The interesting thing about statistics in any aspect is that they can be manipulated to make results suitable for one group and unsuitable for another group. There are many things that can be based off of statistics. One should ask: How can one manipulate statistics?
To answer that question, one would need an example to use. One of the best examples would be polls. When polls are launched, a certain number of people are surveyed with a list of questions.
After the surveys are taken and calculated, the results are in.
However, there are plenty of factors at play. Because of these factors, the accuracy of poll statistics can be questioned. These factors are examples of how statics can become manipulated. The 2008 United States Election year is a prime example on how statistics can become manipulated. There were many polls launched on a variety of topics ranging from the state of the United States economy to immigration. The newsgroups such as CNN, MSNBC, Fox News, the New York Times, and a slew of others constantly conduct polls. Even though there are consistent poll results, the results can easily be manipulated.
One has to ask: How does one conduct a poll? Polls are conducted by asking a certain number of people specific questions. But, there are questions that need to be asked:
How many people were interviewed?
Where did the people come from?
In most polls, between 1,000 and 5,000 are interviewed. While that is considered a lot of people, it is only a small fraction of a state's population. In regards to the United States, that is only a small fraction of the country's population.
When one thinks about it, the statistics have been manipulated.
In regards to polls, the statistics came from a small number of people. If one wants a true statistic, the pollsters would have to interview pretty much the entire country. If it is a statewide poll, everybody in the state would have to be interviewed. That could cost time and money. However, it is the only way to get a really accurate statistic. But, the stats could still be manipulated.
When one looks at statistics, this comes from sampling data. However, samples do not really tell the entire story. More samples need to be taken and tested. But, sampling data is not really accurate because of various factors at play. For this reason, the statistics from the polls are not entirely accurate. On top of that, it shows that the statistics are manipulated.
One could take a poll on the current job approval rating for current United States President Barack Obama.
Consistent poll results show that job approval is pretty high in the upper sixties. However, there is the sampling error in place. While the poll results can be consistent, there are questions to be asked.
How many people participated in the survey for the poll?
Who were those people?
Where did those people come from?
What background do these people have?
These various factors can manipulate a poll's statistics. For the first question, it is apparent that a few thousand people were interviewed. There is a major sampling error in regards to 1,000 to 5,000 being interviewed out of 200 million or 300 million people. The US population is in the nine figures.
For the second question, they could have been registered Democrats, Republicans, or independents.
So far, Obama has garnered much support from Democrats and independents.
For the third question, these people could have come from the northeast or the west coast. The northeast and the west coast are major Democratic strongholds while the southeast is considered a major Republican stronghold.
In regards to various topics in polls, the same thing applies. Overall, the statistics from polls came from a very small sample. On top of that, biases do play a role in statistics manipulation. Most if not all polls are biased because of the sampling data. Many people except for the pollsters and survey takers do not really know who or how many were interviewed. Most of us would not know those answers.
There are other factors at play such as social class, economic class, gender, and ethnicity.
These factors played a major role in the US Democratic Primary contest between Barack Obama and Hillary Clinton.
There were polls that gave the statistics of support leaning to either Obama or Clinton. For those that kept track of the primary, one noticed that Obama got support from the younger voters while Clinton
got support from the older voters.
In conducting the poll, one question should be asked: Where were the ages of those that took such a survey?
The subjects of skin color and gender played a huge role in the Democratic Primary.
But, this is statistical manipulation in regards to polls. I had noticed another form of statistical manipulation in Electoral Maps.
The Democratic Primary contest again is one of the best examples.
While Hillary Clinton was struggling to keep afloat, her husband former President Bill Clinton, said she could still win over Republican Senator John McCain. Bill Clinton used Karl Rove's electoral map as a prime example.
However, the statistics can be manipulated in this respect. The results came from Karl Rove's electoral map.
During the general election between Obama and McCain, Obama's campaign manager drew up a completely new electoral map. This map was completely different from Karl Rove's electoral map. The map utilized former Vermont Governor and former DNC chairman Howard Dean's "50-State Strategy."
In this respect, electoral maps can manipulate statistics. This is due to a universal electoral map not being present. Anybody can come up with an electoral map utilizing different structures and approaches.
When you look at methods to get statistics, one can see that they are implemented by people. People are prone to error which is known as "human error." People are prone to error and manipulation. That would mean statistics are prone to error and manipulation as well.
