Ebola has come to America - CDC confirms first case.

[Al Doggity]

Ok, now let's deal with the part where you seem to be having the most confusion. What they are talking ab out in the blurb that you don't understand is "series victimization". They even define it for you:

high-frequency repeat victimizations (or series victimizations). Series victimizations are similar in type but occur with such frequency that a victim is unable to recall each individual event or describe each event in detail.

Ok. So, now we know that "series victims"= people who are repeated victims of similar crimes. Keeping that in mind, the following passage that you didn't seem to understand should now make sense to you:

The weight counts series incidents as the actual number of incidents reported by the victim, up to a maximum of 10 incidents.

So, what that means is that the report counts up to 10 incidents in a series, even if there are more.

Let me know what about this you don't understand? I can be even clearer if necessary.

[Archer77]

Ok, now let's deal with the part where you seem to be having the most confusion. What they are talking ab out in the blurb that you don't understand is "series victimization". They even define it for you:

Ok. So, that is the paragraph right before the one you keep posting and interpreting incorrectly. It is about series victim= people who are repeated victims of similar crimes. Keeping that in mind, the following passage that you didn't seem to understand should now make sense to you:

So what that means is that the report counts up to 10 incidents in a series, even if there are more.

Let me kn ow what about this you don't understand? I can be even clearer if necessary.

They are just acknowledging the sample size is small, that's all.  I actually acknowledged this in another post. The sample sizes could be bigger.   Despite the small sample size the RCVS is pretty certain they are right.What they aren't saying is that here is a 50% change they are wrong.    Using a small sample size actually is necessary to balance the statistics and prevent them being to high or to low in either direction. If they used straight incident reports the numbers would actually be higher when outliers and repeated incidents are calculated in the totals thus causing an imbalance in the data.


Are you saying this information is completely useless?

[Al Doggity]

 Despite the small sample size the RCVS is pretty certain they are right.What they aren't saying is that here is a 50% change they are wrong.

They literally are saying that.  They literally say that because the sample size is so small,  the standard error of  the estimates is extremely high. LIterally.

Using a small sample size actually is necessary to balance the statistics and prevent them being to high or to low in either direction.

No, it isn't because they use much larger sample sizes throughout the rest of the report.
Right in the link you provided it shows that the survey included over 150,000 people.

Are you saying this information is completely useless?

When it comes to a sample size of less than 10, even they are saying that!

[Archer77]

They literally are saying that.  They literally say that because the sample size is so small,  the standard of  the estimates is extremely high. LIterally.

No, it isn't because they use much larger sample sizes throughout the rest of the report.
Right in the link you provided it shows that the survey included over 150,000 people.
When it comes to a sample size of less than 10, even they are saying that!

They don't use a larger sample throughout the rest of the report.    The report questioned a total of more than 10(150,000 or more) but uses ten as the sample size to correct statistical imbalances. Get it straight, the 10 are a random sampling of the larger pool of individuals surveyed.  For example, they take a survey of 150,000 people who reported sexual abuse.  Of that 150,000 10 are taken from the group and percentages are calculated from that 10.

Being a random sampling of a much larger number, the Coefficient of Variation will naturally be higher.  The variation may change depending on which 10 individuals you chose out of, say 150,000. They admit this but also say that they are still confident the numbers are right because they employ equations to account for that.

[Al Doggity]

They don't use a larger sample throughout the rest of the report.    The report questioned a total of more than 10(150,000 or more).
Of that 150,000 10 are taken from the group and percentages are calculated from that 10.

They admit this but also say that they are still confident the numbers are right because they employ equations to account for that.

LOL...This doesn't even make sense. Why would they interview 150,000 people to get a sample size of 10??

 Methodology says nothing about using a sample of 10. Just that certain categories contain fewer than 10 and aren't reliable.

And, no the BJS is not confident their numbers are right. Also from your link:

To restore the quality of the NCVS data, BJS and the U.S. Census Bureau implemented two large-scale interventions. First, the number of sample cases was increased to improve the stability and precision of national and subgroup estimates of victimization.  

So, to improve the accuracy, they increased the number of sample cases. There is a description about how they are constantly tweaking and trying to improve what they admit is imperfect methodology and data.

[Al Doggity]

They don't use a larger sample throughout the rest of the report.    The report questioned a total of more than 10(150,000 or more) but uses ten as the sample size to correct statistical imbalances.

Please post the exact sentence that makes you think the sample size is corrected for statistical imbalances down to 10. Not a wall of text, but the sentence (or maybe 2) that makes you think that.
I can see how an entire paragraph might go over your head, but if you break it down none of what you are claiming is in that report. I'm seriously wondering how you can misunderstand this so much.

BTW, the the whole report is here. You can
clearly see that the majority of categories include sample sizes in the hundreds or thousands.

[Pray_4_War]

thats because it directly conflicts with college classroom theory.

Real world issues fly in the face of classroom idealism. Its been shown over and over again throughout history.

Try convincing a liberal of that. 

They'll tell you, "our ideas work, we just had the wrong people implementing them".  "Pay no attention to the countless real world examples over the many decades".  "Ignore the dictators, the rampant poverty and the millions of dead bodies, we care about people".

[Archer77]

LOL...This doesn't even make sense. Why would they interview 150,000 people to get a sample size of 10??

 Methodology says nothing about using a sample of 10. Just that certain categories contain fewer than 10 and aren't reliable.

And, no the BJS is not confident their numbers are right. Also from your link:

So, to improve the accuracy, they increased the number of sample cases. There is a description about how they are constantly tweaking and trying to improve what they admit is imperfect methodology and data.

Ive already posted the reason a sample size of ten is used from information about their methodology directly from their website several times. They use a sample size of ten to fix balance issues. Its several postsup.  They provide an entire pdf abour their methodology you can dowload. TheFBI also discusses on thier website about the differences between methodology they employ versus the rcvs.  

They conduct a large number of interviews in intervals of time period of time like three months to six months to a year.  To avoid inflating the numbers they take ten from the pool and calculate the percentages that way.  They warn readers about that not because they feel their conclusion are wrong but to be transparent.  A 50% correlative coefficient doeent mean there is a 50% chance the stats are wrong.

The reason some dont use the 10 sample size has to do with the nature of the statistic. In those cases there isnt a need to correct balancing issues.  For example car theft versus sex abuse.  The chance the same car will be stolen repeatedly isnt as likeky as one person experiencing multiple instances of sexual abuse. The methodology they use is intended to prevent multiple instances of abuse on a single individual being calculated as seperate events thus screwing up the data.


The weight counts series incidents as the actual number of incidents reported by the victim, up to a maximum of 10 incidents. Including series victimizations in national rates results in rather large increases in the level of violent victimization; however, trends in violence are generally similar regardless of whether series victimizations are included.

In 2012, series incidents accounted for about 1% of all victimizations and 4% of all violent victimizations. Weighting series incidents as the number of incidents up to a maximum of 10 incidents produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect of extreme outliers on the rates. Additional information on the series enumeration is detailed in the report Methods for Counting High Frequency Repeat Victimizations in the National Crime Victimization Survey, NCJ 237308, BJS web, April 2012.

[Primemuscle]

There is an excellent article in this month's issue of Vanity Fair regarding ground zero in for the current Ebola outbreak. I highly recommend anyone interested in how this disease has spread so rapidly, read it.

[Al Doggity]

The reason some dont use the 10 sample size has to do with the nature of the statistic. In those cases there isnt a need to correct balancing issues.  For example car theft versus sex abuse.  The chance the same car will be stolen repeatedly isnt as likeky as one person experiencing multiple instances of sexual abuse.

So, this is a screenshot from the study. The entire thing is here. You will notice that beneath the table is an asterisk that notes "estimate is based on 10 or fewer sample cases". You will note that that asterisk is only used six times in the table. None near sexual assault crimes. Obviously what you posted is nonsense.

 The study is based on more than 150,000 respondents, not 10.

[Al Doggity]

They conduct a large number of interviews in intervals of time period of time like three months to six months to a year.  To avoid inflating the numbers they take ten from the pool and calculate the percentages that way.  

Nowhere on the site do they say anything like this, nor does it make sense. No study would ever conduct interviews with 150,000 people, whittle those interviews down to 10 -to get more "accurate" results.   Right in the link you posted it says they've INCREASED the sample size over the years to get more accurate results.

Here is another table from the study. We see that the asterisk is used more more than on the earlier chart. However, the asterisk is never next to a number in the "total" column. Obviously, the total number of cases is more than 10. The asterisk shows up when the category is subdivided. So, the overall sample size is more than 10, but ~when the categories are broken down, the numbers in some of the samples drop below 10.
Once again, it is clear that what you are saying about limiting the pool to 10 is wrong.

[Al Doggity]

 They warn readers about that not because they feel their conclusion are wrong but to be transparent.  A 50% correlative coefficient doeent mean there is a 50% chance the stats are wrong.

The term is "coefficient of variation" and that's exactly what it means. They define the CV as the measure of standard error in a section titled "Computing Standard Error"!

You've said this twice now, so if a high CV doesn't mean a higher probability of error, what does it mean? Seriously, if it's not to calculate error, what does that number mean?

[Shockwave]

Alright, enough, Al and Archer, please stop your buckering about interpreting statistics. Its a no win scenario and its dominated the last 3 pages, that may as well rename this 'Al and Archers statistics pissing match'.

[Al Doggity]

Alright, enough, Al and Archer, please stop your buckering about interpreting statistics. Its a no win scenario and its dominated the last 3 pages, that may as well rename this 'Al and Archers statistics pissing match'.

Newsflash: Not only can you ignore this thread if you don't like the content, but you can open any other thread you like. Interestingly enough almost all of them contain bickering.  

And is it really a no-win situation? Do you honestly believe that there's any survey that interviews 150,000 people  then only counts 10 of those people in the name of accuracy? Do you honestly believe that a survey of 10 people provides an accurate assessment of ANYTHING nationally?

[Archer77]

Newsflash: Not only can you ignore this thread if you don't like the content, but you can open any other thread you like. Interestingly enough almost all of them contain bickering.  

And is it really a no-win situation? Do you honestly believe that there's any survey that interviews 150,000 people  then only counts 10 of those people in the name of accuracy? Do you honestly believe that a survey of 10 people provides an accurate assessment of ANYTHING nationally?

You're absolutely wrong.  Limiting incidents sample to ten (incidents per person) is done to avoid skewing the results.


In 2012, series incidents accounted for about 1% of all victimizations and 4% of all violent victimizations. Weighting series incidents as the number of incidents up to a maximum of 10 incidents produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect of extreme outliers on the rates.

It's that simple.  They aren't admitting to being fifty percent wrong.  What they are saying is the data isn't intended to be interpreted outside of the conclusions made by the RCVS.  Something like frequency of incidents can't be defined accurately by t
he statistics they've calculated.  It's a warning to people not to use the data to make conclusions other than those made by RCVS.
Let me rephrase it, deeba dooba deeba.   Zippity doo daa.

In probability theory and statistics, the coefficient of variation (CV) is a normalized measure of dispersion of a probability distribution or frequency distribution. It is defined as the ratio of the standard deviation to the mean .
The coefficient of variation is useful because the standard deviation of data must always be understood in the context of the mean of the data. In contrast, the actual value of the CV is independent of the unit in which the measurement has been taken, so it is a dimensionless number. For comparison between data sets with different units or widely different means, one should use the coefficient of variation instead of the standard deviation.

[Shockwave]

Newsflash: Not only can you ignore this thread if you don't like the content, but you can open any other thread you like. Interestingly enough almost all of them contain bickering.  

And is it really a no-win situation? Do you honestly believe that there's any survey that interviews 150,000 people  then only counts 10 of those people in the name of accuracy? Do you honestly believe that a survey of 10 people provides an accurate assessment of ANYTHING nationally?

guess what fuckhead, i liked the content of this thread before you decided to fuck it up turning it into your personal pissing match. Thats shits fine on the G&O.

Its no win because youre both interpreting the data in different ways and you both absolutley believe youre interpreting it in the correct way; therefore this will never end. If youre too fucking dense to see that theres no hope for you.

[Al Doggity]

In 2012, series incidents accounted for about 1% of all victimizations and 4% of all violent victimizations. Weighting series incidents as the number of incidents up to a maximum of 10 incidents produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect of extreme outliers on the rates.

So , you keep reposting this bit. What does "series incident" mean?

[Al Doggity]

guess what fuckhead, i liked the content of this thread before you decided to fuck it up turning it into your personal pissing match. Thats shits fine on the G&O.

It's fine on any board. It happens here all the time. If you don't like the content, ignore the thread.

Its no win because youre both interpreting the data in different ways and you both absolutley believe youre interpreting it in the correct way; therefore this will never end. If youre too fucking dense to see that theres no hope for you.

Archer doesn't believe he's interpreting the data correctly. No one on earth believes that you would interview 150,000 people to get a sample size of 10. He doesn't and neither do you.  The concept doesn't even make sense and the blurb he keeps posting says nothing about limiting the sample size. The actual screenshots of the study show that the sample size is not limited.

[Al Doggity]

In 2012, series incidents accounted for about 1% of all victimizations and 4% of all violent victimizations. Weighting series incidents as the number of incidents up to a maximum of 10 incidents produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect of extreme outliers on the rates.

Seriously, Archer, you keep posting this quote and it clearly says  it's in reference to ""series incidents". I posted earlier what "series incidents" means. What do you think it means?

Why do the screenshots of the study show that the respondents exceed 10?

[Archer77]

So , you keep reposting this bit. What does "series incident" mean?

I already posted that.  This is all you need to know to understand what the ten sampling size is and why it is used.   Whats so hard about understanding this?  I didn't make it up.  This is from their website. I think you're playing dumb on purpose to derail the thread.  It's not ten people you moron.

In 2012, series incidents accounted for about 1% of all victimizations and 4% of all violent victimizations. Weighting series incidents as the number of incidents up to a maximum of 10 incidents produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect of extreme outliers on the rates.

Victimization weights used in analysis of NCVS data account for the number of persons present during an incident and for high-frequency repeat victimizations (or series victimizations). Series victimizations are similar in type but occur with such frequency that a victim is unable to recall each individual event or describe each event in detail. Survey procedures allow NCVS interviewers to identify and classify these similar victimizations as series victimizations and to collect detailed information on only the most recent incident in the series.

The weight counts series incidents as the actual number of incidents reported by the victim, up to a maximum of 10 incidents. Including series victimizations in national rates results in rather large increases in the level of violent victimization; however, trends in violence are generally similar regardless of whether series victimizations are included.

The warning is to tell the reader that the series incidents might be greater than 10.

[Al Doggity]

In 2012, series incidents accounted for about 1% of all victimizations and 4% of all violent victimizations.

The weight counts series incidents as the actual number of incidents reported by the victim, up to a maximum of 10 incidents. Including series victimizations in national rates results in rather large increases in the level of violent victimization; however, trends in violence are generally similar regardless of whether series victimizations are included.

The warning is to tell the reader that the series incidents might be greater than 10.

  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.

[Soul Crusher]

Posted on October 19, 2014 at 1:53:02 PM EDT by Nachum

President Barack Obama convened a surprise meeting Saturday night on Ebola. The White House meeting of top officials and advisors took place after Obama played golf in the afternoon at Ft. Belvoir, Virginia with ESPN host Tony Kornheiser.

The urgent Ebola meeting came as Obama chided Americans in his weekly Saturday address for being concerned over three cases of Ebola when thousands are killed by the flu every year.

In a ‘who are you going to believe, me or your own lying eyes’ moment, Obama told the nation there is not an Ebola outbreak in Dallas, but that there may be more “isolated cases.”

“First, what we’re seeing now is not an “outbreak” or an “epidemic” of Ebola in America. We’re a nation of more than 300 million people. To date, we’ve seen three cases of Ebola diagnosed here-the man who contracted the disease in Liberia, came here and sadly died; the two courageous nurses who were infected while they were treating him. Our thoughts and our prayers are with them, and we’re doing everything we can to give them the best care possible. Now, even one infection is too many. At the same time, we have to keep this in perspective. As our public health experts point out, every year thousands of Americans die from the flu.”

(Excerpt) Read more at thegatewaypundit.com ...

[Al Doggity]

They don't use a larger sample throughout the rest of the report.    The report questioned a total of more than 10(150,000 or more) but uses ten as the sample size to correct statistical imbalances. Get it straight, the 10 are a random sampling of the larger pool of individuals surveyed.  For example, they take a survey of 150,000 people who reported sexual abuse.  Of that 150,000 10 are taken from the group and percentages are calculated from that 10.

Why do you keep changing your points? This clearly says that you think 10 people out of 150000 are used as stats. This is not only wrong, but has nothing to do with series incidents. You don't know how to interpret data and you just started pulling random shit out of a hat.

[Archer77]

Why do you keep changing your points? This clearly says that you think 10 people out of 150000 are used as stats. This is not only wrong, but has nothing to do with series incidents. You don't know how to interpret data and you just started pulling random shit out of a hat.

Facepalm

What you call random is right from their website and the pdf on their methodology right from their website.   

[Al Doggity]

Facepalm

What you call random is right from their website and the pdf on their methodology right from their website.   

The website says nothing about paring down the 150,000 respondents to 10. The blurb that you keep posting is specifically for series incidents. Right in the blurb it says that they account for 4% or all incidents. The 150,000 is a sample. They don't turn it into a smaller sample.