kang:  So in summary... there are between several hundred thousand DGUs and several million 
What you write above is problematic, since such a large difference makes reasonable people wonder why such wide variations over a year's time, in such close proximity to each other in time. How can one year produce 1.6 million  dgus by one pollster, then 3.6 million the next year by another pollster using similar standards of measurement?  Why kang? why?
.. or, as you say, the yearly range of dgus go from ~700,000 up to the 3.6 million. How can there be 5 times more dgus in one year than in a recent previous year? There must be some dichotomy in polls which produce this kind of margin of error, 400%.
In the 12 polls there tends to be an inconstancy of homologous standards of measurement which in good part produces the wide variation in results.  This suggests the surveys are subject to a false equivalency &/or misinterpretation or false positives or telescoping or even manipulation.
I apologize for not including the link to guncite in my last post,  a clumsy oversight on my part, since it was a crucial part of my evidence. It is below, & I've added by edit to my previous post #2. http://www.guncite.com/kleckandgertztable1.html 
Here is a rough copy I made. Note that DMIa DMIb comprise one survey, a & b to separate dgus against animals. And is that 'Mauser' what I'm thinking of? shouldn't they recuse themselves?
Are these the 12 dgu studies which bob Aldridge is referring to, kang? Note that two have 'not available' for their dgu estimates. Three evidently pertain to states of calif, Illinois & ohio. One is by 'mauser'. Three have yearly dgu figures of ~750,000, way off the LATimes 3.6 million.
Tarrance & LATimes both surveyed in 1994, yet LAT had 3.6 million dgus, while Tarrance only 764,000 dgus. LATimes included dgus against animals, but tarrance pertained to evidently respondents in a household, LAT just the respondent, how does that explain it?
FREQUENCY OF DEFENSIVE GUN USE IN PREVIOUS SURVEYS 
Survey: 
 Time/CNN ..Mauser ..Gallup ..Gallup ..L.A. Times ..Tarrance {=6}
Area: 
U.S. U.S. U.S. U.S. U.S. U.S. 
Year of Interview 
1989 1990 1991 1993 1994 1994 
Population Covered 
"Firearm owners"  Residents  NoninstAdult NoninstAd NoninstAd NoninstAd 
Gun Type Covered 
All guns All guns All guns All guns All guns All guns 
Recall Period: 
Ever 5 yrs. Ever Ever Ever 5 yrs. 
Excluded Uses Against Animals 
No Yes No No No Yes 
Excluded Military, Police Uses? 
Yes Yes No Yes Yes Yes 
Defensive question asked of: 
Gun owners ..All Rs.. Rs in hgun hshlds ..Gun owners ..All ..All 
Defensive question refers to: 
Respondent.. Hshld.. Respondent.. Resp.. Resp ..Respondent/Hhold 
% Who Used 
n.a. 3.79 8 11 8[c] 1/2[d] 
 Implied number of def. gun uses 
n.a.. 1,487,342..  777,153..  1,621,377..  3,609,682.. 764,036 
FREQUENCY OF DEFENSIVE GUN USE IN PREVIOUS SURVEYS [80] 
Survey: 
 Field.. Bordua ..CambridgeReports ..DMIa DMIb ..Hart ..Ohio {=6}
Area: 
California.. Illinois ..US.. US US... US .. Ohio 
Year of Interview 
1976 .. 1977.. 1978 ..1978 1978 ..1981.. 1982 
Recall Period: 
Ever/1, 2 yrs. Ever Ever Ever Ever 5 yrs. Ever 
Excluded Uses Against Animals 
No No No No Yes Yes No 
Excluded Military, Police Uses? 
Yes No No Yes Yes Yes No 
Defensive question asked of: 
All Rs.. All Rs ..Protection hgun owners ..All Rs All Rs.. All Rs ..Rs in hgun households
Implied number of def. gun uses
3,052,717 ..1,414,544.. n.a.. 2,141,512 - 1,098,409 ..1,797,461 ..771,043