In the last 8 years has their been a change in the way unemployment is calculated? Is there a nuance that has changed in the polling info? Many of my right wing comrades claim the Obama administration has changed the calculation in some way to make it seem as though the number is lower than what it should be?
The most likely explanation for discrepancies that political partisans like to pick at is that presidents like to cite different measures of unemployment. The BLS has 6 alternative categories of labor under-utilization. The U-6 measure of unemployment with seasonal adjustment stands at 9.7%. Although the official rate may be under 5% as in the BLS's September 2016 release here, many people may not feel like the economy is much better, and partisans can argue over which measure of employment or labor force participation rate is most useful.
Most politicians are not part of insidious, cynical political schemes, as the media may make them out to be. Just that when incidents happen, we hear about it non-stop, and it reinforces the public's view that political workers are untrustworthy as a whole, allowing for weird specious claims from the left and right. Take your friends' words with a grain of salt, and when you hear any president make a claim about the unemployment rate, you can look up the BLS's numbers yourselves and see how they are measured and which one the president is citing.
It is also worth noting that the BLS regularly updates how they measure unemployment.
Okay here is an answer. Nothing in this is simple. But this is how you lie with statistics.
The U-3 number is the one posted in the news all the time. It is a simple percentage of people unemployed when taken from the total number of people "participating in the workforce"
That's where the confusion, deliberate or otherwise, sets in. Change the number of people "participating in the workforce" and the U-3 percentage may not change, yet the actual number of people who are not holding down a job and contributing to the economy can go through the roof.
For instance: current headlines. Unemployment rate lowest since 2000. This is a phony comparison. In 2000 the percentage of working age Americans "participating in the workforce" was nearly 67%. TODAY that percentage is only 62%.
In terms of total numbers let's look at a guestimate of numbers. Let's say that 1/3 of America is of the age to work, i.e. between 16 and 60. That would be approximately 87 million. So in 2000 with a 67% participation rate you had a workforce of approx. 58 million. Thus a 4.1 percent unemployment meant you had a total of 23 million people officially unemployed.
That is interesting. Now in 2016 the participation rate was only 62%. That means your number of people who are still in the work force is only 53 million. So you do the calculation on that and you get a total of 22 million unemployed. And, everybody is all happy and stuff.
The problem is that in addition to that you now have to figure out why you have FIVE MILLION fewer people in your workforce. Now to make matters worse. In 2010, to hide this decline in workforce participation the Obama admin made a significant change in methodology.
In 2000 you were dropped out of work force participation if you had not found a job in 99 weeks. In 2010, 2 yrs into the Obama admin that was increased to 260 weeks. This skewed the workforce participation rate so badly that comparing unemployment numbers in 2017 to those of 2000 and going, "YIPPEEEEE what great news!", should be prefaced with, "Once upon a time", to make sure you get credit for being a great story teller instead of being mistaken for a bare faced liar.
This fairy tale skews the number of people "no longer participating in the workforce" by totals up in the millions. In a straight line comparison we only had to figure out where the extra FIVE million jobs went. But with this change in 2010 the actual number of people not participating in the work force goes up by even MORE millions, if calculated using pre-2010 methodolgy.
That change in methodology in 2010 means that ANY comparison of unemployment between unemployment rates in 2017 with any year prior to 2010 is just a meaningless exercise in story telling. Furthermore with a workforce participation rate of only 62% the U-3 number is just a farce.
Let me finish, that my grasp of all of this is pretty weak. If someone can offer better or more complete interpretation of all of this PLEASE feel free to chime in.
In answer to the original question - not there has not been any change in the way unemployment is calculated. You can see how it is done here https://www.bls.gov/cps/cps_htgm.htm The way data is collected will change from time to time with technology but the way it is calculated has not. The last significant change in the BLS employment situation summary was in 1994 and even then it did not impact the frontline unemployment rate (U3).
Now in response to some misinformation that has come from other answers.
"It is also worth noting that the BLS regularly updates how they measure unemployment" - no they do not. The last change in any measurement was in 1994 - and even then that did not in any way impact the frontline U3 rate that is most often quoted (including by the BLS) as the unemployment rate.
"In 2000 you were dropped out of work force participation if you had not found a job in 99 weeks. In 2010, 2 yrs into the Obama admin that was increased to 260 weeks" This is pure fiction. In 2000, 2010 and now there was and is no limit to how long someone can remain in the workforce. If one is either working or not working but looking and available for work, they are a member of the workforce. How long they have been in that situation is simply not a factor.
"In terms of total numbers let's look at a guestimate of numbers. Let's say that 1/3 of America is of the age to work, i.e. between 16 and 60. That would be approximately 87 million. So in 2000 with a 67% participation rate you had a workforce of approx. 58 million. Thus a 4.1 percent unemployment meant you had a total of 23 million people officially unemployed." None of these numbers is even close to correct. Why are you making "guessitmates" that are completely wrong when the real data is readily available? First off - the upper age limit of 60 is something you have just invented. The BLS uses the adult civilian non-institutional population - which does not exclude everyone over 60. Second - in Dec 2000 the workforce was 141m, not 58m and the number unemployed was 5.8m, not 23m. https://www.bls.gov/news.release/history/empsit_01052001.txt