Crosstab Editor Options
Overview
This page describes the options available in the crosstab editor. For more on the calculations options in MarketSight crosstabs, click here.
Turn off Auto-Update in the Crosstab Editor
- Open the crosstab editor
- Deselect the auto-update button in the right-hand corner of the editor
- Click the "Update" button as desired to refresh the preview window
Edit Variables Directly from the Crosstab Editor
- Right-click the variable of interest within the left column of the designer
- Make any desired changes in the Edit Variable dialog
- Click OK to confirm and return to the crosstab editor
Freeze Column Headers in the Crosstab Editor
- In the gray toolbar, check the box for "Freeze column header," or
- Click on your name in the upper right corner
- Click on Options
- Select the Crosstabs tab
- On the Basic tab, check the box for "Freeze column header"
- Click "Save All Options" at the bottom of the page
Design Tab Options
| Option | Description |
|---|---|
| Name | Title of the crosstab. Required field. |
| Description | Optional description field to detail the variables the crosstab includes or why it might be important. |
| Variable Display Boxes | Choose variables from the left box and drag them into the columns, rows, filters, index or weighting areas on the right.Clicking the + sign to the left of a category will expand that category to show variables it contains. Clicking the + sign next to a variable name will expand the list of values for discrete variables, or display value ranges for continuous variables. |
| Display | This drop down will cause the dialog and crosstab to show variable names such as "Job Occupation," variable codes such as "job_occ," or variable codes and names. |
| Sort | Sorts the list of variables in the designer, displaying them either alphabetically or the order in which they appear in the survey. |
| Search | Search for variables by using keywords. |
| Collapse/Expand All | Collapse/Expand all will minimize or expand the variable categories in the left box. |
| Columns Sub-Tab | |
| Number of Columns | Indicates the total number of columns in the crosstab design, which includes the totals column, the divider columns between tables, and the column containing the row variable's labels.There is a limit of 20 columns for pairwise testing, and 50 columns for contrast testing.PDF exports are limited to 20 columns. |
| Show Totals | Displays a totals column in the crosstab when checked. |
| Include Missing Values | Checking this will display user-missing values for the column variables. |
| Options | Appears to the right of a discrete column variable.Click to edit the column header labels and reverse the order of variable values. Settings here are stored in the crosstab and do not permanently change the variable itself. |
| Create Value Ranges | Appears to the right of a continuous column variable.Click to specify value ranges for the column variable. |
| Up/Down arrows on the side | Click to reorder variables in the list. |
| Rows Sub-Tab | |
| Number of Rows | Indicates the total number of rows. Calculated by taking the sum of the number of variable display options (i.e., mean, median) selected for each variable, plus the product of the number of values times the number of value display options (i.e., count).Variables that have all values deselected add no rows. |
| Include All Variables | Check this box to include all variables in your dataset as row variables. |
| Include Missing Values | Check this box to display user-missing values for row variables. |
| Collapse/Expand All | Click expand or collapse to show or hide all values for all row variables. If the number of row variables included is large, the expand option may be grayed out. |
| Hide/Omit Values | Appears after you have deselected values.Hide will include the value in all calculations and statistics, but the value itself will not be displayed. This option is typically selected to reduce the amount of data shown in the crosstab, enabling users to focus on specific results.Omit will exclude the value from all calculations and statistics, and the value itself will not be displayed. This option is typically selected to exclude "don't know" or other variable values to avoid skewing results. |
| Filters Sub-Tab | |
| Filters | Click and drag filter variables to this area to use them, or create filters on the fly by using any variable to analyze a subset of your data.Select the values representing respondents you want to include in the crosstab. The remaining respondents will be filtered out.When using random variables, MarketSight assumes the Boolean operator "AND" between selected values of two or more different variables, and the Boolean operator "OR" between selected values within the same variable. |
| Collapse/Expand All | Click expand or collapse to show or hide all values for all filter variables. |
| Select/Deselect All Values | Click to select or deselect all values for included filter variables. |
| Index Sub-Tab | |
| Use Total column in each table as Index | When no other variable has been added, this option is chosen automatically. The Index on Column % calculation divides the Column % number in a particular column by the Column % number in the Total column, and multiplies by 100. It works similarly for Index on Row %. |
| Select Index variable and value below | Drag and drop an existing variable and check the box next to the value on which you want to base your index. |
| Weighting Sub-Tab | |
| Use weight variable | If you have a weight variable set at the dataset level, it will be automatically displayed here. Otherwise, drag any variable in to use as a weight variable.Deselecting the checkbox or removing the weight variable will run an unweighted crosstab instead. |
Display Tab Options
| Option | Description |
|---|---|
| Combine similar categorical and MRV calculations | Select this checkbox to combine % Sample Size with Column%,and Mentions with Count when both categorical and multiple response variables are present in a crosstab |
| Replicate row variable name on each table row | Replicates the row variable name on each table row. Only available when calculations are displayed in columns. For more information on layout options, click here. |
| Show variable question text/description | Displays the variable description in the crosstab output. This option is not available when "Replicate row variable on each table row" is chosen. |
| Number format | Options for controlling the number of decimals displayed and for suppressing the percentage sign in crosstabs. The up or down arrows will modify the number of decimals to display for percentages, counts, and all other display options.Deselecting the box next to "Show %" enables you to hide the percentage sign otherwise shown next to calculations. |
| Highlight cells with small counts | Specify a threshold at which MarketSight highlights low counts in red (first threshold) or blue (second threshold).Choose whether the thresholds are based on count or unweighted count with the drop downs, and type in the threshold values in the boxes. |
| Highlight cells with small sample sizes | Specify a threshold at which MarketSight highlights cells with small sample sizes in red.Choose whether the thresholds are based on sample size or unweighted sample size using the drop downs, and type in the threshold values in the boxes. |
| Highlight cells with index value above or below a certain threshold | Specify a threshold at which MarketSight highlights index values.Select the boxes to indicate in your crosstab if a cell is below the specified threshold with green shading, or above the specified threshold with red shading. |
Statistics Tab Options
| Option | Description |
|---|---|
| Run tests with confidence level of ... | Check this box to run statistical tests at the desired confidence level. MarketSight automatically runs the appropriate statistical tests based on the column variables, row variables, and display options you have chosen on the Design tab.For more information on which statistical tests are run, please see: Statistics in Crosstabs.For more information on confidence levels, please see: Confidence Levels. |
| Show colors for significant results | Select this option to highlight statistically significant results with colored shading. If this option is deselected, no shading will appear.For more information on how to interpret the shading used in MarketSight, see: Interpreting Significance. |
| Show only significant results | When selected, only row variables that contain statistically significant results are displayed. Only a portion of the included row variables may be displayed. |
| Show significance pop-ups | Displays pop-ups when you mouse over statistically significant cells. Enables significance comments to appear when exporting crosstab results to Excel.The pop-ups display information about the statistical relationship between cells |
| Show test names and p-values | Causes the significance pop-ups to also display the names tests that were run as well as p-values.This option can be enabled for significant results, or for all results regardless of significance. |
| Contrast test (results in yellow shading) | Contrast testing compares each cell of a row variable to the group of all other cells in the same row. For more information on contrast testing in MarketSight, see: Contrast and Pairwise Testing. |
| Pairwise test (results in blue shading) | Pairwise testing compares each cell of a row variable to every other individual cell in the same row. For more information on pairwise testing in MarketSight, see Contrast and Pairwise Testing. |
| Use column letters | Select this option to have a letter identify each column in crosstab results. Lettered columns may be used to identify statistically significant differences that result from pairwise testing. For more information on interpreting the letters in crosstab results, see: Interpreting Significance. |
| Use a second confidence level of ... | Select this option to run pairwise tests using a second confidence level. This option is only available if you use column letters. |
| Treat each pair of columns as a self-contained independent sample | Select this option to run pairwise tests of means (t-tests) that treat each pair of columns as a self-contained independent sample. MarketSight does not run ANOVA in this configuration. Equal variance is assumed between each two samples compared. Variance is estimated only from the two categories compared. |
| Correct for Type I errors in all comparisons(more conservative) | Select this option to correct for the higher chance of Type I error when running multiple statistical tests.By selecting this option, MarketSight automatically adjusts statistical tests to control the overall Type I error rate for a set of related tests. Correcting for Type I error is not necessarily ”better,” however, because it increases Type II error and therefore reduces the power of tests.For more information, see: Correcting for Type I errors. |
| Enforce strict data sufficiency rules(more conservative) | Select this option to enforce strict data sufficiency rules for Chi-squared tests on Row Variables displaying Column % and/or Counts.For more information on strict data sufficiency, see: Strict Data Sufficiency.For more information on Chi-squared tests, see: Statistical Tests in MarketSight. |
| Calculate row % only on the innermost nested column variable | Calculates row % only on the inner-most nested column variable. |
| Sample base for statistical calculations when a weight variable is used | Weighted baseSelect this option to use the weighted base in calculations of test statistics when weighting is in effect.Unweighted base Select this option to use the unweighted base in calculations of test statistics when weighting is in effect. Effective base Select this option to use the effective base in calculations of test statistics when weighting is in effect. Effective base = (sum of weight factors) squared / sum of the squared weight. Using the effective base provides for accurate statistical results for weighted data. |
Interactivity Tab Options
| Option | Description |
|---|---|
| Row Variable Interactivity | If your crosstab contains more than one row variable, you can turn interactivity on or off with this checkbox. If it is selected, a dropdown will appear at the top of the crosstab enabling users to change the variables displayed in the crosstab. |
| Row Variable Label | Edit the displayed text for the interactive dropdown. It is "Select a Row Variable" by default. |
| Column Variable Interactivity | If your crosstab contains more than column row variable, you can turn interactivity on or off with this checkbox. If it is selected, a dropdown will appear at the top of the crosstab enabling users to change the variables displayed in the crosstab. |
| Column Variable Label | Edit the displayed text for the interactive dropdown. It is "Select a Column Variable" by default. |
| Filter Interactivity | If your crosstab contains one or more filter variables, you can turn interactivity on or off with this checkbox. If it is selected, a dropdown will appear at the top of the crosstab enabling users to change the filters used in the crosstab. |
| Filter Label | Edit the displayed text for the interactive dropdown. It is "Filter" by default. |