Wednesday, March 20, 2013 3:15 am
How low-wage workforce poll was conducted
By The Associated Press
The AP-NORC Center poll of lower-wage workers was conducted Aug. 1 to Sept. 6, 2012, among a random sample of 1,606 workers. The nationally representative random sample of workers included people age 18 to 74, who were employed full-time or part-time or were on a temporary layoff of six months or less. Respondents were not self-employed and reported earning an annual equivalent of $35,000 or less.
Online interviews were conducted with 1,250 respondents by GfK using the GfK KnowledgePanel. Phone interviews with 356 workers reached via landline or cellular telephone were conducted by NORC at the University of Chicago. No more than 1 time in 20 should chance variations in the sample cause the results to vary by more than plus or minus 2.9 percentage points.
GfK's KnowledgePanel was recruited using traditional telephone and mail polling methods. People chosen for the study who had no Internet access were given it for free. With a probability basis and coverage of people who otherwise couldn't access the Internet, online surveys conducted using KnowledgePanel are nationally representative.
As is done routinely in surveys, results were weighted, or adjusted, to ensure that responses accurately reflect the population's makeup by factors such as age, sex, education, region and race. In addition, the weighting took into account patterns of phone use.
The survey of employers was conducted Nov. 12, 2102, to Jan. 31, 2013, among a random sample of 1,487 employers. Qualified employers included those with at least three employees and at least one lower-wage worker - defined as an employee making the full-time equivalent of $35,000 per year or less.
NORC selected the sample from the Dun and Bradstreet business database. The sample was stratified by business size and sector.
All interviews were conducted by NORC at the University of Chicago. A total of 826 interviews were completed over the telephone and 661 completed the survey online. Respondents were offered a $50 monetary incentive for participating.
The sample was weighted to adjust for unequal sampling probabilities across strata, unknown eligibility, nonresponse and regional distribution. The base weight reflects the selection probabilities under the sample design. The base weight was then adjusted to compensate for records that are unknown eligibility and for nonresponse. A final adjustment was applied to the weight to adjust the regional distribution of the sample.
No more than 1 time in 20 should chance variations in the sample cause the results to vary by more than plus or minus 4.5 percentage points for the survey of employers. There are other sources of potential error in polls, including the wording and order of questions.
The project on America's lower-wage workers was conducted with funding from the Joyce Foundation, the Hitachi Foundation and NORC at the University of Chicago. The Joyce Foundation's Employment program works to improve workforce development and education systems to help underprepared adults learn important skills, earn credentials and pursue jobs in their communities. The Hitachi Foundation aims to expand business practices that measurably improve economic opportunities for low-wealth individuals and enhance long-term business value.
The questions and results are available at http://www.apnorc.org.