Calculate the weighting factors

In the example used, there are more female respondents than male respondents, and the instructions show how to change the sample proportions to represent half the respondents as female and half as male, without changing the current base.

1. Click to display the Analysis Definition dialog for a table.
2. In Analysis, enter the variable that you want to weight. Build the table and note the totals for each code.

3. Calculate the Weighting Factors for each code. To calculate the value, divide the target figure (384÷2=192) by the actual figure.
   • Code 1, Male = 192 ÷ 152 = 1.263
   • Code 2, Female = 192 ÷ 232 = 0.828

Create the weight

1. Click to display the Weights window.
2. Click to add a new weight and specify the Weight Details as follows:
   • Name: WT1
   • Label: Weight male/female
   • Decimal places: 3
   • Number of codes: 2

3. Click to save the weight.

Use the weight in an analysis

1. Click to redefine the table or to display a new Analysis Definition dialog.
2. In Analysis, enter the variable that you want to weight.
3. Specify the Weight as WT1 (Q12), which tells Snap to use weight WT1 based on the results for Q12.
4. Select the Cells tab. In the Accuracy section set the Calculations d.p. to 3 decimal places to avoid rounding errors.
5. Click OK to build the table. Check that the table shows the Weighted Values correctly, in this example there should be 103 males and 103 females. Any errors are due to rounding errors or the incorrectly calculated weighting factors.

The table shows the Unweighted Base and the Weighted Base. You can exclude the Unweighted Base in the Tailor | Analysis option.

The weighting can be applied to any other tables and charts. However, there are 2 limitations that multiple-response variables cannot be weighted and you can only use one weight in an analysis.

The post Changing the proportions or balance of a sample group appeared first on SnapSurveys.

The following example shows exporting weights from a survey to a file. The weights can also be exported via the clipboard.

1. Click   to open the Weights window.
2. Select File | Export to display the Weight Export dialog.

3. Specify the Format as SNAP Interchange Format (SnIF).
4. Set the Destination as Clipboard.
5. Specify the names of the weights you wish to export in the Content field. (e.g. Score). Leave the field blank to export all weights.
6. Click OK to export the selected weight.
7. A confirmation message is displayed when the export is complete. Click Done.
8. Open the survey to which you wish to add the weight or select File | New Survey.
9. Click   to open the Weights window and select File | Import to display the Weight Import dialog.
10. Set the Source to File, entering the file name from the export, and click OK to import the weight(s).

11. A confirmation message will be displayed when the import has completed successfully and the new weight(s) will be added to the survey.

The post Moving weights between surveys appeared first on SnapSurveys.

A default report title is entered which you can update with a more suitable title.

This dialog has three different report formats:

• Detailed, single column
• Produces a report showing all the elements of each weight, as specified in the Weight Details window, in a single column format

• Detailed, double column
• Produces a report showing all the elements of each weight, as specified in the Weight Details window, in a double column format

• Summary
• Produces an overview-type report that matches the data in the Weights Window

The report settings can be changed by selecting Setup. Click Print and the report will be printed.

The post Weight reports appeared first on SnapSurveys.

You can use weights to subtract the negative from the positive values of a rating scale variable. By using this data in a statistics table, you can see whether the ratings were generally positive or negative.

This example assumes a five-code scale, in which the middle value is neutral, such as Q6a (speed of service). Because the rating scale only contains counts of responses, you need to set up a derived variable in order to calculate the statistics of the scored response.

1. Click to display the Weights window.
2. Click to add a new weight and specify the Weight Details as follows:
   • Name: WT6
   • Label: Weight pos/neg
   • Decimal places: 0
   • Number of codes: 5
3. Specify the Code Details:

4. Click to save the Weight.
5. Click to display the Variables window.
6. Click to add a new variable.
7. Specify the Variable Details:
   • Name: V10
   • Label: Calculated difference
   • Type: Derived (the variable will derive its data from other existing variables).
   • Response: Quantity (the response will be a numeric value for each case).
8. Specify the Code Details to combine the Weight with the Speed of service rating variable:

9. Click on to save the variable.
10. Click to display the Analysis Definition dialog to create a table.
11. Specify the Analysis as V10, which is the variable to be weighted.
12. Select Statistics table from the list for the Break (or enter STATS).
13. Click OK and the table will be built.

The Sum will be the difference between the positive and negative values.

The post Using weights to calculate the difference in rating appeared first on SnapSurveys.

The questions to be scored are 6.a to 6.e in the Crocodile Rock Cafe survey.

1. To check which question this is, open the Crocodile Rock Cafe survey and scroll down to question 6.

2. Click or press Ctrl+R to open the variables window.
3. Find the grid questions and look at the Code list so you know which labels seen by the respondent are assigned to which codes.

4. Open the Weights window by clicking or pressing Ctrl+W.
5. Click to add a new weight and specify the Weight Details as follows:
   • Name: scoreSat
   • Label: Satisfaction score -2 to +2
   • Decimal places: 3
   • Number of codes: 5
6. Specify the Code Details

7. Click to save your weight.

Using the score in analysis

1. Click to open the Analysis Definition window for a table.
2. Set the analysis value to be Q6.a ~Q6.e (this includes all questions from Q.6a to Q6.e)
3. Select the Summary Statistics tab and move Mean from the Available column into the Used column. Type the name of your score (scoreSat) in the Summary Score box. This tells Snap to create a summary of your analysis, scoring the codes with the specified values.
4. Click OK to display the table.
5. There is a column displaying the mean scored values.

The top line (Base) shows the total number of all respondents for all the questions. The mean satisfaction of everyone with all the services is 0.13.

The other rows show the average satisfaction for each service. You can see that people are generally positive about the Cleanliness, Quality of food and Choice of food and unhappy about the Speed of Service and Parking.

Including Don’t Know responses

When the question contains a Don’t Know response you can choose to include these responses in your analysis or exclude these responses from the analysis.

• Including the response add a score, such as 0, for the code associated with the Don’t Know response
• Excluding the response set the score to NR to give it a score of No Reply. This means that it won’t affect the analysis

The post Using a weight to convert responses into numerical values appeared first on SnapSurveys.

If you simply want to increase or decrease the number of cases in the survey, you could follow the steps below, which divides the target number by the actual number. In the example below, 384 is the actual number of respondents and 500 is the target number:

1. Click to display the Variables window.
2. Click to add a new variable.
3. Specify the Variable Details:
   • Name: V8
   • Label: Inflated cases
   • Type: Derived (the variable will derive its data from other existing variables).
   • Response: Quantity (the response will be a numeric value for each case).
   • Decimals: To access the Decimals field, click on . Change the figure from 0 to 3.
4. Specify the Code Details to calculate the higher base. As with the creation of a weight, the calculation is going to divide the target figure by the actual figure.

5. Click to save the variable.
6. Click to display the Analysis Definition dialog to create a table.
7. Specify the Analysis as Q2, which is the variable to be weighted.
8. Specify the Weight as V8.

9. Click OK and the weighted table will be built.

The inclusion of the weight will multiply the number of respondents in each response code by the value in V8.

You may also need to set the Decimal Places of Calculations to 3 to avoid rounding errors. Select the Cells tab on the Analysis Definition dialog and alter the Calculations d.p. number, if required.

The post Increasing or decreasing the size of a sample group appeared first on SnapSurveys.

In Snap XMP, RIM weighting provides the ability to set target totals for one or more variable so you can achieve, as closely as possible, a target distribution of results across the survey’s data.

Assessing sample proportions

First, use an analysis table to view the proportions of the survey dataset.

1. Click Analysis Table to display the Analysis Definition for a table.
2. In Analysis, enter the variable that you want to balance, for example, age ranges. Make sure Counts is selected as well as Respondents on the Base/Labels tab.
3. Click Apply to display the table.

Creating the RIM Weight

Next, create the RIM weight to change the proportion of each age range.

1. Click Analysis Variables  to open the Analysis Variables window.
2. Click New Analysis Variables Item  , which displays a menu of analysis variable types to choose from.
3. Click New RIM Weight to open the RIM Weight window. Note there is an initial error in the status bar as there are no variables references yet.
4. In Name, enter a name which describes the RIM weight.
5. In Label, enter a description of the RIM weight.

6. Click Add Variable  to open the Select Variable dialog.
7. In the Name drop-down, select the age range variable as the weighting variable, and click OK.

The grid shows the ratio, expected count and percentage as well as the actual count and percentage.

Even distribution example

This example creates an even distribution for each age range.

In Target total, select Valid cases the required option. Valid cases bases the RIM weight on the valid data responses in the survey. The default ratio is 1 to give an equal distribution for all variable codes. 

After entering the ratios you need to build the RIM weight.

Demographic distribution example

This example creates a distribution based on a census result.

In Target total, select Valid cases the required option. Valid cases bases the RIM weight on the valid data responses in the survey. The default ratio is 1 to give an equal distribution for all variable codes. 

Change the ratio to that of the population for each age range.

After entering the ratios you need to build the RIM weight.

Build the RIM Weight

The status of the RIM Weight displays in the status bar at the bottom of the RIM Weight window. When the RIM Weight is created the status shows as Not built.

1. Click Build RIM Weight  on the RIM Weight toolbar to build the RIM weight.
2. The Status changes to show Built. If there is an error in the RIM weight the status bar will show an error message and the RIM weight is not built.
3. Click Save  to save the RIM weight.

If the survey receives more responses after you have built the RIM weight, then you will need to rebuild the RIM weight manually as the weight is not updated automatically.

Using the RIM Weight

Create the table analysis using the RIM weight to change the proportions of the sample group.

1. Open the Analysis table created for the age range.
2. In Weight, enter the name of the RIM weight.

4. Click OK to build the rim-weighted analysis.

The RIM weight can be saved and used on any other table or chart.

The post Using RIM weighting to balance sample proportions appeared first on SnapSurveys.

1. Click to display the Analysis Definition dialog box.
2. Specify the Analysis as Q6a to build a table of Q6a, “How do you rate the speed of service?”
3. Specify the Weight as Q3a, “How many people were in your party today?”

4. Click OK and the weighted table will be built.

The inclusion of the weight will multiply the number of respondents in each response code by the value in Q3a for each respondent, namely the number of adults in the party.

The table shows the Unweighted Base and the Weighted Base. The Unweighted Base can be excluded using the Tailor | Analysis option.

The post Counting items other than respondents appeared first on SnapSurveys.

• To create means in mean score tables
• To count items other than respondents, such as, amounts of money
• To change the proportions/balance of a sample group
• To increase/decrease the size of a sample group
• As part of a calculation

Multiple Response variables cannot be weighted and you cannot use more than one Weight in a table.

When tables have weights applied to them, and statistics produced, such as significances and confidence intervals, they are calculated using the weighted data not the unweighted.

Defining a weight

Weights consist of a series of codes. They should have the same number of codes as the variable that you are applying the weight to.

1. Click to display the Weights window.
2. Click to add a new weight or on the Weights window toolbar to display the Weight Details dialog for an existing weight.

The image below shows a weight used to transform a five-code rating score into positive and negative values.

3. Enter a unique name in the Name field.
4. Enter a description of your weight in the Label field.
5. Enter the number of decimal places used in a calculation in the Decimal places field.
6. Enter the Value for each code in your weight. (Press Tab to create a new code). If you wish to exclude a variable code from the weighting, (e.g.”Don’t Know”) set its value to NA.
7. Click to save the weight.

The post Using weights appeared first on SnapSurveys.