Subscribe to Blog via Email
Good Stats Bad Stats
Search Text
April 2025 S M T W T F S 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 -
Recent Posts
Good Stats Bad Stats
goodstatsbadstats.com
“The private sector is doing just fine.”
With the words “The private sector is doing just fine.” Obama was pounced upon by the Republicans as not being in touch with the economy. What seems to have been left out of the discussion in the ensuing discussion was the full context of Obama’s remarks. For the record he said:
The private sector is doing fine. Where we’re seeing weaknesses in our economy have to do with state and local government. Oftentimes cuts initiated by, you know, Governors or mayors who are not getting the kind of help that they have in the past from the federal government and who don’t have the same kind of flexibility as the federal government in dealing with fewer revenues coming in.
So what is the story as told by the data? Reviewing the employment numbers from the Bureau of Labor Statistics a number of points are clear. When Obama took office private sector employment was at 113.8 million while last month BLS put the number at 111.4 million. So Obama is wrong the private sector is not doing just fine. Look deeper. At the depth of the recession when private sector employment reached it lowest point in December of 2009 employment at 105.6 million workers. In the last 18 months employment in the private sector has grown by 5.8 million. That is a respectable number but it has not been enough to bring the economy out of trouble. Unemployment remains unacceptable high at an official number of 12.7 million.
Let’s look at the government side of the equation. When Obama took office there were 21.0 million government workers. Employment in that sector bottomed out in April of this year with the loss of 1.1 million workers for a low of 19.9 million workers. May saw the numbers recover to 20.1 million. In this regard government sector employment from Obama’s perspective has been a real problem. The Republicans on the other hand are likely rejoicing in those numbers as they think that government is too large.
While Obama must understand is that neither sector, private or government are doing “fine,” the Republicans need to realize that there are two sectors to the economy and when the government sector loses a million workers they are a million people who are without an income, these are a million people that the Republicans must realize that in their paradigm the private sector much create jobs for in the mist of an already difficult time. This is a case where it seems that both sides are out of touch with reality.
Posted in Telling the Full Story, Where's the Data?
Romney and Job Creation
Much has been said about Romney’s record on creating jobs both as governor of the state of Massachusetts and while he was at Bain Capital. Just last week Romney said in one of his speeches:
“While I was governor of Massachusetts we took the unemployment rate from 5.7 percent down to 4.6 percent.”
But was this an accomplishment for which he deserves credit? He could make similar claims about job creation. But does he deserve credit for the jobs created in the state during that time? He seems to think he does. His supporters seem to think he does.
The factor that has not entered into the debates is that state employment and unemployment rates usually reflect what is going on at the national level. With that simple fact a better measure of the effectiveness of a sitting governor is how well the state fares as compared to the nation as a whole.
The graph as the right shows the unemployment rates for Massachusetts and for the nation during Romney’s term as governor. The two rates track very closely. During that period the rate in Massachusetts fell 1.1 percent as Romney has claimed. However the national unemployment rate 1.4 percent from 5.8 percent to 4.4 percent.
At the start of Romney’s term Massachusetts was doing better than the nation as a whole while at the end of his term the state was doing worst than the nation. This does not seem to something that he should be claiming as a great accomplishment without an explanation of why the state has fared worst than the nation.
Posted in Methodolgy Issues, Telling the Full Story
Legislating Science
It seems that in North Carolina that some legislators think that science can be made by rule of law. These members of the legislator don’t like some of the estimates of sea level rise that have been put forward for the rest of this century. So they have decided they they can dictate through the law how those projected levels should be measured. A copy of the draft bill can be found here. Published descriptions can be found here. It seems to me that is what the state hires good employees with solid training in science and data analysis to do.
The key part of the bill is in section 2(e):
The Division of Coastal Management shall be the only State agency authorized to develop rates of sea-level rise and shall do so only at the request of the Commission. These rates shall only be determined using historical data, and these data shall be limited to the time period following the year 1900.
So for planning purposes they have decided that linear extrapolation is the way to go. Nature may not agree with them. And at this point the scientific evidence does not agree with them. Methods of estimation should not be determined by legislative fiat. Rarely do things in science actually behave in a linear fashion. Perhaps they should estimate the temperature in Raleigh next January using linear extrapolation base on data for January and May of this year. If so then someone needs to buy some good air conditioners. Or they can forget about Hurricane preparedness for this summer as it is extremely unlikely that any hurricane will have a linear trajectory toward the North Carolina coast during the first few day after formation.
This is not an issue of global warming. It is an issue of science. If linear extrapolation works, use it. If the exponential extrapolation is the best model then use it. If the models give an estimate that is less than that given by linear extrapolation then use that estimate. But to assume that linear extrapolation is the way to go, and to codify that in the law is foolishness.
Posted in How to Lie with Statisics, Methodolgy Issues