comScore overview
comScore reports on panelists in 37 countries, with panelists residing in more than 165 countries. The reportable countries represent 83% of the world’s machines used to access the Internet. Panelists per reportable country in the December 2008 survey:
| United States | 547,452 | Japan | 9,871 |
| Argentina | 8,796 | Korea (South) | 1,112 |
| Asia Pacific all other | 26,769 | Latin America all other | 6,058 |
| Australia | 14,727 | Malaysia | 7,357 |
| Austria | 2,980 | Mexico | 14,320 |
| Belgium | 6,220 | Middle East-Africa all other | 35,233 |
| Brazil | 65,385 | Netherlands | 12,538 |
| Canada | 37,118 | New Zealand | 2,535 |
| Chile | 4,684 | Norway | 3,241 |
| China | 34,752 | Portugal | 6,771 |
| Colombia | 3,684 | Puerto Rico | 2,296 |
| Denmark | 3,881 | Russian Federation | 6,015 |
| Eastern Europe all other | 26,616 | Singapore | 4,829 |
| Finland | 1,421 | South Africa | 3,223 |
| France | 20,094 | Spain | 17,923 |
| Germany | 34,799 | Sweden | 4,569 |
| Hong Kong | 2,424 | Switzerland | 3,112 |
| India | 28,297 | Taiwan | 1,943 |
| Ireland | 2,523 | United Kingdom | 63,134 |
| Israel | 6,664 | Venezuela | 2,751 |
| Italy | 19,209 | Western Europe all other | 4,211 |
The Streaming Video Codec Report uses comScore's behavioral machine* panel. The panel is made up of more than 1 million personal computers worldwide. The universe and machines in comScore's panel do not represent shared use machines such as library or Internet cafe machines. The following information relates to comScore’s behavioral machine panel.
*Full capture machine panel defined as the number of computers onto which comScore tracking software has been downloaded and successfully installed. Not all full capture machines are eligible for comScore reporting panels in a given month.
Recruitment
comScore uses an array of online recruitment techniques to acquire the members of its panel. These include affiliate programs and partnering with third-party application providers who meet comScore's quality standards. In all cases, panelists opt in through an online registration process that includes a stringent privacy practice.
Data collection
During registration, panelists configure a software agent that allows comScore to observe activity on the machine resulting in a view of the user experience, as opposed to site-centric measurement. This software yields not only the URLs of web pages requested by users but also information such as search strings, products purchased, ads seen, videos streamed, and referral requests. As a result, comScore can capture just about anything exchanged using common Internet protocols.
All of the panelists' Internet activity is captured regardless of type of browser used (Note: The comScore panel includes only computers running a Windows® operating system, not those using other operating systems such as Mac OS or Linux®). Activity is captured regardless of whether an Internet connection is established via a commercial Internet Service Provider (ISP) or an office-hosted LAN.
Data capture and reporting are conducted in adherence to strict, industry-leading privacy protection policies. Data about user identity is stored in an encrypted, access-controlled database. Internet audience and behavior data is reported only in aggregate form.
Weighting and projection
comScore calculates and applies weights to the data accumulated from panelists when aggregating the data for publication. One purpose of these weights is to project measurements of the Internet users in the panel to the much larger number who are not. The other purpose is to eliminate bias that may occur if online recruitment yields disproportionate numbers of people from any segments of Internet users (for example, too many intensive Internet users or too few Internet users from a certain geographic area). Panelists from a segment that is underrepresented get bigger weights and those from a segment that is overrepresented get smaller weights.
Targets for the establishment of these weights are derived from comScore estimates of the Internet universe.
comScore streaming overview and methodology
The following describes the collection and reporting methodology for comScore's streaming media data stream from which the Streaming Video Codec Report is pulled.
Overview
In order to accurately measure the complexities of audio and video streaming content, comScore partnered with industry leaders to develop methods for capturing key pieces of information on every audio and video streaming request viewed by our panelists. Importantly, comScore gathers data on requests sent via streaming servers as well as download methods.
The comScore streaming measurement methodology analyzes the media request itself — not an image beacon, cookie, or server log. As panelists surf the web, comScore dissects each request to identify important information about the URL location of the file and data contained within it. comScore gathers data on the file's duration, mime type, codec, and format, and the software used to play the file.
Defining "streaming media"
The term "streaming media" has different meanings to different user groups. comScore's streaming media reports represent all audio and video consumption on the web using the technologies and methods listed below. comScore refers to audio and video files as "streams" regardless of what technology was used to transmit it.
Collection
comScore’s streaming reports include data captured using two key technical methods commonly referred to as "streaming" and "download."
- Streaming — A live client/server connection is maintained during the transmission, and media files do not remain on the user's computer when the connection ends.
- Download — The user downloads an audio or video to his or her computer and that file is stored on that computer (usually temporarily) for playback. In the vast majority of cases, downloaded audio or video clips are called "progressive downloads" because most media players can play the file as it is being downloaded.
comScore monitors standard ports (and in some cases, nonstandard ports when requested) used by the following protocols:
- RTSP
- RTSP over HTTP
- MMS
- MMS over HTTP
- HTTP
- Ultravox
- PNM
- RTMP
- RTMP over HTTP
comScore parses out the header of known streaming file types to extract data about the file, including duration and whether the file is audio or video. comScore includes all major streaming file formats including Windows Media, Real, MP3, AAC, SWF, and so on.
comScore also measures and collects data downloaded or streamed outside a web browser. This includes users who are streaming files in Windows Media Player, Real Player, iTunes, and so on.
The following data is not included:
- Streams protected using DRM encryption methodologies when comScore is unable to parse out the file due to the encryption; this technology is most commonly used for purchased audio or video files, such as those found in the iTunes Music Store
- Data acquired via peer-to-peer networks such as BitTorrent
- Data transmitted in a nonstandard way, such as using unusual port configurations
- Traffic on Mac OS systems
comScore Video Streaming Codec Report
comScore can decipher the codec used to encode the content at the point of file creation. This is possible when such information is included as part of the file header. comScore rolls up multiple versions and subversions of known codecs into general groupings based on the type of codec (video) and the format (such as FLV or Windows Media).
At present, the Adobe report includes video codecs only (no audio).
| Unique streamers | Number of machines that streamed at least one file during the reporting period for the codec |
| Reach-streamers | Machines that used the given codec expressed as a percentage of all machines that streamed |
| Streams initiated | The number of unique streams for a given codec |
| Percent share streams initiated | The percentage of streams for the given codec compared to the total number of streams in the reporting period |
| Streams per streamer | The number of unique streams per machine for the given codec |
| Total streaming minutes | An aggregate of all minutes associated with the streams for the given codec |
| Minutes per streamer | The aggregate of all minutes associated with the streams for the given codec divided by the number of unique streamers for the given codec |
