That’s the beauty of it!
This only indicates that you do not understand statistical data collection, it does not mean that for the educated professionals the interim data is worthless or that the reporting is a problem.
It is the opinon of the persons unedcuated in the statistics, of the same nature as the doubters on flouridation or the climate change data.
Indeed yes.
or not wrong, but not right on the full information.
Or neither, rather there is utility to the professionals to have the early data even though the final more solid data takes more time to publish.
Unfortunately like the reporting on the science research, journalists are often - usually not educated to report in a way that is understandable (or maybe it is not really possible) to the general public who do not have any education in the statistics or the particular science and so gross misinterpretations and frustrations arise from fundamental misunderstanding.
Agencies such as the CDC, NASA, NOAA, and others are generally obligated to make raw data available (barring specific data involved in ongoing failure investigations and so forth), and nearly all data that is not of a classified national security nature can be obtained through a Freedom of Information Act (FOIA) request. Errors or manipulations in data interpretation have occurred, particularly by third party or academic researchers, government contractors, and political think-tanks but intentionally “faking” (e.g. fabricating) data on a systemic level would be nearly impossible due to the number of people involved and the amount of oversight non-classified projects receive. Even just concealing information is very difficult; witness the Pentagon Papers or the information relased by Daniel Snowden from NSA archives; it just takes one person to blow the lid off of a conspiracy to conceal or misdirect.
Stranger
Like who, for example?
Of course “they” could get in trouble. Who do you think “they” is? It’s just people in jobs. They can lose their jobs. Some government employees can face criminal punishments for committing fraud in their duties.
If you ask someone to start generating data for you, it’s because you need it. Between flying blind or practicing data-driven decision making, most people would rather not be blind.
Which is really a different thing. The Census is the original ‘statistic’ the US govt was required to publish, and the methodology has been debated and modified for ages. That has nothing IMO to do with the question. IMO ‘fabricating statistics’ means you could publish what everyone would be reasonably sure are the right ones, but you consciously choose to substitute made up ones.
That has happened in other countries. Greece was mentioned, phony Argentine official inflation statistics under the Kirchners are another relatively recent example.
The reasons it generally doesn’t happen in countries with a tradition of relatively strong rule and law and relatively low corruption have been mentioned, but to summarize
- if it were commanded by political leaders, who indeed could have an interest in govt stats, particularly economic stats, saying what they want, the professional people compiling the stats generally have no interest in keeping that command a secret.
- the professional people compiling the stats as a rule have no way to personally benefit otherwise from publishing phony stats.
The likely exception would seem to be stats related to emotional social issues. Climate stats might be an example. I’m not saying any have been fabricated, but it’s an issue which for some people has gone beyond scientific to an emotional social kind of issue (as in ‘we’re going to die in 12 yrs’ to quote a much quoted new face on one side of the US political spectrum). It’s not inconceivable to me that some bureaucrats in relevant US govt agencies could come to feel it their ‘duty’ to falsify climate stats to make them look worse than they are and get the public to ‘finally do something about it’. But still, it’s probably more likely co-workers would see this as counter productive to warning the public about the real threat, and/or simply be too honest to go along with it. Again the current US political leadership might want to fake climate stats in the opposite direction, but there’s essentially zero likelihood the people who compile them would go along or keep the request secret: they are overwhelmingly on the other side politically in their personal lives.
Sorry, but this means you aren’t very good at Conspiracy Bingo. 
It is fervently believed by those suffering from varying degrees of paranoia that mass manipulation/suppressing/altering of data goes on commonly in government, no matter how wildly improbable such plots are.
As for the CDC, antivaxers have been struggling to get as much mileage as possible over the publication of a CDC-sponsored research paper that found no link between MMR vaccination and autism. The paper omitted one finding that all the authors (except one) felt was a statistical outlier that didn’t make sense, and indeed subsequent analysis has borne that out. The scientist who disagreed decided to contact an influential antivaxer to share his concerns and the matter got blown up into the “CDC Whistleblower Conspiracy”.
You would have to believe that substantial numbers of reputable scientists not only would conceal vital facts to protect their reputations/agenda, but that they’re downright Evil to boot.* More than a few people think this way.
*note that all of the data that went into the paper was preserved for interested parties to analyze and interpret/misinterpret.
BTW, here is a great site that pretty much aggregates all US government statistics. www.usafacts.org
It seems like in these cases where the political leadership does not like research, they ban it or refuse to fund it. For example, banning the CDC from studying gun violence, or forbidding agencies from talking to the press. You can debate the merits of these examples, but that is beside the point. The idea is that the Whitehouse doesn’t call up NOAA and say, “hey, you need to knock 2 degrees off those ocean temperature measurements,” instead they defund the research.
Agreed. Flat out lying like the Greeks and Argies did is rare, and in those two cases it can be said those who were being lied to, wanted to believe what they were being told.
Far more common is misrepresentation of facially true statistics. You see this in climate change claims by supporters. Sex crimes cases have been mentioned, there the amount of misrepresentation is mind boggling.
(This does not mean that the causes are false, they deserve support, but for some reason supporters feel the need to do this).
Yes, it is very important to draw a distinction between scientific publications from research agencies, and policy statements published to achieve a political end. As an example, many of the statistics about human trafficking thrown around are completely made up. That is where it is important for the reader to use some judgement. Which is more believable, a 200 page report on the findings of a multi-year study on cancer treatment from the NCI, or a tweet from a Congressional Representative?
That can be very difficult, as it is easy for ones own biases to get in the way of believing stuff. I’m strongly in favor of cat ownership. But then the National Institute on Pets comes out and says that recently convicted felons were 15 times more likely to be cat owners than the general population. It’s much easier for me to claim the NIP is lying, just hates cats, and as always been a dog biased agency, than to accept that this is a striking finding and may point to major problems in the cat owning community. Just because you don’t like some numbers, doesn’t mean they are wrong.
Fabricating data and cherry picking data are two very different things. Fabricating scientific data is hard to get away with and something almost all scientists would naturally avoid.
Cherry picking subsets of data and presenting only those that appear to support your cause, on the other hand, is a universally great temptation that even otherwise good scientists too often give in to.
The answers to scientific questions are not reliably found simply by searching for what pops out of data sets as they become available. To get a reliable answers, one must frame a question, define terms, agree on which data are relevant and valid, choose a method of analysis, and decide what conclusions follow from each of the possible outcomes of the analysis. In the cat example, you would define “cat owner” and “recently convicted felon.” You would decide what population you are interested in defined by time (2019, 2018, an earlier year or a span of years) and space (US, North America, the world). You would have to do a comprehensive and systematic search for all relevant data and have a transparent and unbiased procedure for validating the data. You would need to decide what measure of association between cat ownership and conviction for felony you were going to calculate and then what your conclusion would be after each of the possible outcomes. If you don’t go through all these steps before looking at your data, all you will have is an interesting hypothesis—you will have proved nothing.
What “conspiracy” does the OP want us to help prove? That would narrow things down a lot.
Given the broad range of CTs associated with the CDC (vaccination and gunshot deaths), NASA (global warming, lunar landings, UFOs), and the NOAA (just global warming?) I suspect some friend of the OP is assembling a Grand Unified Conspiracy Theory. I got pretty far along with my own, ten or fifteen years back, but I got distra–Ooh! Shiny!
Well, nothing other than being the center of a giant scandal, losing their job and their professional reputation when their fabrication is discovered.
Besides which, for most of these people, the knowledge is itself the goal, and the government position just a means to that end. It’s like asking why rich people don’t burn money in their fireplace: Sure, they could, but why would they want to?
The problem with the unemployment rate isn’t the number. No doubt, out of the “actively looking” job seekers, only <some number under 5%> haven’t found a job. The problem is that it doesn’t factor in:
underemployed. People who need the income of a full time position (or, since minimum wage pays so little, they really need to have 2 jobs) but only have an intermittent contract job or a part time low paying job.
People who have given up. Now, some of these people are collecting money from various forms of pension and social security, and don’t *need *a job.
But, uh, what do homeless people who desperately would take a job but won’t get one because they can’t get a haircut and a shower count as? Or paroled prisoners getting by on food stamps and petty crime and who no one will hire? What bucket do they fall in?
This is why “workforce participation” is a far more honest metric. If we simply tracked this, or took everyone who is not collecting social security, over the age of 18, and not in full time college or prison, this would be a far more accurate metric.
I thought everyone knew that 86.894% of all statistics were just fabricated.
To the extent gov’t entities decide what types of data they seek from what sources, they could reach results skewed towards specific political agendas. They could choose to rely on a superficial survey, or drill down deeply in order to identify seeming exceptions.
Then, the gov’t bodies are able to decide what data they rely on in enacting various laws/policies.
I’m reminded of a recent conversation I had re: lying to Congress/investigators. There can be a mighty fine line between knowingly expressing a factual falsehood and expressing an opinion.
FYI, here’s an example of fabricated statistics from a government official.