After you create an analysis, the modify menu will become available. The options in the modify menu allow you to interact with the settings in your analysis. In this menu, you can also find statistical test options available in Infotools Harmoni.
 As with other menus, the options in this menu are contextual.
 Actions will apply to the entire table by default unless you select a particular column, row or cell.
 Icons will turn orange when the selected analysis options are applied.
 Icons that display an arrow at the bottom allow expanding. This will open the menu and give you further analysis options.
 During your session, your setting preferences are retained across projects and analyses unless:
 You end your session (log out or close browser).
 You load an analysis that already has presaved settings (i.e. Story)
In this article
The modify menu options:
 Total  Analysis total options
 Format  Values displayed in an analysis
 Selected columns/rows
 Suppress items from analysis
 Test  Statistical Testing
 Analysis Layout
 Color Scheme
 Data Labels
 Information Bar
1. Total  Analysis total options
Unless different bases have been set on the elements included in the analysis, the Total is used as the denominator The number the count is divided by, when calculating a percentage. or base See bases article for more information on changing bases. for the percentage and/or average calculations in the body of the table.
There are four different Total options, Source Total is usually the default. You can change the total option using the modify menu. Infotools Harmoni calculates one Total option at a time, but multiple value types can be displayed.
Total Options 
Description 
Total 
All records in the source (s)A source is a collection of data that is either in an imported file or in a connected data store. Learn more about Data Sources. 
Total (qualified) 
All records excluding missing values. 
Total (analysis) 
All records excluding missing values and unselected items. 
Total (view) 
All records excluding missing values, unselected and hidden items. 
Learn more about Analysis Total Options.
2. Format  Values displayed in an analysis
Any row or column in an analysis can display different values including counts (weighted or unweighted), percentages, averages, etc.
a) Add/Replace Switch
The add/replace switch allows creating tables that show either complimentary data displays or singular displays. This can be applied to a selected part of the table/chart or the whole table/chart.
Replace
While the add/replace switch is on replace any data calculation change will replace the currently selected item (or the entire table if nothing is selected).
Add
If the switch is moved to add:
 Any new data type/calculation selected will be added as a new column or row next to the item you have selected.
 If nothing is selected, the default will be to create a new column of the data display for each of the original columns of that table.
b) Analysis setting options
Standard Axes  Value Types
Icon  Name  Description 
Weighted count 
Standard Axes Weighted count of data records for an element (if there are no weights this is the same as u123). Measures Weighted count of data records with a value. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Unweighted count 
Standard Axes Unweighted count of data records for an element. Measures Unweighted count of data records with a value. 

Percentage 
Standard Axes Weighted count divided by the weighted Base. Measures Weighted count of data records with a value in the measure divided by the count of data records included in all measures in the dimension. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Unweighted base 
Standard Axes Unweighted count of data records in Base. Measures Unweighted count of data records included in any measure in the dimension. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Base 
Standard Axes Weighted count of data records in Base (if there are no weights this is the same as uB). Measures Weighted count of data records included in any measure in the dimension. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. Base Options Cell Base
Total Base
Learn more about Analysis Bases. 

Differences 
Standard Axes Difference between an element and the reference. Measures Difference between the measure and the reference. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. Format Option Reference format allows switching difference format to the default format.


Indices 
Standard Axes Proportional difference between an element and the reference. Measures Proportional difference between a measure and the reference. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Averages 
Standard Axes with values and Measures Summ of values divided by the count of data records with value. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Sum Value Percentage 
Measures Weighted sum divided by the sum of values across all measures in the dimension. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Sum Value 
Measures The weighted sum of values in the measure. Sum of values over all respondents in the cell. Decimal Options The arrow at the bottom allows expanding to set up decimal places. Up to four decimal places can be selected. 

Unweighted Sum Value 
Measures The unweighted sum of values in the measure. 
Infotools Harmoni applies bankers rounding.
Bankers Rounding is an algorithm for rounding quantities to integers, in which numbers which are equidistant from the two nearest integers are rounded to the nearest even integer.
We round 0.5 to the nearest even digit.
Example:
 7.5 rounds up to 8 (because 8 is an even number)
 but 6.5 rounds down to 6 (because 6 is an even number)
3. Selected column or row settings
You can select any row or column in the analysis and display different value types, including counts (weighted or unweighted), percentages, averages, etc.
a) Hide
In addition, you can select columns and/or rows and hide these items from the view.
 The items will still remain part of the overall calculation of the analysis, but hidden from the view.
 If you want these items to be excluded from the overall calculation of the analysis, you can use the View Total Option.
b) Rank
Icon  Name  Description 
HIDE  Hide 
It only appears when something is selected. Hides the selected items. Hidden items can be added back into the analysis by using the select menu. 
Rank 
It only appears when a header column or row is selected. Rank from high to low, low to high and off/default. Column selection will rank items within that column (vertical). Selecting a row will rank items within that row (horizontal). 
4. Suppress items from analysis
Icon  Name  Description 
ZERO  Suppress zero 
Suppresses items where the numerator is zero. It only appears when nothing is selected in your analysis. 
NULL  Suppress null 
Suppresses items where both the numerator and denominator are zero. It only appears when nothing is selected in your analysis. 
5. Tests  Statistical Testing
a) Reference for testing
It only appears when a single row or column is selected. The selected row or column becomes the reference for significant difference calculations.
Previous Period Referencing
When a timebased axis is a part of your analysis (generally columns/across) you have the ability to set significance testing to be based on a rolling period. This could be the previous period or even the previous 4th period. Depending on your time axis this allows you to set a reference to a preferred time period.
b) Significant Differences
Statistical significance of the differences between the results for one group and a reference group (often the total sample).
An analysis can contain discrete or continuous variables.
 Discrete variables contain labeled responses that have a yes, no or missing response for each record. The statistical significance of the difference between two discrete variables is calculated using both counts, and bases. The statistical test used is the chisquared test.
 Continuous variables are called 'measures' or 'values'. Measures in an analysis may either be sums (i.e. totals) or averages. For comparisons of continuous variables (measures or averages), the statistical test used to determine significant differences is a ttest. The test is applied using numerical values.
The significance level is the likelihood that the difference is not due purely to chance. If the confidence level is set to 95%, the change is said to be statistically significant, as there’s only a 5% chance that the difference is due purely to chance.
Infotools Harmoni applies an automatic (and generally industry default) confidence level of 95% for significant difference calculation. As a user, you can override this via the modify menu. Site settings can be controlled via a site key.
Significance testing is represented using red and green arrows.
c) Low sample size
When the percentages shown in tables are based on small subsamples, the results may be unreliable. The Low Sample Size feature hides or fades values in the table when the unweighted value of the denominator used for percentage calculations is lower than a specified level.
 Low Sample warnings:
 Low sample  Transparency is applied to the results to match the warning threshold.
 Too low sample  An asterisk (*) is shown instead of the result.
d) Effective Base
Traditionally the test for significance uses the unweighted base in the calculation. This can lead to an overestimate of the significant difference between the two data cells when extreme weighting has been applied. By using the effective base instead of the unweighted base in the calculation we can adjust for this overestimate.
The effective base can be set through personalization keys.
Tests Summary Table
Icon  Name  Description 
Reference 
It only appears when a single row or column is selected. The selected row or column becomes the reference for significant difference calculations. Previous Period Referencing When a timebased axis is a part of your analysis (generally columns/across) you have the ability to set significance testing to be based on a rolling period. 

Significance test 
It only appears when nothing is selected.
Confidence Level The confidence level used for the significant difference calculation can be set independently for each saved view. A date axis can have an offset reference set for significant difference calculations. 

Low sample size 
It only appears when nothing is selected. Low sample size values can be set.

6. Analysis Layout
a) Flip analysis
Flips the analysis so what was across moves into the down dimension, and vice versa. This has the same functionality as clicking on the by between the across and down dimensions in the analysis title.
b) Reorder columns and rows
Allows shifting selected columns and/or rows in an analysis table view. For this option to become available, you must first select the column(s) or row(s) you want to move.
Once you have moved:
 The order is retained when saving a view.
 The order applies to Excel and PowerPoint exports.
 Selectors will also reflect the order of the items. The selector won’t reflect the change if the order change is different across nested items.
Reorder is not available when ranking has been used. If you want to change the order of items you first need to remove the ranking.
Icon  Name  Description 
Flip 
Appears when you have items in both the across and down dimension of an analysis. Flips the analysis so what was across moves into the down dimension, and vice versa. This has the same functionality as clicking on the by between the across and down dimensions in the analysis title. 

Reorder

Allows shifting selected columns and/or rows. You must first select the column or row for the option to become available.

7. Color Scheme
Changes scheme for the session from dark to light, and vice versa.
8. Data Labels
a) Discover
Discover is a tool to profile a target group against a set of descriptor variables. It will show you which of these descriptors best describe your group and compare these with other groups.
On a table view, select the group you are interested in profiling. Selecting will enable the Discover option located in the modify menu under data labels.
b) Heat map
Heat map is another statistical test option available in Infotools Harmoni. Using a stepped coloring from green to red, you can quickly see the highs and lows of significance in the results as a visual distribution across the entire table.
The heat map colors the cells in the table to show the quintile distribution within a row. It compares the data within each row and divides into 5 quintiles (colors).
A quintile is a statistical value of a data set that represents 20% of a given population, so the first quintile represents the lowest fifth of the data (1% to 20%); the second quintile represents the second fifth (21% to 40%) and so on.
c) Chart labels
You can interact with the data labels you want to display in your chart. Options include:
 Show/hide labels.
 Show only labels for items where there is a significant difference to the reference.
 Show the value gap of items when compared to the reference.
 Optimize the label placement to avoid overlaps that would make labels hard to read.
d) Show Sample Size on Charts
When creating an analysis and choosing to visualise you can display the sample size within the graph.
The show sample size option can be switched on or off through the modify menu. The setting selected is sticky for all the analyses in your session.
If your percentages are based on the down dimension, which is the default behavior, the sample size appears in the across legend labels. If you have changed the percentages to be based on the across dimension, the sample size will appear in the graph legend. The sample size is the unweighted base.
The sample size is the unweighted base (n=uB).
Sample size on export to PowerPoint and Excel
When show sample size option is on:
 In Excel  the sample size is available even when you can't see in the Infotools Harmoni table.
 In PowerPoint  the sample size is part of the labels on export to PowerPoint.
Show sample size option is not available:
 When using apply base Learn more about how to apply a base in the project tree. in the project tree at the element level results in different bases across the elements in the percentages dimension. The sample size option is only available when the uB of all elements is the same.
 In an analysis where the percentage base has been set to a dimension Learn more about percentage base zone. that is not plotted in the graph (i.e. not down or across).
 For space graph, plot graph and single number display visualise options.
 Pie graph  this will be available in the future.
Data Labels Summary Table
Icon  Name  Description 
Discover 
Selecting will enable the Discover option. Discover is a tool to profile a target group against a set of descriptor variables. 

Heat map 
The heat map colors the cells in the table to show the quintile distribution within a row. 

Show/Hide All 
Allows to show or hide values from a chart. 

Show Only Sig Diff. 
Allows to only show significant differences in a chart. 

Add/Hide Diff.  Allows to show or hide value gaps or differences from a chart.  
Optimise Label Placement  Optimize label placement avoiding overlaps that would make labels hard to read. This option is on by default as part of the smart graph setting.  
Show Sample Size 
Allows displaying the sample size next to chart axis labels or legend item labels based on the direction/dimension of the percentage base. The sample size is equal to the unweighted base. n=<uB>. Where uB belongs to an element in the direction of the base. 
9. Information Bar
The information bar (ibar) is located on the bottom right of the analysis pane. It provides contextual information about the settings that have been applied in your analysis. This information will flow through to any visualization, story, dashboard, export and the information bar in the mobile alerts dashboard.
Comments
0 comments
Please sign in to leave a comment.