Groups
When administrating a panel, you usually do not work with individual panelists, but with larger groups of panelists whose members are selected according to specific criteria.
People offers you diverse options for creating groups (also called “subpanels”).
You can select panelists with specific properties from the result data record of a survey, extract their “u_other_id” or “pseudonym” into a list, and then import this list to create a new group.
Groups in People are static, i.e. they are created at a time X and will not be dynamically filled. However, you can update the composition of groups by repeating the creation process or by running a daily automatic update according to selfdefined rules.
You have various options for editing the members of a group as a whole: For example, you can send e-mails to all members, change the status of all members or assign bonus points to them.
The groups are used as the basis for drawing project-specific samples.
The larger a panel becomes and the longer you work with it, the more groups will be formed. To improve clarity, it is possible to merge groups into panel group categories. These categories are merely organizational aids; they are of no further importance to the process of creating groups and drawing samples.
Please note: The number and size of groups can have a negative effect on the performance of your installation. The following guiding values provide orientation:
It is recommended to use not more than 1.000 active groups per installation. Therefore, please archive superfluous panel groups regularly.
The number of archived groups which are kept available on an installation should not be higher than 1,500.
The number of panelists’ group memberships should not be higher than 20,000,000.
Groups
If you open the People → Groups menu, you will see an overview of the existing groups. The table lists the existing groups, including their title, description, category, group status, size, creation date, and the name of the user who created the group.
As usual, simple and extended search are available. Using extended search, you can search the table for a search term, a group category or a combination of both. Browsing for a search term will return all groups containing the term entered, i.e. entering the search term “calendar week” will output all groups for specific calendar weeks (“calendar week 1”, “calendar week 2”, etc.).
You can sort the list by title and by creation date in ascending or descending order. To do so, click on the title of the respective table column.
Actions
The buttons located above the group list enable you to create new groups and sort the existing groups.
Create group: Opens the dialog for creating a new group.
Import group: Groups can also be created via import.
Manage categories: To improve the overview, sort the groups of your panel by topic.
Via the icons in the “Actions” column you can edit individual groups:
Edit group: You can change the category, title and description of a group.
History: Opens the tab which lists all actions performed when creating the group and where parts of the work process can be repeated.
Copy group: Clicking on this icon allows you to copy a group.
Move group to archive: You can use this feature to archive superfluous groups
Delete group: If you click on this icon and then confirm the pop-up message, the respective group will be removed.
You can also edit multiple groups at once. To do so, select the desired groups using the “Actions” checkbox and then select the desired action from the drop-down list, as usual.
Delete groups: The selected groups will be deleted.
Empty groups: All members will be removed from the selected groups.
Archive/Reactivate groups: Depending on their state, the selected groups will either be archived or reactivated.
By default, each list of groups contains the group “Default panel group”. This group contains all panelists and, in contrast to the groups created by users, is continuously updated. It cannot be deleted.
Organizing Groups into Categories
Group categories allow you to sort groups by topic and arrange them in a clearly structured way. Especially for large panels with many groups, it will be helpful if you are able to narrow the list of groups to specific categories.
Managing group categories
The dialog for managing the group categories is opened via the Manage categories button in the group overview. It will direct you to an overview of the already existing categories along with their corresponding labels, descriptions and number of assigned groups. As usual, you can either browse the list of categories, or sort the list by simply clicking on the title of the respective table column.
Via the Create category button, you can open a dialog which allows to create new categories. All you need to enter is category title and description.
To edit existing categories, use the icons in the “Actions” column:
Edit: You can change the title and description of a category, via the same dialog that was used for creating the category.
Delete: If you click on the icon and confirm the pop-up message that follows, the respective category will be removed.
Assigning groups to categories
You have two options to assign a group to a category:
When creating a new group, select the appropriate category in the drop-down list.
To change the category of an existing group, open the properties dialog of this group via the Edit group icon in the list of groups (People → Groups) and change the selection in the drop-down list.
Marking Groups for Better Monitoring
One of the purposes of group creation is to simplify the tracking of groups of panelists with characteristics that are important for the panel. The “Monitored groups” feature allows you to select such groups:
These groups will be specially marked in the list of groups.
Using the drop-down list “Monitored groups” in the extended search function of the list of groups, you can limit the display to the monitored groups or exclude them from the display.
The group history shows the development of the goup size over time (Breadcrumb: People → Statistics → Panel overview → tab Group history).
To mark a group as described above, find the desired group in the People → Groups menu and click on the Edit group icon. This will open the dialog familiar from creating groups. Tick the checkbox “Mark group as 'monitored group'” and confirm by clicking on Save group.
Creating Groups
You can create any number of groups.
Click on the Create group button in the Groups menu.
Select the category to which the group is to belong. In the example, select the “Default” category which contains all the groups of the panel.
Enter the title and description into the form. Both of these are for internal orientation and will not be externally visible to the panelists.
Optionally, you can mark this group as “monitored group”.
Confirm by clicking on Create group.
You have now created an empty group with no participants, which can now be filled using various filters and set-generating algorithms.
Importing Groups
The function for importing new groups is used if you wish to form a new group, e.g. based on characteristics in the result data record of a survey. Please proceed as follows:
Decide which linking variable is to be used. An overview of the available variables is given in the following section.
Extract the linking variable for those panelists you wish to include in the group from the source record. List the values of the listing variable for all desired panelists vertically in a CSV file.
Click on the Import group button in the People → Groups menu.
Set the desired category. Enter a title for the group and add a description.
Optionally, you can mark this group as “monitored group”.
Select the linking variable.
Locate the file on the hard disk.
Confirm by clicking on Next.
Now, a group will be created whose members are the panelists with the pseudonyms you have uploaded.
Please mind: empty rows or the value “0” will be ignored during import.
Linking variables
Linking variable | Description | Export data record |
---|---|---|
u_other_id | During participant import, you can use this field to save any identification number that serves to identify the respective participant. | Address record |
u_account | Account name | Address record |
pseudonym | Internal identification number within the panel which allows you to uniquely identify each panelist. | Address record, master data record |
panelist_code | Panelist code | Master data record |
u_email | E-mail address | Address record |
Editing Groups
Create a new group or click on the title of an existing group in the overview. This opens the edit dialog of the respective group.
It contains several tabs with information and actions sorted by topic:
Manage members
Basic information on the group, list of members, set operations, functions for editing the members, and mail functions.
Apart from the basic infromation on the group, the Manage members tab also features the list of members, various functions for further editing group structure, and actions that can be applied to individual or selected members.
Actions for editing the group structure
You can further modify the group structure using the buttons located above the list of members:
Import members: It is possible to subsequently import further members into the group. The import is performed in the same manner as the import of entire groups by uploading a linking variable.
Add groups: With this set operation you can add already existing groups to the current group. If you click on the Add groups button, a list of the existing groups will be displayed. Select the group or groups whose members you wish to add.
Subtract groups: With this set operation you can subtract already existing groups from the current group.
Generate intersection: This set operation allows you to generate intersections from your group and existing groups, i.e. your group will be reduced to those members who are also members of specific other groups.
Draw random sample from group: You can further restrict the group at random.
Empty group: All group members will be deleted following a confirmation prompt.
Actions for editing the members
In the list of members you will find all the panelists of the group with their names, panel statuses, e-mail addresses and number of bonus points. By clicking on the name or on the Edit icon you can switch to the detail view of the respective panelist.
The following actions can be applied to all or individual members. To do so, select the respective members using the “Actions” checkbox and then select the desired action from the drop-down list.
Assign bonus points: Bonus points can be assigned to or subtracted from selected members. This action is only available if you have write rights for “panelbonus”.
Write e-mail: Opens the standard mail template for contacting the selected members. If you are operating a community panel, you can post messages into the group members’ inbox on the website without dispatching e-mails.
Remove panelists from group: The selected members will be removed from the group.
Delete panelists: The selected panelists will be deleted completely with all their address and master data. Please mind: This action cannot be reversed.
Change status to: The panel status of the selected members can be changed. Please mind: the action “Delete panelists” is used to assign the status “deleted”.
Filter
The group structure is defined using filters.
The filters presented in the following chapter are used to select the group members.
These filters are usually applied to filter out panelists from a already filled group, restricting membership to those group members who fulfill the filter criteria (Filter mode: “Create intersection with current panel group”). E.g. if the filtering instruction “restrict to panelists with panel status ,active‘” is applied to an already filled group, only the group members with panel status “active” will remain in the group.
If you want to filter on the whole panel, adding all panelists who fulfil the criteria to the group, the option “Apply to all panelists in panel” should be activated.
Performance data filter
The performance data filter selects the group members using performance data from the default group (= overall pool of panel participants). Performance data is data based on the participation behavior of the panelists.
Filter criterion | Explanation |
---|---|
Number of invitations | Evaluates the tracking variable “track_num_starte”. |
Number of survey starts | Evaluates the tracking variable “track_num_completed”. |
Number of completed surveys (based on variable track_num_complete) | Evaluates the tracking variable “track_num_completed”. |
Number of bonus points | Evaluates the current sum of the bonus points in the panelist’s account. |
deprecated: Number of completed surveys (based on variable n_poll) | Evaluates the “n_poll” system data item which is increased by 1 when the final page of a panel survey is reached, i.e. only completed surveys are taken into account. Please mind: If tracking is activated, evaluation of “track_num_completed” (see above) provides you with more precise information. |
Ignored invitations since <date> | If a panelist ignores a survey, he will have disposition code 11 or 12 in a sampling, i.e. he has not seen the first page of the survey. |
Last completed interview at least / not longer than x months ago | Evaluates the system data item “last_poll_date”, which is filled with the current system date when the final page of a panel survey is reached. |
Date of entry should be between (including) <date> and (including) <date> | Evaluates the system data item “penter_date”, which is set while filling in the registration or during manual import. |
Panel member since x months | Calculates the time for which the panelist has been a panel member based on the system data item “penter_date”. |
Date of last panel status change not longer than x months ago | Interprets the system data item “pstatus_date”, i.e. the last change of panel status. This value is currently set only during manual import. |
Way of entry | Selects the panelists according to the way of entry that has been chosen. |
Panel status | Current role and activity of the panelist. |
E-mail verification status | Status of the double opt-in e-mail verification. |
PCI value | Describes the reliability of panelists. |
Mark | Evaluates whether the panelist was marked via the search or not (system data item “marked”). |
Master data filter
In most panels, master data surveys are conducted at the beginning of each new panel membership. The master data filter checks your group for characteristics collected during this first survey.
Clicking on the Master data filter button will display a listing of all master data defined in the system. u_gender, pstatus and site_id are available as well.
Optionally you can restrict the selection to a specific master data category.
Select one or more variables from the master data list, in the example above the year of birth. The list is often very long. Therefore, the “Tick all checkboxes” checkbox is available as an entry aid in the upper section of the form: Ticking this checkbox allows you to select all master data via JavaScript and manually de-select individual fields.
Confirm by clicking on Continue.
You will now see the characteristics of master data with answer categories (“int” variable type). Select the characteristics the panelist has to meet in order to be included in the group. In the example shown, you would have to choose all age groups < 30.
For text variables and decimals (variable type “text” and “float”), an input field will be displayed instead. Here, you can enter the desired characteristic for your master data variable. The available operators are “equal”, “not equal”, “greater than”, “less than”, “less than or equal”, “greater than or equal” and, for text variables “contains”.
After that, confirm by clicking on Continue.
Sample data filter
The sample data filter allows you to limit the group to members of a specific project with a given disposition code. If the disposition code is not relevant for your purposes, you can also select all available disposition codes instead.
Project filter
Using the project filter you can restrict the groups first to the members of a project you have chosen. Then, you can filter the group using specific answer values from the selected project.
When you open the filter menu, a list of the projects including project type and status will be displayed. Click on the title of the desired project. Select the answer values to be used for filtering. During the filtering process, all participants who have given the corresponding answer in the project will be assigned to the new group.
Quick poll filter
The Quick poll filter allows you to use the results of panelist polls for creating panel groups.
Open the filter dialog and select one or more quick polls. In the next step, select one or more answer options of these polls. Depending on the filtering mode, the filter will then select all those panelist from the current group or from the entire panel who have given at least one of the listed answers in one of the polls.
Grouping filter
The grouping filter (previously: “alternative grouping filter”) allows you to combine a variety of filter criteria:
The filter has access to all address and master data.
The following system data can be used: pstatus, pinput, last_poll_date, pcredit_points, penter_date, n_poll, reg_code, doub_stat, md_update_date, md_invite_date, md_num_invites, md_num_updates, pci, site_id, date_last_login, remode_addr, remote_host, last_mod_date and last_mod_user.
Tracking variables can be used as well.
It is possible to perform date calculations.
Both AND conjunctions and OR conjunctions can be used.
You can define complex filter conditions by combining conjunctions.
Administrating grouping filters
You can access the previously defined grouping filters of your panel and the functions for creating new filters
from the Filters tab in a group’s edit dialog via the Grouping filter button.
via the People → Groups → Grouping filters menu.
Defining grouping filters
The filter editor’s structure and function structure corresponds to the condition editor in the EFS Survey project area. The procedure for defining conditions is identical, too: Click on the Create filter condition button and define the filter condition by selecting the desired variable, condition, code and conjunction. In addition, you have the following options:
You can save the filter in order to use it again in future grouping and sampling processes.
A list allows you to view, edit, apply and delete the filters that have already been created.
Clicking on the Test filter condition icon in the list of filters triggers a checking mechanism which determines the number of panelists a grouping filter would apply to.
Date calculations
It is possible to perform date calculations in the grouping filter. Wildcards are used to define dates. To access a list of the available wildcards, click on the Question mark icon in the condition editor.
Examples:
#NOW#: Current date and time.
#LOCAL_NOW#: Current date and time in local panel time.
#CURRENT_QUARTER_START#: Beginning of the current quarter.
#CURRENT_HOUR_START#: Beginning of the current hour
.
More complex date calculations can be realized by entering the number of days, weeks etc. to be added or subtracted directly inside the wildcard.
Subtraction and addition are permitted. As filters usually refer to past events, in most cases, subtraction is used.
Permitted units are d = days, m = months, y = years, w = weeks and h = hours. If no unit is explicitly named, the system will use days as default unit.
Examples:
#WILDCARD#: The date defined by the wildcard is inserted directly.
#WILDCARD-3#: The date minus 3 days is used. For example,
#PREVIOUS_WEEK_START-3# subtracts 3 days from the beginning of the previous week.
#WILDCARD-3d#: The date minus 3 days is used.
#WILDCARD-3w#: The date minus 3 weeks is used.
#WILDCARD-3m#: The date minus 3 months is used.
#WILDCARD-3y#: The date minus 3 years is used.
#WILDCARD-3h#: The date minus 3 hours is used.
Import filter
The import filter allows to import any number of the characteristics of a numeric master data variable or address variable and use it as criterion for group membership. Optionally, you can restrict the size of the resulting group further. Optionally, the size of the resulting group can be restricted further.
This filter is useful if, for example, you want to group panelists with specific zip codes, e.g. 20,000 of 40,000 US zip codes. All you need to do is create a list of the zip codes required and match it to u_zip during import - the system will find all suitable panelists.
Formatting the import file
The import file should be a simple CSV list, containing those characteristics of the linking variable which are to be selected.
Applying the import filter
Please proceed as follows:
Open the Filters tab in the edit dialog of the group.
Click on the Import filter button to open the new dialog.
Select the appropriate linking variable. All numeric master data variables and the address variable u_zip are available.
Choose the character set.
Choose the import file.
In the “Maximum size of random selection from file” field, you can restrict the size of the resulting group further: If you use this option, a random selection will be applied to the panelists found. If you do not enter a maximum size, all panelists found will be adopted into the group.
Select the appropriate filter mode. As usual, you can choose either “Apply to all panelists in panel” or “Create intersection with current panel group”.
Click on Continue.
Now, the system identifies the suitable panelists and adds them to the current group resp. creates an intersection with the group.
History
On this tab, you can view previous actions and repeat parts of the work process. On the History tab, all the actions that you performed in a group are logged. It is possible to repeat the work process in whole or in parts. To do so, select the action up to which the process is to be repeated, and click on Repeat grouping.
Tracking
Tracking data of the group. See Chapter "Panel Tracking".
Statistics
On the Statistics tab, you can create statistics for the current group. The desired evaluation variable and two split variables can be selected.
Usage
In order to check the dependency when deleting panel groups, each group has a "Usage" tab, which shows the EFS functions, where this group is used.
Grouping filters
You can access the previously defined grouping filters of your panel and the functions for creating new filters
from the Filters tab in a group’s edit dialog via the Grouping filter button.
via the People → Groups → Grouping filters menu.
The filter editor’s structure and function structure corresponds to the condition editor in the EFS Survey project area (see “EFS Survey”, Chapter 7.1.3). The procedure for defining conditions is identical, too: Click on the Create filter condition button and define the filter condition by selecting the desired variable, condition, code and conjunction. In addition, you have the following options:
You can save the filter in order to use it again in future grouping and sampling processes.
A list allows you to view, edit, apply and delete the filters that have already been created.
Clicking on the Test filter condition icon in the list of filters triggers a checking mechanism which determines the number of panelists a grouping filter would apply to.
Administrating grouping filters
You can access the previously defined grouping filters of your panel and the functions for creating new filters
from the Filters tab in a group’s edit dialog via the Grouping filter button.
via the People → Groups → Grouping filters menu.
Defining grouping filters
The filter editor’s structure and function structure corresponds to the condition editor in the EFS Survey project area. The procedure for defining conditions is identical, too: Click on the Create filter condition button and define the filter condition by selecting the desired variable, condition, code and conjunction.
In addition, you have the following options:
You can save the filter in order to use it again in future grouping and sampling processes.
A list allows you to view, edit, apply and delete the filters that have already been created.
Clicking on the Test filter condition icon in the list of filters triggers a checking mechanism which determines the number of panelists a grouping filter would apply to.
Date calculations
It is possible to perform date calculations in the grouping filter. Wildcards are used to define dates. To access a list of the available wildcards, click on the Question mark icon in the condition editor.
Examples:
#NOW#: Current date and time.
#LOCAL_NOW#: Current date and time in local panel time.
#CURRENT_QUARTER_START#: Beginning of the current quarter.
#CURRENT_HOUR_START#: Beginning of the current hour.
More complex date calculations can be realized by entering the number of days, weeks etc. to be added or subtracted directly inside the wildcard.
Subtraction and addition are permitted. As filters usually refer to past events, in most cases, subtraction is used.
Permitted units are d = days, m = months, y = years, w = weeks and h = hours. If no unit is explicitly named, the system will use days as default unit.
Examples:
#WILDCARD#: The date defined by the wildcard is inserted directly.
#WILDCARD-3#: The date minus 3 days is used. For example,
#PREVIOUS_WEEK_START-3# subtracts 3 days from the beginning of the previous week.
#WILDCARD-3d#: The date minus 3 days is used.
#WILDCARD-3w#: The date minus 3 weeks is used.
#WILDCARD-3m#: The date minus 3 months is used.
#WILDCARD-3y#: The date minus 3 years is used.
#WILDCARD-3h#: The date minus 3 hours is used.
Update Rules
The static panel groups can be updated automatically or manually according to selfdefined rules. Open the People → Groups → Update rules menu. If update rules have already been created, you will find them in an overview table in the bottom area.
The table contains the following information for each rule:
Name of rule
Number of actions: List of the actions that are to be executed when updating.
Group: Indicates the name of the group to which the update applies.
Group size: Indicates the number of panelists.
Automated: If this option is activated, the update will be executed automatically in a user-defined interval.
Last executed automatically: Indicates the date and time of the last update.
The icons in the “Actions” column provide you with the following editing options:
Edit: Opens the menu for editing the update rule, see the following chapter for a description.
Copy: The rule is copied.
Apply rule now: Triggers the respective update.
Delete: The rule will be deleted after a confirmation window.
Creating New Rules
Defining basic parameters
Click on the Create update rule button.
Provide the following details:
Name of rule
Comment
Group category: Name of the category in which the group to be updated is located.
Panel group to be updated: Indicates the name of the group to which the update should apply.
Execute rule automatically: If this option is activated, the update will be executed automatically in a user-defined interval (default setting: no).
Start time: Date and time can be set.
Interval type: The possible intervals are hours, days, weeks, months and years.
Interval value: Specifies the size of the interval. For example, with the interval type “days” and the interval value “3” the update rhythm will be every three days.
Defining actions
Specify the basic parameters first, then define the actions to be executed in an update. There is more to defining actions than simply choosing the desired action. Often an advanced configuration is required, for example execution conditions. The following actions can be selected:
Action | Description | Parameter to be defined |
---|---|---|
Empty group | This action empties the panel group. | |
Apply group filter | This action applies a grouping filter to the current group. | The grouping filter should be created. In addition the “filter operation” field is used to set what should happen to the panelists that fulfill the filter condition: You can add to and remove from the group, an intersection between the group members to date and the panelists determined by the filter can be formed. |
Add group | This action adds panelists from another panel group to the group. | The source group must be chosen. |
Subtract group | This action removes panelists that are present in another group from the current group. | The source group must be chosen. |
Create intersection | This action forms the intersection of the current group and another panel group. | The second group must be chosen. |
Draw random sample | This action draws a random sample with a definable size from the current group. All panelists that are not part of the random sample will be removed from the group. | The size of the random sample must be chosen. |
Give bonus points | This action gives all panelists in the current group bonus points. | The number of bonus points must be chosen. A reason can be entered. |
Deduct bonus points | Use this action to deduct bonus points from all panelists in the group. If the deduction results in a negative value, the deduction will be reduced for that panelist, so that the resulting count of bonus points will be 0. | The number of bonus points must be chosen. A reason can be entered. |
Send mail | This action sends an e-mail to all members of the group. | A suitable mail template should be created. |
Execute automatic recoding | This action executes the automatic recoding of master data for all panelists in the group. | Further detailed information can be found in Chapter "Recoding Master Data". |
Execute recodings from recoding group | This action executes all recodings from the recoding group selected. | More detailed information on recoding groups can be found in Chapter "Recoding Master Data". |
Change panel status | This action changes the panel status of the panelists in the current group. | You can choose from the panel statuses “Admitted temporarily”, “active” and “inactive”. |
Modify master data item | This action changes the value of a master data item. | Master data item and characteristic must be chosen. |
Increase / decrease value of master data variable | This action increases or reduces the value of the selected master data variable. | The master data item and the size of the change interval must be chosen. |
Add to sample | This action adds the users chosen during the previous steps to the selected sample. | The sample must be chosen. You must also specify whether panelists are to be added only if the survey is in the field and active. This option is activated by default. The following parameters should be defined:
|
Send survey invitation mail | This action will send a survey invitation mail or a reminder mail to panelists from a specific sample in a panel survey or master data survey. Mails will be sent only to panelists who are already in the sample and who have a certain disposition code in it. | The following parameters should be defined:
|
Copy from survey to group | This action adds panelists from a selected survey with a given disposition code to a panel group. | The following parameters should be defined:
|
Invalidate e-mail addresses | This action allows you to invalidate e-mail addresses of panelists by attaching the content of an arbitrary master data item to the e-mail address. This is advisable, for example, in the following cases:
| The master data item to be attached to the e-mail addresses must be chosen. |
Execute quota | Updates the actual values of a panel quota. | Quota, see Chapter "Statistics". |
Set PCI value | This action assigns the selected PCI value to all panelists in the group. | PCI value, see Chapter "Contacts". |
Running an Update
Triggering the update manually
You can trigger the execution of an individual rule manually by clicking on the Apply rule now icon.
Triggering the update automatically
If you tick the “Execute rule automatically” checkbox in the menu of a rule and select a start time, the respective rule will be processed at the set time.
© 2024 Tivian XI GmbH