Actually, the BJS report has quite a bit of additional information, for example, burglary is defined broadly in the report and includes entry when no one has the right of being there (trespass) along with the traditional forcible entry and a theft of any item is attempted or succeeds. Under the normal NCVS reports, each crime incident is reported separately and not as related occurrences during the same incident. Thus, a murder that happens during a burglary has both a murder and burglary reported as a crime in its annual report but not as a burglary that resulted in murder or a murder that resulted in a burglary depending on which came first.
This is from the footnotes of the report cited in the O/P
"This report, Victimization During Household Burglary, differs from other NCVS reports in that a different coding approach was constructed to combine burglaries where a household member was present and experienced a violent crime with burglaries that took place while no one was home. Presenting the analyses in this manner allows for a comparison of the characteristics of present and non-present burglaries and the examination of the co-occurrence of a resident’s presence and subsequent victimization.
As a result, estimates presented in this report are not comparable to victimization estimates of burglary or personal crime contained in other NCVS reports. This approach was used previously in Household Burglary, 1985 (NCJ 96021)."
For fun, I also read the earlier 1985 Report which covers 1973-82 data. This is the original source of the 12.7 percent of burglaries occur around or in occupied dwellings. It appears to use a similar format but has much less detail as the computational requirements to churn the survey data were much less available at the time.
Delving into the more recent report itself, one will discover that nighttime burglaries have a much higher rate of occupied house burglaries and this is precisely when a burglar might expect to encounter someone. Only 1/3 of nightime 6PM to 6AM happen when no one is present in the residence where the other 2/3 of the night time burglaries occur in occupied residences. But, this often makes sense when you consider this includes garages, outbuildings such as sheds, and cars on the premises. Quite a bit of the reports come from multi-residences, condos, apartments, etc., which can get problematic when defining them as occupied especially outbuildings, cars, etc.
Examining the details in the report, the burglary as defined in the report includes coming in through open doors and windows, forcible entry, or being admitted to the residence itself. OVER HALF OF THE BURGLARIES when the residence was occupied involved open doors and windows or unlocked doors and windows (see Table 7). Strictly speaking, in the law of most states, this is defined as trespass, not burglary until something is taken. However the BJS report indicates that nothing was taken in about half of their survey incidents when someone was present. This could mean a serious over reporting problem exists where trespass is being equated with burglary.
Reading the report, one can also see that the rate of burglaries differs by income level whether the residence is occupied or not with lower income respondents indicating higher levels of burglary compared with more wealthy respondents. When you separate by whether the residence was occupied, one finds significantly lower rates for occupied burglaries as you ascend on the income scale. Concurrently, one can also observe that the rate of occupied burglaries for homeowners is about half the rate of renters (Table 3).
What the BJS report conflates is home invasions and theft by deception with burglaries as about 30 percent of "burglaries" where the residence was occupied involve entry by shoving past someone or being let in by the resident (Table 7). This by no means is the regular definition of a "burglary". In addition, the definition includes garages, outbuildings, cars, etc. which can account for a substantial amount of burglaries where the residence is occupied.
Now, I believe the purpose of the O/P appears to try to be using the Socratic method to argue that the presence of firearms does not discourage burglars entering homes. This is to create support for gun control legislation. However, the conflation of concealed carry permits which is a license to carry outside of the home is not a reliable proxy of whether a firearm is present in the home. And, I seriously doubt that any city or town of (1,000 or more population) has over 50% issuance of concealed carry permits. However, if you conflate FOID cards required in Illinois which is not a concealed carry permit, I can see some places easily reaching that std. of owning one or more firearms particularly in the southern and western parts of the state.
The more proper use would be ownership by household of any firearm, Gallup fortunately has asked that question and in 1978, they had the rate of firearm ownership in a household rate at 51% which held constant until about early 1990's where it dropped since then to 36% reported in 2016.
https://www.washingtonpost.com/news...erican-gun-ownership-is-now-at-a-30-year-low/
Using that argument, instead of the unreliable proxy of ccw possession, one could argue that the decrease has led to less deterrence of burglars. Anecdotally, illegal residences and felons that are forbidden to own firearms, have both increased as a percentage of the population and are clustered at the lower levels of income/residence ownership/bad neighborhoods/and working in a cash economy. It is very possible that the increase in households that cannot own firearms has led to targeting of such by criminals as easy prey. The prevalence of drug use of harder substances such as meth and opiods that started about that time also play a part. Thus, you would see a secular change in the conditions between 73-82 and the 2003-07 data used for the later report which says little about firearm deterrence of "hot" burglaries aka home invasions. Speaking as someone who has done time-series analysis and had to de-trend the data to reach reliable variable estimates over 30-40 year intervals, it is very difficult to do and requires much more work than the O/P has done.
The O/P's posts also have little to say about Kleck's or Kopel's work as well.
O/P, what you are essentially doing is applying macro level data to individual decisionmaking of burglars whether or not to commit a crime based on the potential presence of a firearm. This is a classic ecological fallacy. Instead, it is easier and far more reliable to use interviews with criminals on target selection information to determine that feature. For example, this study,
https://airef.org/wp-content/uploads/2018/10/BurglarSurveyStudyFinalReport.pdf and this THR thread addresses some of the studies about what individual burglars fear
https://www.thehighroad.org/index.p...st-afraid-of-armed-citizens-need-help.457182/
From Kopel, "To the contrary, well-designed studies of prisoners (J. Wright & Rossi, 1994), of active burglars (R. Wright & Decker, 1994), and of ordinary citizens (Ikeda, et al., 1997) consistently find that active armed defense against burglars by victims is frequent and is feared by most burglars."
http://www.davekopel.com/2A/ch/Comment-on-The-Effects-of-Gun-Prevalence-on-Burglary.htm
Full citation of first study cited by Kopel, Wright, James D., and Peter H. Rossi. “PUBLICATIONS.”
NCJRS Abstract - National Criminal Justice Reference Service, 1994,
www.ncjrs.gov/App/Publications/abstract.aspx?ID=155885.
There are similar studies of muggers, robbers, etc. that indicate these felons try to avoid victims that can fight back including whether or not they might have firearms.