This reference page for the Measuring Fixed Broadband - Twelfth Report provides the validated data set that was used to produce this Report. It includes the detailed results for 13 tests that formed the basis for the Report, in addition to the results of a separate quality metric run to assess which reporting units were available. The validation approaches used for the September-October 2021 data set are discussed in more detail in the Report, the Technical Appendix, and the webpage Open Methodology. Note that the materials below are applicable to data 2021collected in September-October 2021 for the Twelfth Report.
- Statistical Averages
- Data Processing Flow
- Data Dictionary
- Data Cleansing
- SQL Cleanup Scripts
- R Processing Scripts
- Unit Profile
- Excluded Units
- Geocoded Units
- Validated September Data
Files
Statistical Averages
Description
Excel workbook of tabulated test results data. Contains the tabular test data from the validated September-October 2021 test results data. A full range of statistical measures for each test is given.
Why We Provide It
This is the comprehensive analysis of all tests conducted within this study comprising test results contained within the report and additional tests described but not presented in the report. Researchers can use this data to do further analysis of broadband performance in the United States.
Additional Information
See the section on methodology, exclusions and scripts for more information on how the statistical analysis was performed and how the data set was processed.
Download
Statistical-Averages-sept.2021.xlsx
Data Processing Flow Chart
Description
This document provides an overview of how data is processed, from its raw form to the resulting report.
Why We Provide It
This file is provided as documentation to aid researchers and other parties interested in replicating or analyzing the data processing flow.
Download
data-processing-flow-sept2021.pdf
Data Dictionary
Description
This document provides a brief explanation of each field included in the data set, and is provided as documentation on the structure of the data.
Why We Provide It
This file is provided as documentation on the structure of the data. This one file describes the contents on this page.
Download
Data Cleansing
Description
This document outlines the data cleansing processes used to generate the "clean" data set from the "raw" data set.
Why We Provide It
This file is provided as documentation of how the raw data was cleansed to be included in the validated data set.
Download
SQL Cleanup Scripts
Description
SQL scripts associated with the data cleansing process used to prepare raw September 2021 data for statistical analysis.
Why We Provide It
These scripts are provided for full transparency and to facilitate researchers in examining the data. The SQL scripts are necessary to prepare the dataset for R processing operations.
Download
sql-cleanup-scripts-sept2021.tar
R Processing Scripts
Description
R scripts used to prepare the per-unit averages used for statistical processing of the September 2021 data.
Why We Provide It
These scripts are provided for full transparency and to facilitate researchers in examining the data. The SQL scripts process data to to prepare for use in statistical processing.
Download
r-processing-scripts-sept2021.tar
Unit Profile
Description
This document identifies the various details of each test unit, including ISP, technology, service tier, and general location. Each unit represents one volunteer panelist. The unit ID's were random, which served to protect the anonymity of the volunteer panelists. Technical note: This is a large, normalized data set which expands to multiple files. To use this data you will most likely need to import it into a relational databases.
Why We Provide It
This data is provided as reference look up information for the individual units running tests.
Download
Excluded Units
Description
A listing of units excluded from the statistical analysis with brief notations as to why.
Why We Provide It
This list is provided for transparency and identifies those units that were excluded from the validated September-October results used for the statistical analysis. Test results from the excluded units are contained in the Raw Bulk Data files.
Download
Geocoded Units
Description
This spreadsheet identifies the census block in which each unit running test is located. Census block is from 2000 census and is in the FIPS code format. We have used block FIPS codes for blocks that contains more than 1,000 people. For blocks with less than 1,000 people we have aggregated to the next highest level, i.e. tract and used the Tract FIPS code, provided there are more than 1,000 people in the tract. In cases where there are less than 1,000 people in a tract we have aggregated to County level. This level of anonymization is done for privacy purposes so as not to expose PII
Why We Provide It
To identify the general geographic location of the unit running the test.
Download
Validated September Data
Description
This document contains the validated September-October 2021 data set that was used to produce the Report. It includes the detailed results for all 13 tests that were run as part of this study, in addition to the results of a separate quality metric run to assess which reporting units were available. The September-October 2021 data set was validated as discussed in the Report and Technical Appendix; further details can be found in the Methodology section of this page. Technical note: This is a large, normalized data set which expands to multiple files. To use this data you will most likely need to import it into a relational database.
Why We Provide It
The validated data is provided for full transparency and to enable researchers to conduct their own analysis from a common pool of data.