The Fox News Version of Events

[Archer77]

It proves what I"ve been saying all along:  A SUBGROUP DOES NOT HAVE ENOUGH SAMPLE CASES TO MAKE STATISTICALLY RELIABLE CONCLUSIONS!!

What does it mean? How many individuals?   When they refer to incomplete data they aren't referring to perpetrators but victims so the stats on offenders are accurate as are the national victim statistics overall.  What they are saying is they have an excellent understanding of national trends but not local trends.  Who is raping who is accurate but where for example is not certain. Therefore its difficult to get the needed help where it needs to go.  They aren't commenting on the veracity of national statistics.

[Al Doggity]

LOL...

but it does not measure the low incidence events of rape and sexual assault with the precision needed for policy and research purposes. Comparisons across subgroups and years are particularly problematic.

This literally means "The number of sample cases in certain subcategories is too small to be statistically reliable". Exactly what I've been saying.

In a few more posts, it will be  what you've been saying. Since December. 

[Archer77]

LOL...
This literally means "The number of sample cases in certain subcategories is too small to be statistically reliable". Exactly what I've been saying.

In a few more posts, it will be  what you've been saying. Since December. 

Ten sample cases means ten individual respondents?  Whats the actual number of individuals? Does it refer to individuals?  It's subpopulations, meaning non-national or subnational statistics.  They stand behind their national statistics on perps and victims. 

[Al Doggity]

Ten sample cases means ten individual respondents?  Whats the actual number of individuals? Does it refer to individuals?  It's subpopulations, meaning non-national or subnational statistics.  They stand behind their national statistics on perps and victims. 

  "Subpopulations"' means categories: Native Americans, Blacks, Asians, etc. They are literally saying certain categories they don't have enough data and they CAN'T stand behind their statistics.


 

[Archer77]

"Subpopulations"' means categories: Native Americans, Blacks, Asians, etc. They are literally saying certain categories they don't have enough data and they CAN'T stand behind their statistics.


 

The limitations are specifically where to allocate resources and identifying potential victims. It's not specific enough to tell us certain things that are necessary for allocating resources.  The national statistics they consider valid, particularly perpetrator which is what this was all about to begin with.  If anything, this would indicate the number of black rapists and murders and minority victims are underestimated. This would also have very little effect on the statistics pertaining to interracial rape and murder.  What does sample cases mean to you?  Does ten sample cases mean ten individual respondents.


As important as national-level estimates of rape and sexual assault are, there is an equivalent need for quality estimates for certain subpopulations to ascertain which demographic groups are more “at risk” to become victims and to look at regional differences in criminal victimization levels. These more focused estimates are important because they allow for better allocation of resources to prevent crime and support victims

Unfortunately, the sampling error for estimates of victimization rates for many subpopulations of interest can become quite large on the NCVS because there are very few affirmative responses to questions about serious violent criminal victimization in the sampled groups. Thus, BJS does not provide estimates for rape and sexual assault for these subpopulations; they only provide estimates for the larger category, serious violent crimes.

For the aggregated category, serious violent crime, Table 7-2 shows that the CVs at the national level are approximately 6 percent. However, the CVs for important subpopulations are much higher because of their smaller sample sizes. For example, the NCVS estimates that blacks experienced an estimated serious violent victimization rate of 10.8 percent in 2011, which were 65 percent higher than that experienced by whites, 6.5 percent, and the CVs for blacks were high (13 percent). American Indians/Alaska Natives experienced an estimated serious violent victimization rate of 47.3 percent in 2010 and 12.6 percent in 2011, and the CVs for those years were 24 and 51 percent, respectively. It is clear that the sampling errors for these important “at-risk” subpopulation were large and the estimates were very unstable from year to year.

[Al Doggity]

The limitations are specifically where to allocate resources and identifying potential victims. It's not specific enough to tell us certain things that are necessary for allocating resources.  The national statistics they consider valid, particularly perpetrator which is what this was all about to begin with.

That's what it's all about?! I thought it was about series incidents. 


Also, this is crap:
http://www.ncbi.nlm.nih.gov/books/NBK202273/

However, for rape and sexual assault (as measured in 2011), which account for only 13 percent of all serious violent crime victimizations, the CVs for the number of those victimizations are approximately 14 percent at the national level with more year-to-year variation.

In other words, national level rape statistics aren't particularly reliable, either.

[Archer77]

That's what it's all about?! I thought it was about series incidents.  


Also, this is crap:
http://www.ncbi.nlm.nih.gov/books/NBK202273/

However, for rape and sexual assault (as measured in 2011), which account for only 13 percent of all serious violent crime victimizations, the CVs for the number of those victimizations are approximately 14 percent at the national level with more year-to-year variation.

In other words, national level rape statistics aren't particularly reliable, either.

As I said way back in December, unreliable as in they underestimate levels of rape.  In particular minority victims by minority perps. Black on white crime should be fairly accurate. Even if we assume the incidents of rape are higher it only makes the imbalance of black on white sexual assaults even worse or at the very least doesn't change the imbalance at all. You know what this is about.  You denied blacks commit crimes against whites at higher rates.  What does sample cases mean to you?  Does ten sample cases mean ten individual respondents total as you continue to claim?

[Al Doggity]

As I said way back in December, unreliable as in they underestimate levels of rape.  In particular minority victims by minority perps. Black on white crime should be fairly accurate. Even if we assume the incidents of rape are higher it only makes the imbalance of black on white sexual assaults even worse or at the very least doesn't change the imbalance at all. You know what this is about.  You denied blacks commit crimes against whites at higher rates.  What does sample cases mean to you?  Does ten sample cases mean ten individual respondents total as you continue to claim?

LOL... I thought this was about "series incidents"?  And, of course, you're wrong.

I never denied that blacks commit crimes against whites at a higher rate. Statistically, that doesn't
even make sense.  What I did say is that your interpretation of this study was wildly inaccurate and exaggerated. And the type of assumptions you just posted is further proof of that. Not only are they just assumptions based on no data whatsoever, they are actually antithetical to what is posted  in the report.

The report is here:
http://www.bjs.gov/content/pub/pdf/cvus08.pdf

Table 42 has the rape and race statistics, the accurate perp statistics that they stand behind  
I have posted screencaps of the data and the key here.

Based on the methodology that you have posted several times, we know that the sample cases available for black victims is 4: 3 black perpetrators and 1 of undetermined race. 75% black on black, 25% other.The breakdown  the same for white victims: 75% white perpetrators, 25% others. We also know based on the methodology that the number of  sample cases of black perps/white victims is 6, a number so low that the study gives it a standard error rate above 50%.

 Furthermore, as the key points out, these aren't even limited to incidents of rape. The stats also include verbal threats.

I have posted all of this before. You will keep lying though. You will baldfaced lie about understanding what this means. You will baldfaced lie about me saying things I didn't say. You will baldfaced lie about you saying things you didn't say. I posted 10 quotes of me saying the study had 150000 respondents and 10 of you saying the study had only 10, yet you still try to paint it as the opposite. You're just a liar and an idiot.
 

[Archer77]

LOL... I thought this was about "series incidents"?  And, of course, you're wrong.

I never denied that blacks commit crimes against whites at a higher rate. Statistically, that doesn't
even make sense.  What I did say is that your interpretation of this study was wildly inaccurate and exaggerated. And the type of assumptions you just posted is further proof of that. Not only are they just assumptions based on no data whatsoever, they are actually antithetical to what is posted  in the report.

The report is here:
http://www.bjs.gov/content/pub/pdf/cvus08.pdf

Table 42 has the rape and race statistics, the accurate perp statistics that they stand behind  
I have posted screencaps of the data and the key here.

Based on the methodology that you have posted several times, we know that the sample cases available for black victims is 4: 3 black perpetrators and 1 of undetermined race. 75% black on black, 25% other.The breakdown  the same for white victims: 75% white perpetrators, 25% others. We also know based on the methodology that the number of  sample cases of black perps/white victims is 6, a number so low that the study gives it a standard error rate above 50%.

 Furthermore, as the key points out, these aren't even limited to incidents of rape. The stats also include verbal threats.

I have posted all of this before. You will keep lying though. You will baldfaced lie about understanding what this means. You will baldfaced lie about me saying things I didn't say. You will baldfaced lie about you saying things you didn't say. I posted 10 quotes of me saying the study had 150000 respondents and 10 of you saying the study had only 10, yet you still try to paint it as the opposite. You're just a liar and an idiot.
 

Yes you did.  That was why you were upset.  You though the survey was inaccurate and that blacks don't commit more crimes like rape against whites than whites do against blacks.  I don't even understand what there is to dispute here.

You used the term respondents multiple times.  You said the sample size was ten respondents, meaning ten individuals.  You said it in this very thread. There is no other way around that.  You may have acknowledged the 150,000 but you stated it was only ten respondents out of that number.

What does case samples mean to you?  Is sample case the same as sample respondent in your mind? Why did you repeatedly write ten respondents?

[Archer77]

So, in other words  you agree that the survey included 150000 respondents and that survey also limited series incidents to 10 so that the data was not skewed. Good, so we're on the same page with this?

You even tried to agree with me.  Now you are again denying incidents are important.

[Al Doggity]

Yes you did.  That was why you were upset.  You though the survey was inaccurate and that blacks don't commit more crimes like rape against whites than whites do against blacks.  I don't even understand what there is to dispute here.

You said this:
http://www.getbig.com/boards/index.php?topic=546754.msg7673228#msg7673228

What is staggering is that blacks commit assaults against whites twenty-fives more than whites on blacks.  In terms of aggravated assault, the number is two-hundred times more. The National Crime Victimization Survey reported approximately 13,000 black-on-white rapes and 39,000 black-on-white robberies – both violent crimes. The statistics show that the number of white on black rapes and violent robberies were so small that they had to be rounded to the nearest whole number, and that whole number was zero.  This is the reality we live in and not the fantasy land perpetuate by fools where racist white cops killing blacks is an epidemic.

This is not at all what that report says. When another member commented that there should be a huge national uproar over those numbers, I pointed out that those numbers aren't accurate.

Me:
http://www.getbig.com/boards/index.php?topic=546754.msg7673773#msg7673773

Not if you take into account how dubious those numbers are. This study has been discussed in detail for years on the net. The entire thing can be found here:
http://www.bjs.gov/content/pub/pdf/cvus08.pdf

The table relevant to this discussion is number 46 and the methodology of the study is discussed on the final pages.

Firstly, the "rapes" are not rapes. They are statistics of COMBINED reported and unreported rapes, attempted rapes, threats of rapes and sexual assaults that include anything from verbal assaults to fondling.


Then you continue to lie, swerve, evade, accuse me of being every member you can think of, despite me actually posting a link to the actual report. The info is right there.

[Archer77]

Fine. You know what series incidents are. This whole discussion stems from you posting info from this study and claiming that certain black-on-white crime was at epidemic levels while white-on-black crime in those categories was statistically zero. When I pointed out that these stats were derived from fewer than 10 cases, you claimed that didn't make a difference. Right in the link you provided, it says categories with fewer than 10 case studies are unreliable. It literally says that. It's simply common sense. More case studies makes a survey more reliable.


None of this has anything to do with series incident. You keep posting that blurb, but it is completely irrelevant. Limiting the number of series incidents to 10 has nothing to do with the reliability of 10 or fewer case studies.

Why do you use the word respondent interchangeably with samples cases?  What does sample cases mean to you?   Is the same as respondents?  You won't answer the question

[Al Doggity]

Why do you use the word respondent interchangeably with samples cases?  What does sample cases mean to you?   Is the same as respondents?  You won't answer the question

I don't. Some categories have 10 case studies, but the total study- ALL CATEGORIES COMBINED- had 150000 respondents.

Understand?

[Archer77]

You confirm your confusion in the exchange with OMR.  YOu interpret samples cases as individual respondents.

Depends, is that sample group composed of Sheldon Adelson and 9 other of America's richest people?

Another way of looking at it is by asking if the 538 presidential electors that make up the electoral college are a reliable-enough group to chose our president every 4 years (Keeping in mind that there are 316 Million Americans)?

No is the straightforward answer to your question. A sample of 10 people can not give you a reliable national statistic.

"1"

Exactly. But more importantly... that's exactly what quoted text from the study is saying, too.

[Archer77]

They do limit series incidents to 10 to prevent skewing national data.  I may be wrong that this applies to the 10 sample cases references but you are absolutely incorrect when you claim the number 10 means 10 individual respondents.  You used the word respondents many times.

[Al Doggity]

You confirm your confusion in the exchange with OMR.  YOu interpret samples cases as individual respondents.

OMR:
A sample of 10 people can not give you a reliable national statistic.

Me:
Exactly. But more importantly... that's exactly what quoted text from the study is saying, too.

[Al Doggity]

They do limit series incidents to 10 to prevent skewing national data.  I may be wrong that this applies to the 10 sample cases references but you are absolutely incorrect when you claim the number 10 means 10 individual respondents.  You used the word respondents many times.

There are over 150000 respondents. The only time I referred to 10 respondents was to correct you and inform you of the actual number.

[Archer77]

What does sample cases mean to you?   Why did you repeatedly use the 10 respondents?  They are not the same.   Nice edit.  OMR gives a direct answer to your question and you confirm he is correct.   OMR refers to 10 people and you confirm that is exactly what you meant.

OMR:
A sample of 10 people can not give you a reliable national statistic.

Me:
Exactly. But more importantly... that's exactly what quoted text from the study is saying, too.

Depends, is that sample group composed of Sheldon Adelson and 9 other of America's richest people?

Another way of looking at it is by asking if the 538 presidential electors that make up the electoral college are a reliable-enough group to chose our president every 4 years (Keeping in mind that there are 316 Million Americans)?

No is the straightforward answer to your question. A sample of 10 people can not give you a reliable national statistic.

"1"

Exactly. But more importantly... that's exactly what quoted text from the study is saying, too.

[Al Doggity]

I may be wrong that this applies to the 10 sample cases references

You are wrong about everything. Your interpretation of the study is wrong. You lied about what I said. you lied about what you said. Your understanding of how series incidents relates to sample cases is wrong. Your conclusions on rape based on the data from the study were wrong.

[Archer77]

You are wrong about everything. Your interpretation of the study is wrong. You lied about what I said. you lied about what you said. Your understanding of how series incidents relates to sample cases is wrong. Your conclusions on rape based on the data from the study were wrong.

hat does sample cases mean to you?   Why did you repeatedly use the 10 respondents?  They are not the same.   Nice edit.  OMR gives a direct answer to your question and you confirm he is correct.   OMR refers to 10 people and you confirm that is exactly what you meant.

OMR:
A sample of 10 people can not give you a reliable national statistic.

Me:
Exactly. But more importantly... that's exactly what quoted text from the study is saying, too.

Depends, is that sample group composed of Sheldon Adelson and 9 other of America's richest people?

Another way of looking at it is by asking if the 538 presidential electors that make up the electoral college are a reliable-enough group to chose our president every 4 years (Keeping in mind that there are 316 Million Americans)?

No is the straightforward answer to your question. A sample of 10 people can not give you a reliable national statistic.

"1"

Exactly. But more importantly... that's exactly what quoted text from the study is saying, too.

And you're a liar kore.  A fat zit faced liar.  The survey text never says ten people or respondents.  So this confirms you confused sample cases with respondents.

[Archer77]

You are wrong about everything. Your interpretation of the study is wrong. You lied about what I said. you lied about what you said. Your understanding of how series incidents relates to sample cases is wrong. Your conclusions on rape based on the data from the study were wrong.

The conclusion on rape are the surveys not my own.  

[Al Doggity]

What does sample cases mean to you?   Why did you repeatedly use the 10 respondents?  They are not the same.   Nice edit.  OMR gives a direct answer to your question and you confirm he is correct.   OMR refers to 10 people and you confirm that is exactly what you meant.

There was no edit. OMR and Hulkotron both quote me. The post has been the same since December.


Sample cases are the cases used in the study. There were 150000 respondents and some categories- once broken down by subgroups- had less than 10 sample cases.

[Archer77]

OMR says ten people and you confirm that is what you meant.  You can't get around that.  The survey never says ten people or respondents.

There was no edit. OMR and Hulkotron both quote me. The post has been the same since December.


Sample cases are the cases used in the study. There were 150000 respondents and some categories- once broken down by subgroups- had less than 10 sample cases.

This is not an answer.  How many individuals in a case sample?  Is case sample the same as respondents?  Ten sample cases as in ten people only? You seem to think so.

[Al Doggity]

And you're a liar kore.  A fat zit faced liar.  The survey text never says ten people or respondents.  So this confirms you confused sample cases with respondents.

This is right from the study.

You can find the whole thing here:
http://www.bjs.gov/content/pub/pdf/cvus08.pdf

[Al Doggity]

It says "Estimate is based on 10 or fewer sample cases."