Saving data as text only works for legacy forms, and frankly doesn't help much when it comes to collating data.
Greg Maxey and myself have spent the past few weeks collaborating on the updating of an add-in we jointly developed to extract data from both legacy form fields and forms with content controls, and also included new functions to aid with the creation of the forms in the first place to ensure that you get meaningful data extraction.
Greg's version (not that linked in Andrew's post - but
http://gregmaxey.com/word_tip_pages/...rom_forms.html) has a different set of emphases from my own published version -
http://www.gmayor.com/ExtractDataFromForms.htm but the core processes are essentially the same. My version exports only to Excel, so can be a little more forgiving of badly designed forms.
Things that we did discover during the update, are that it is much faster to extract data from content controls than from legacy fields (not much of a problem if you only have a handful of fields to process) and Word 2010 processes legacy forms up to six times faster than later versionsm which can be significant if you have more than 100 fields.
I have always liked the convenience of legacy forms, but this exercise has made me seriously doubt the wisdom of them now that content controls are available.