Colin and Bob, thanks alot for the information. Colin, I will respond to you first.
When I initially discovered the discrepancy I immediately thought of this...because I was selecting the ENTIRE sheet and copying it from a template to a summary file that is place-specific. SO, I have a template to produce a given cross tabulation and the template is just an excel sheet with formatting and formulas in it. I dump in a raw data file (txt) and the template sheet produces a nice summary table that pertains to that place. Each time you add a new raw data set, the template recomputes the table. I then, select the entire sheet including the formatted table in the excel template and then I COPY it, and I PASTE-SPECIAL it into a summary file (excel) that pertains to that specific place. So, yes, I do end up pasting in the ENTIRE sheet. So, I went to each file and made sure that entire sheets were not selected in the saved files....none seemed to be. Wierd thing is, I did this exact procedure for 31 different regional summaries...and more summary files are small...but 3 or 4 blew up to 4 or 5 megs when most are less than 300 k.
Bob, I will respond to your post now...assuming you are not exhausted from my reply to Colin.
Question 1 - I am taking templates that are already made in Excel 03 for Windows, OPENing them into Excel 08 and adjusting them before dropping raw data into them. NOT using the gallery.
Question 2 - when I say "imported" what I mean is PASTED. I have a program called Beyond 2020 running on a PC here (does not come in a mac version) and this program is used by many nations to distribute their census data. The program allows me to produce a given cross tabulation and then SAVE-AS an Excel 03 file. It is the contents of these excel 03 files which I drop into the templates to produce a summary table. No importing per se...sorry for the imprecise language.
Question 3 - regarding repeated importing to one file. Keep in mind I used these templates to conduct this analysis back in 02 for the previous census...but I used a PC...and this did not occur. As for the process, I do have ONE summary file for each region and each summary file contains 40 separate sheets...each sheet is a different tabulation for that region. The end result will be a summary file for a region which includes 40ish cross tabulations in .xlsx format. So, I do PASTE in these 40ish tables into each summary table...but again, I have followed this process for all 30 regions and only a few have ballooned...and I can't see a difference between the small ones and these large ones save for size.
Not sure what the ghosties are. I have started to use SHIFT CONTROL END to highly only the range I have actually used as opposed to going out the edges of the sheets. To distribute the summaries, I have simply used WinZip on my PC and made sure that clients have access to the WinZip extractor...this way, the entire set of 30 summary tables only amounts about 2 megs...when the bloated files themselves are each approaching 5 megs on their own. Whatever the ghosties are, the compression algorithm in WinZip strips them out.
Thanks for your help....you too COlin.
Cheers, Rick