office: +44 (0) 1491 825644
fax: +44 (0) 1491 832376
QPSMR Companion Data Drill Down
• Drag and drop entries to table columns and rows
• Show figures, percentages, missing, counts, means
• Include filters, weighted data, SIG tests etc
• Discrete SIG tests to remove overlapping records for tests
|QPSMR Companion Data drill down allows you to interrogate a data file, usually by creating cross tables.
Many entries can be used as the table rows and columns, simply by dragging and dropping from the list of entries, or directly from the “Main” window; a table will be produced from each combination.
The tables are shown on screen one at a time with toolbar buttons to move up and down the row entries or between column entries. Toolbar icons allow you to view table figures, column percentages and/or row percentages.
Tables can be filtered as required and if respondent weighting or quantity weighting has been used you can choose to see the raw (unweighted) figures or weighted figures simply by clicking the toolbar icon.
Hovering your mouse over a table cell gives details of that cell including confidence intervals of any percentages shown.
As well as standard rows and columns your tables can show:
• Missing - records that do not have a response in
• the entry (blanks)
• Count - the total number of responses given to
• a multi-coded entry (blanks)
• Unweighted total figures - if data is weighted
• ESS effective sample size total figures - if data
• is weighted
• Mean scores - if the responses are values or
• have score values applied
• Standard deviation - if mean scores are used
|Settings can be applied
globally (so they are used on all tables created), you can:
• choose to output all tables in one single
• spreadsheet, instead of the default separate
• sheet for each table
• choose not to show the "Missing", "Count" and
• "Effective sample size" figures
• suppress blank (empty) rows and columns
• set the widths for row labels and table columns
• set the decimal places to be used for figures,
• percentages, means and standard deviations
• change the significant level percentages used
• suppress the significant difference colouring of
• set the minimum sample needed
There are two types of significance testing used on Data drill down tables:
• Compared to the rest - each column is tested
• against "the rest" (the total column minus the
• column being tested) and significant differences
• shown, coloured appropriately (green higher,
• red lower). For mean scores (averages) and
• standard table rows, as default, a t test is used.
• Column identifier markers - if columns have
• identifiers (a letter in parentheses at the end of
• the label) then each column under a header will
• be tested against all the other columns under
• the same header. There is also an "All columns"
• option to test every column with a marker
• against all other columns with a marker (not
• only the ones under the same header).
• Significant differences are shown by including
• the relevant letters under a cell, meaning that
• the cell is significantly higher than the columns
TIP: When using multi-coded questions as the columns in Data drill down you can apply discrete SIG tests, to remove any overlapping records (respondents that that appear in more than one column under the same header) so the comparison is only between those who are in one column, but not another.
As well as creating tables for analysis, QPSMR Companion also has a Data distinction facility that allows you to compare a group of respondents with the rest of your sample; any significant differences found within the entire questionnaire are displayed.
Using standard filter descriptions simply specify the respondents you wish to test as Sample A (for example, those from Area 1) and the program will automatically test these respondents against Sample B - all respondents who are not in Sample A (in our example those not from Area 1).
|The entire data file is then automatically scanned for any entry where Sample A is significantly different to Sample B, and the results are shown on screen with the significance percentage level in descending order.
The Data distinction facility checks:
• Individual responses to all single-coded and
• multi-coded entries, using a t test
• (although a Z test may be requested)
• The mean scores of all entries that have score
• values attached to responses, using a t test
• The mean values of all value (integer or float)
• entries, using a t test
If respondent weighting or quantity weighting has been applied you can use the raw (unweighted) figures or the weighted figures, and global filters may be applied, if required.