Michigan Penetration of broadband Internet by County

Value for Michigan (Percent): 79.0%

Data item: Households with a broadband Internet subscription, percent, 2014-2018

Sources: U.S. Census Bureau, American Community Survey (ACS) and Puerto Rico Community Survey (PRCS), 5-Year Estimates. The PRCS is part of the Census Bureau's ACS, customized for Puerto Rico. Both Surveys are updated every year.

Concept History:

The computer and Internet use questions were added to the ACS in 2013 and were mandated by the 2008 Broadband Improvement Act. Data about computer and Internet use were asked of all occupied housing units.

About

The computer use question asked if anyone in the household owned or used a computer and included four response categories for a desktop or laptop, a smartphone, a tablet or other portable wireless computer, or some other type of computer. Respondents who checked “Yes” for the some other type of computer category are asked to write in descriptions of their computer types. These are mostly used for internal purposes, although some people may write in a type of computer that can be reclassified as a desktop or laptop, a smartphone, or a tablet or other portable wireless computer. The Internet question asked if any member of the household accesses the Internet. “Access” refers to whether or not someone in the household uses or connects to the Internet, regardless of whether or not they pay for the service. For the complete definition, go to ACS subject definitions "Computer and Internet Use."

Limitation of the Data

These questions are not asked for the group quarters population, so would not include data about people living in housing such as dorms, prisons, nursing homes, etc.

Comparability

Data prior to 2013 are not available because 2013 was the first year that these questions were collected using the ACS. Data about computer and Internet use also has been collected sporadically from the Current Population Survey (CPS) since 1984. Both surveys exclude those living in group quarters. However, users should note CPS data is not necessarily comparable to ACS data in several important ways. First, unlike the ACS, some CPS questions are asked at the person level. In addition, the CPS questions and answer categories have changed multiple times over the years. Therefore, comparable data may not be available for certain questions during some years. In addition, some questions may appear to have similar wording as the ACS questions, but may not have been asked of the same type of people.

Source and Accuracy

This fact is based on data collected in the American Community Survey (ACS) and the Puerto Rico Community Survey (PRCS) conducted annually by the U.S. Census Bureau. A sample of over 3.5 million housing unit addresses is interviewed each year over a 12 month period. This Fact (estimate) is based on five years of ACS and PRCS sample data and describes the average value of person, household and housing unit characteristics over this period of collection.

Statistics from all surveys are subject to sampling and nonsampling error. Sampling error is the uncertainty between an estimate based on a sample and the corresponding value that would be obtained if the estimate were based on the entire population (as from a census). Measures of sampling error are provided in the form of margins of error for all estimates included with ACS and PRCS published products. The Census Bureau recommends that data users incorporate this information into their analyses, as sampling error in survey estimates could impact the conclusions drawn from the results. The data for each geographic area are presented together with margins of error at Using margins of error. A more detailed explanation of margins of error and a demonstration of how to use them is provided below.

For more information on sampling and estimation methodology, confidentiality, and sampling and nonsampling errors, please see the Multiyear Accuracy (US) and the Multiyear Accuracy (Puerto Rico) documents at "Documentation - Accuracy of the data."

Margin of Error

As mentioned above, ACS estimates are based on a sample and are subject to sampling error. The margin of error measures the degree of uncertainty caused by sampling error. The margin of error is used with an ACS estimate to construct a confidence interval about the estimate. The interval is formed by adding the margin of error to the estimate (the upper bound) and subtracting the margin of error from the estimate (the lower bound). It is expected with 90 percent confidence that the interval will contain the full population value of the estimate. The following example is for demonstrating purposes only. Suppose the ACS reported that the percentage of people in a state who were 25 years and older with a bachelor's degree was 21.3 percent and that the margin of error associated with this estimate was 0.7 percent. By adding and subtracting the margin of error from the estimate, we calculate the 90-percent confidence interval for this estimate:

21.3% - 0.7% = 20.6% => Lower-bound estimate
21.3% + 0.7% = 22.0% => Upper-bound estimate

Therefore, we can be 90 percent confident that the percent of the population 25 years and older having a bachelor's degree in a state falls somewhere between 20.6 percent and 22.0 percent.

For this Fact, its estimates and margins of error along with percents and percent margins of errors can be found on American Community Survey, Data Profiles-Social Characteristics

More Information

• Questions on the ACS form and why we ask
• ACS methodology
• Computer and Internet Use data from other sources