Summary Tables

Mean Summary

A Mean Summary table can show the summary statistics for a series of rating scales on one table. For example, if you had four questions and each asked respondents to rate the magazine Computer Life on one of these four attributes: Interest, Originality, Readability and Coverage of Issues. Additionally, you want to have one table that reports the summary statistics for all four questions.

For the table, you would select Row Options in Setup|Tables for each of the four questions and select desired statistics for each. Following are the row text and the locations for each:

Choose the Mean row option for each of the four rows using Setup|Tables|Row Options.

The values in parentheses indicate the base for each row. If you want a different filter for a row, edit the row logic. You can have complex logic on a row.

Variable data example:

If GENDER (1) is MALE and you want INTEREST Q1A (1-5) to be reported for MALE only, you would write:

GENDER (1) AND Q1A (1-5)

for the logic for the INTEREST row.

If you wrote: Q1A (1-5) and GENDER (1) the statistics would be calculated on the MALE location, since the last location in the logic is the location on which summary statistics are calculated.

ASCII data example:

For example, if location 1/6 (1) is MALE and you want INTEREST 1/25 (1-5) to be reported for MALE only, you would write:

1/6 (1) AND 1/25 (1-5)

for the logic for the INTEREST row.

If you wrote: 1/25 (1-5) and 1/6 (1) the statistics would be calculated on the MALE location, since the last location in the logic is the location on which summary statistics are calculated.

The entire table can be filtered using Filter in Setup |Tables.

You can report the grand mean for the table in the following way:

The grand mean will appear before the summary means. Create row text for the GRAND MEAN using Add Row(s) in Setup|Tables. For the logic on that row, write SUB logic and choose the Mean row option from Row Options in Setup|Tables. When SUB logic refers to rows with means, the grand mean is calculated. It is the sum of values, across all variables, divided by the number of responses across all variables.

For our example, the following rows and their logic:

create a grand mean, which is reported in the first row for the 4 following rows.

Top Box Summary

If you want to create a summary of just the high rating, your rows would be defined as follows:

The first four rows are the "basing" rows. The last four rows are the rows indicating the "high" rating. The "high" rows will need to be percentaged off each corresponding "basing" row using the Percentage Off option of Setup|Tables|Table Options.

Since the first four rows are only used for the calculation, you won't need to show them on your report and can use the Hide row option (HR) of Setup|Tables|Row Options.

Related topics:

Table types

SUB Feature