4 ms·

No, it’s more complicated than that. The BLS does a robust survey. They also track discouraged workers who have left the labor force. https://www.bls.gov/home.h

No, it’s more complicated than that. The BLS does a robust survey. They also track discouraged workers who have left the labor force. https://www.bls.gov/home.htm https://www.bls.gov/home.htm

salawat 7y ago
I've been thumbing through their survey methodologies, and am actually hard pressed to find anyone in any social circle who has directly or even indirectly to the best of their knowledge participated in these surveys. Specifically the CPS which tracks the unemployment vs. employment rates. Given that it takes the form of a monthly survey, I'd expect more people to actually remember having done so. Now I could understand inferring employment via tax documents cross referenced with unemployment papers, but to my understanding that level of information sharing is not generally done. Where do the statistics come from? Here's the answer from BLS's Current Population Survey FAQ at w.r.t. where the data comes from. https://www.bls.gov/cps/faq.htm#Ques2 https://www.bls.gov/cps/faq.htm#Ques2 >Because unemployment insurance records relate only to people who have applied for such benefits, and since it is impractical to count every unemployed person each month, the government conducts a monthly survey called the Current Population Survey (CPS) to measure the extent of unemployment in the country. The CPS has been conducted in the United States every month since 1940, when it began as a Work Projects Administration program. In 1942, the U.S. Census Bureau took over responsibility for the CPS. The survey has been expanded and modified several times since then. So what do the statistics being gathered actually represent and where do they come from? More digging required... Ah, here it is. Survey methodology. ho! https://www.census.gov/programs-surveys/cps/technical-documentation/methodology.html https://www.census.gov/programs-surveys/cps/technical-docume... I'm not super Survey methodology savvy, (more familiar with anti-patterns for it honestly, and figuring out when the Map starts diverging from the landscape), but in terms of what it is actually measuring, and how much divergence is introduced between reality and the convenient model inferred from the survey's statistical finger waggling... I leave that to the more Stats inclined among us. It seems an okay methodology, but it doesn't strike me in terms of being the most accurate measurement, merely getting enough data that someone feels that the divergences can be safely tucked under the rug. I will say, the folks putting this together seem to have done a great job documenting it. I wish my developers could be relied upon to do as such.

No, it’s more complicated than that. The BLS does a robust survey. They also track discouraged workers who have left the labor force. https://www.bls.gov/home.htm https://www.bls.gov/home.htm

salawat 7y ago

I've been thumbing through their survey methodologies, and am actually hard pressed to find anyone in any social circle who has directly or even indirectly to the best of their knowledge participated in these surveys. Specifically the CPS which tracks the unemployment vs. employment rates. Given that it takes the form of a monthly survey, I'd expect more people to actually remember having done so. Now I could understand inferring employment via tax documents cross referenced with unemployment papers, but to my understanding that level of information sharing is not generally done. Where do the statistics come from? Here's the answer from BLS's Current Population Survey FAQ at w.r.t. where the data comes from. https://www.bls.gov/cps/faq.htm#Ques2 https://www.bls.gov/cps/faq.htm#Ques2 >Because unemployment insurance records relate only to people who have applied for such benefits, and since it is impractical to count every unemployed person each month, the government conducts a monthly survey called the Current Population Survey (CPS) to measure the extent of unemployment in the country. The CPS has been conducted in the United States every month since 1940, when it began as a Work Projects Administration program. In 1942, the U.S. Census Bureau took over responsibility for the CPS. The survey has been expanded and modified several times since then. So what do the statistics being gathered actually represent and where do they come from? More digging required... Ah, here it is. Survey methodology. ho! https://www.census.gov/programs-surveys/cps/technical-documentation/methodology.html https://www.census.gov/programs-surveys/cps/technical-docume... I'm not super Survey methodology savvy, (more familiar with anti-patterns for it honestly, and figuring out when the Map starts diverging from the landscape), but in terms of what it is actually measuring, and how much divergence is introduced between reality and the convenient model inferred from the survey's statistical finger waggling... I leave that to the more Stats inclined among us. It seems an okay methodology, but it doesn't strike me in terms of being the most accurate measurement, merely getting enough data that someone feels that the divergences can be safely tucked under the rug. I will say, the folks putting this together seem to have done a great job documenting it. I wish my developers could be relied upon to do as such.