Polls and Statistics - Twins of Evil Data Collecting

It seems as though every five minutes some group releases a new political poll in recent weeks. It has gone so bad that as many as a dozen or more different polls are released every day.

In each poll, one person is sometimes substantially ahead of their opponent and in another poll, the numbers favor the other person.

Of course every poll is skewed toward the group that produces the poll. For example, if a Democrat group takes a poll a Democratic candidate gets the better percentage. This is not by accident. It is done to purposely make people think that their candidate is the better person.

It is the number of people who are polled that is the important part of any poll. This sample size has to be large enough to represent the group it is supposed to represent, but not so large as to alter the results. This margin of error should be small enough to make the poll truthful and not favoring any particular result.

So at this point, you may be asking how does this relate to Staples?

Its simple. Staples used all of the customers to determine the amount of rewards cards that employee takes in during their shift. As the day goes on, the lower the percentage becomes. Early in the day and the percentage of customers with cards is high, at night customers rarely had the rewards cards.

So if Staples only took a random sample throughout the day of rewards cards, the numbers would be much more realistic as far as percentage of customers who actually had the cards. But like a crooked poll, Staples believes in the “poll everything” approach, which unfortunately gives an unfair advantage to those who worked later in the day like myself.

As we get closer to Election Day, everybody should pay close attention to the number of people polled and who does the polling. Chances are that the poll is as crooked as the group performing the polling. Just like Staples does everyday with every employee.