As a Crosstabs user, you can define and implement weighting schemes in Crosstabs without the aid of a Data Processor or external programs. Weights can be easily rerun if new data is collected, and you can run weights on one or more subsets of your data.
RIM weighting is only available for compat 135+ surveys.
What is RIM Weighting?
Random Iterative Method (RIM) Weighting allows researchers to adjust multiple characteristics in a dataset while keeping the different characteristics of the data itself proportionate as a whole. This means that you can run separate weights by data subsets and tailor your weighting schemes by wave, country, or any other data segment available within your survey.
There are two options for adding RIM Weighting to your survey data: by selecting questions and assigning weighting percentages to each response in the provided tool or by uploading a pre-designed weighting file. Both options are accessed via the Crosstabs menu after you have run a report.
This article details how to create and apply a weighting variable. Click here if you would like to learn more about uploading an existing weighting scheme.
Accessing RIM Weighting in Crosstabs
You will need report:edit access to access the RIM weighting features within Crosstabs.
There are two ways to manually enter weights for your project. You can either click "New Weighting Scheme" within the Crosstabs "Actions" menu:
Or, you can click "New" under "Weighting" within the report settings menu:
Regardless of the access method you choose, the Weighting Scheme Editor will appear in a new window.
Overview of the Weighting Scheme Editor
The Weighting Scheme Editor allows you to create and apply a weighting variable within Crosstabs. There are a variety of tools available here to help you customize your survey weights and run test cases to ensure that your final reports are calculated exactly as needed.
Here is a quick outline of the functions available within the Weighting Scheme Editor:
1. Variable Name: This is where you will name your weighting variable for use in the reports and data downloads.
2. Learn More: Clicking here will direct you back to this guide at any time during weighting.
3. Filters: All selected filters will be displayed in this box. Clicking "Edit" will allow you to add or edit filters.
4. Weighting Dimensions: Use this drop-down menu to select which survey question you would like to use for data weighting.
Only closed-ended, single-dimension questions and variables are displayed in the Basic Weight Editor. Click here to learn how to add weights to more complex question types using the Advanced Weight Editor
5. + Dimension: Clicking here will create an additional drop-down menu, allowing you to select another question to use in your weighting scheme.
6. Switch to Advanced Weight Editor: Clicking here will direct you to the Advanced Weight Editor, where you can programmatically adjust the weights and filters for your weighting scheme. Any weights and filters you've added in the Basic Weight Editor will remain intact. Click here to learn more about using the Advanced Weight Editor.
7. Efficiency icon: Depending on your current weighting scheme, this can be a green checkmark or a red exclamation point. Hovering over this icon will allow you to view the estimated efficiency rating for the current weighting scheme, along with any error messages. Clicking the hyperlink displayed under the efficiency rating will open a new window with the results of computed weights.
8. Save: Clicking here will add your weighting variable to the Crosstabs "Weighting" menu, where it can be applied to the report at any time.
Building a Weighting Scheme
To start building out your new weighting scheme, you will first need to provide a name for the weighting variable to be applied. Enter this at the top of the editor window, using the "Variable Name" box:
Once you have a name, you can use the drop-down box under “Weighting Dimensions” to start adding weights. To add a survey question to your scheme, click the box and select the desired question from the list:
The drop-down box has a limit of 199 questions. Surveys with 200+ questions will not display anything in the drop-down, and will not include a datamap when using the Advanced Weight Editor.
Then, enter your weights as a percentage for each of the question's response options. The system will sum all responses as they are entered and display a dynamic total for your reference:
The Weighting Scheme Editor will not accept decimal values for weights. To use decimal values within weights, switch to the Advanced Weight Editor.
To apply multiple questions to your weighting scheme, click "+ Dimension" and select an additional question from the new drop-down that appears:
Repeat this process to add as many questions as needed.
You can use the "Filters" box to add filters based on respondent status and any previous Crosstabs reports you've created. This can be helpful for applying weights only to certain question, row, or column data, or to specific logic blocks.
To add filters to your weighting scheme, first click "Edit":
Then choose a respondent status and any other saved Crosstabs report you'd like applied to your weighting scheme:
The "Respondent Status" filter allows you to filter the data by respondent status:
- Qualified: Respondents who have completed your survey as a qualified user.
- Terminated: Respondents who have completed your survey as a terminate user.
- Overquota: Respondents who have completed your survey as an overquota user.
- All Complete Respondents: Respondents who have completed your survey with any completion status.
The "Additional Filter" field allows you to add filters based on previously run Crosstabs reports and is pre-populated with Crosstabs reports you or your project collaborators have already created. If you or your teammates haven't created any Crosstabs reports, this field will appear locked as "None".
Crosstabs reports must be set to "View Only" or "Editable" access in order for other project collaborators to view them.
Click here for more information on creating reports within Crosstabs.
It is possible to add report logic to further customize your weighting scheme. For advanced scheme building options, click "Switch to Advanced Weight Editor" at the bottom left of the screen:
To learn more about programming weights manually using the Advanced Weight Editor, click here.
Weights are run automatically as you build out your weighting scheme, but are not applied until you click "Save" at the bottom right. Each time you add a weighting dimension or filter, the system will automatically re-run your weighting scheme and display an icon at the bottom of the screen indicating your current scheme's efficiency rating:
The icon's appearance will change depending on the estimated efficiency of your current weighting scheme. A green checkmark means that your weighting scheme has met at least 70% efficiency and should be safe to apply to your data. A red exclamation point means that your weighting scheme has not met at least 70% efficiency.
While it is possible to apply weighting schemes with low efficiency ratings, we recommend achieving an efficiency of 70% or higher before applying a weighting scheme to your data.
Hover over the icon to view the estimated efficiency for your current weighting scheme, and click the hyperlink there to view the full calculation as it will be applied to your data:
Viewing Calculation Details
Clicking the "Click This Link for Details" hyperlink will display the system calculations for your scheme's current Weight Factor Efficiency Output and Root Mean Square (RMS).
Evaluating the Efficiency Output
Viewing these calculations allows you to evaluate your efficiency rating before actually applying any weights to your data. If the efficiency output is not what you expected, you can adjust your weighting scheme and re-calculate it as many times as you like.
The system will calculate the efficiency output as close as possible, so the maximum difference will never be more than 0.5%.
Evaluating the Root Mean Square (RMS)
The "Details" box also provides the RMS and number of iterations. The Decipher Weighting tool will run a maximum of 10 iterations of the RIM weighting algorithm, stopping only if convergence is reached. Convergence is reached once the RMS is less than 0.000005.
For example, the results of the weighting scheme above indicate convergence was met after one iteration and, therefore, the weighting scheme is considered a success:
If the weighting scheme does not reach convergence after 10 iterations, it is considered a failure. A failed weighting scheme can still be applied, but researchers should be cautious about drawing conclusions from a data set with a failed weighting scheme applied.
Once you are satisfied with your weighting scheme, click the "Save" button at the bottom right to add your new weighting variable to Crosstabs:
If your Total does not equal 100%, the "Save" button will be disabled and you will not be able to apply your current weighting scheme.
To apply a saved weighting variable, return to the report settings page within Crosstabs. Then select the desired weighting variable under the "Weighting" menu, and click the "Apply" button at the bottom of the page:
After you add your weighting variable using the Weighting Scheme Editor, it will be available within the Crosstabs report settings, and you can toggle it on or apply a different weighting scheme at any time. To create another weighting scheme, just repeat this process for each individual variable.
Editing Existing Weighting Schemes
Once you have applied a weighting variable within Crosstabs, you can return to edit it at any time, or even remove it altogether. To view the options for existing weighting variables, just hover over the variable name within the "Weighting" drop-down in the report settings menu: