly more selections for the given time period for sample 2 in the pair.
-
Data Proportions. Raw data for the TCATA Curves.
-
Data Difference Proportions. Raw data for the TCATA Difference Curves.
- Band Plots. Band plots are a visual representation (one graph per sample) of the duration significant descriptors/attributes were selected by the panel for a sample. The descriptors/attributes that have a value higher than the significance value will be graphed. The y-axis shows the significance level of the attribute. Hover over a band (strip of colour) to reveal the values.
In the example image below we can see that for the first sample, Apple flavor was selected roughly between 33rd and 85th points on a standardized scale, then it was deselected, and then it was selected again at the 90th point and remained selected to the end (100th point).
The x-axis values will reflect whether you selected time standardization or not before generating the report. For additional details please read the corresponding steps outlined above for selecting the Band plot options when generating the report.
- Band Plot Data. Raw data for the Band Plots.
Related Articles
Temporal Choice Workbook: TOS
Overview The Temporal Order of Sensations (TOS) data can be analyzed directly in Compusense by generating the Temporal Choice Workbook report. On this page you will learn how to generate the report and what various options exist to help you meet your ...
Temporal Choice Workbook: TDS
Overview The Temporal Dominance of Sensations (TDS) data can be analyzed directly in Compusense by generating the Temporal Choice Workbook report. On this page you will learn how to generate the report and what various options exist to help you meet ...
Standard Test: TCATA
Overview In Temporal Check All That Apply (TCATA) questions, panelists start an on-screen timer and then select all of the perceived attributes or sensations during the sampling time. They may check as many of the perceived attributes simultaneously, ...
Descriptive Analysis Workbook Product Means Radar Graphs
What is it? Product Means Radar Graphs is one of the sheets in the Descriptive Analysis Workbook. It is a visual overview of your product intensities for each attribute. Why would I use it? It can be helpful for initial product performance ...
PCA Workbook
Overview PCA (Principal Component Analysis) allows for a better understanding of the multi-dimensional relationships among samples and attributes. The visual representation allows for an easier interpretation of these relationships. The PCA is ...