User Guide > Analysis

Companion and Reflect

You can also inspect the data using the Data drill down window.

Tables

This facility builds tables (cross tabulations) from selected entries. Rows, columns, filters and weights (respondent or quantity) are selected by dragging them from a pane on the window or from the main window.

Additional filters can be added by dragging one or more cells from a table to produce tables only for the respondents in those cells.

Tables can use respondent weighting, quantity weighting, and score values.

At any time you can view all the data for any records that you have selected with applied filters.

You can select the table to view and can request to see:

•Raw (unweighted) figures

•Weighted figures

•Row percentages

•Column percentages

Additional rows and columns will be produced in addition to the responses in the entries tabulated:

•Missing - records that do not have a response in this entry

•Count - the total number of responses given to a multi-coded entry (rows only unless horizontal percentages used)

•Unweighted - if the data is weighted the unweighted total figures are also shown

•ESS - if the data is weighted then the Effective Sample Size total figures are also shown

•Mean score - if the responses are values or have scores attached then the average will be shown

•Standard deviation - optional with mean score

You can output individual tables or all the tables to a spreadsheet .xlsx file.

If needed, you can set the name of your spreadsheet program in [Tools] [Options].

Significance testing

Each column is tested against "the rest" and significant differences shown. "The rest" is the total column minus the column being tested. This methodology makes sure that there are no overlapping samples.

Weighting effects are taken into account when doing the testing.

The statistic formulae used can be selected in Tables global.

The t-tests or Z tests on a column are not done if the total in that column or "the rest" is less than format MCT. If the data is weighted then the ESS (effective sample size) is tested against this minimum value. With format NESS the unweighted base is used.

If the column labels have identifiers (letters in parentheses) at the end, then columns are also tested against all other columns under the same header.

Distinction analysis

For this you select two subsets of the data for comparison using filter definitions.

Companion then scans all the entries in the project to find entries where there is a significant difference between these two subsets.