<< Click to Display Table of Contents >> Navigation:  Reference Manual > Questions > Float question

A float question is one for which the answer is not always a whole number - it may contain a decimal point.  In the QPSMR Classic programs this is referred to as a Weight question.

You can fix the number of decimal places which are used when displaying, saving or exporting data.

You will need to set the width to the number of characters necessary to hold the largest answer you expect.  You must always allow room for the decimal places (at least one if not set), for example the number 100 always needs a width of 5 because it will be stored as "100.0".  If set to 2 decimal places then 100 will need a width of 6 because it will be stored as "100.00".  If negative values are allowed, you should include an extra location to hold the minus sign.

A float question may occupy a width of up to 20 data locations.

It is also possible to set a range of valid answers and up to 3 further numbers which have special meanings, by using the "plus" boxes, for example 998 (Refused) and 999 (Don't know).  The range limits and extra values must be whole numbers.

If the decimal places has not been set, it will not include trailing zero decimal places which means the decimal point will not always be in the same place (floating).  Some decimal places may also be removed if there is not enough space for the digits before the decimal point.

If decimal places has been set then this will always be the number of decimal places used when storing the data.

The program will only allow valid answers to be entered for a float question.

A data check report will error questions with a value out of range.

TIP: float questions that are filtered past or have been skipped will be treated as "undefined" in arithmetic calculations.