Systems, methods, and articles of manufacture to measure online audiences
View Patent ↗Methods and apparatus to monitor media content at a content display site are described. An example method includes obtaining panelist data corresponding to a plurality of panelists accessing web pages at measured locations, classifying the panelist data according to demographic information of the panelists, generating a virtual panel based on an estimate of web page access at unmeasured locations, and classifying page view data associated with the unmeasured locations based on data corresponding to the virtual panel.
1. A method to classify audience data, comprising:
obtaining panelist data corresponding to a plurality of panelists accessing web pages at measured locations;
classifying the panelist data according to demographic information of the panelists;
generating, via a processor, a virtual panel based on an estimate of web page accesses at unmeasured locations, the generating including:
determining demographics of an individual unmeasured person that accesses a first one of the web pages from a first one of the unmeasured locations, the panelist data not including media monitoring data for the individual unmeasured person,
identifying an individual panelist that accesses the first one of the web pages at a first one of the measured locations, the individual panelist selected to be representative of the individual unmeasured person based on the individual panelist and the individual unmeasured person having a same demographic characteristic, and
adding a copy of a portion of the panelist data corresponding to the individual panelist to the virtual panel to include the individual unmeasured person in the virtual panel, wherein generating the virtual panel is based on a census-based count of page views, the panelist data, and an estimated universe; and
classifying page view data associated with the unmeasured locations based on the virtual panel including the individual unmeasured person.
2. The method of claim 1 , wherein the virtual panel includes a subset of the plurality of panelists.
3. The method of claim 1 , wherein the generating of the virtual panel includes estimating a total audience.
4. The method of claim 1 , wherein the generating of the virtual panel includes estimating an audience at the unmeasured locations using a negative binomial distribution.
5. The method of claim 1 , wherein the generating of the virtual panel is based on at least one of a demographic target or an activity target.
6. The method of claim 1 , wherein the classifying of the panelist data includes estimating a first portion of an online audience based on the panelist data.
7. The method of claim 6 , wherein the classifying of the page view data includes estimating a second portion of the online audience, the estimates of the first portion and the second portion identifying numbers of unique audience members.
8. The method of claim 7 , further including:
determining a difference between thea census-based count of page views and a number of page views associated with the first and second portions of the page view data; and
estimating additional audience activity not represented by the plurality of panelists and the virtual panel based on the difference.
9. The method of claim 1 , wherein determining the demographics of the individual unmeasured person includes extracting the demographics from a survey response received from the individual unmeasured person.
10. An apparatus, comprising:
a panel data collector to collect panelist data corresponding to a plurality of panelists accessing web pages at measured locations;
a negative binomial distribution calculator to estimate an audience at unmeasured locations based on the plurality of panelists and a universe estimate;
a virtual panel generator to generate a virtual panel based on an estimate of web page accesses at the unmeasured locations, the virtual panel generator to generate the virtual panel by:
determining demographics of an individual unmeasured person that accesses at least one of the web pages from at least one of the unmeasured locations, the panelist data not including media monitoring data for the individual unmeasured person,
identifying an individual panelist that accesses at least one of the web pages at at least one of the measured locations, the individual panelist selected to be representative of the individual unmeasured person based on the individual panelist and the individual unmeasured person having a corresponding demographic profile, and
adding a copy of a portion of the panelist data corresponding to the individual panelist to the virtual panel to include the individual unmeasured person in the virtual panel; and
an audience classifier to classify the panelist data according to demographic information of the panelists and to classify page view data associated with the unmeasured locations based on data corresponding to individual unmeasured persons in the virtual panel, wherein at least one of the panel data collector, the negative binomial distribution calculator, the virtual panel generator, or the audience classifier includes a logic circuit.
11. The apparatus of claim 10 , further including an activity fulfiller to determine online activity based on the plurality of panelists, the audience classifier to classify a third portion of the page view data based on the online activity.
12. The apparatus of claim 10 , further including a census-based data collector to receive the page view data representing a number of page views of a web page from measured and unmeasured locations.
13. An apparatus, comprising:
a panel data collector to collect panelist data corresponding to a plurality of panelists accessing web pages at measured locations;
a virtual panel generator to generate a virtual panel based on an estimate of web page accesses at unmeasured locations;
a survey-based data collector to collect a universe estimate, the virtual panel generator to:
determine demographics of an individual unmeasured person that accesses a first one of the web pages from a first one of the unmeasured locations, the panelist data not including media monitoring data for the individual unmeasured person,
identify an individual panelist that accesses the first one of the web pages at a first one of the measured locations, the individual panelist selected to be representative of the individual unmeasured person based on the individual panelist and the individual unmeasured person having corresponding demographic characteristics, and
add a copy of a portion of the panelist data corresponding to the individual panelist to the virtual panel to include the individual unmeasured person in the virtual panel; and
an audience classifier to classify the panelist data according to demographic information of the panelists and to classify page view data associated with the unmeasured locations based on data corresponding to individual unmeasured persons of the virtual panel.
14. A tangible computer readable storage medium comprising machine readable instructions which, when executed, cause a processor to at least:
obtain panelist data corresponding to a plurality of panelists accessing web pages at measured locations;
classify the panelist data according to demographic information of the panelists;
generate a virtual panel based on an estimate of web page accesses at unmeasured locations, a census-based count of page views, the panelist data, and an estimated universe by:
determining demographics of an individual unmeasured person that accesses a first one of the web pages from at least one of the unmeasured locations, the panelist data not including media monitoring data for the individual unmeasured person,
identifying an individual panelist that accesses the first one of the web pages at at least one of the measured locations, the individual panelist selected to be representative of the individual unmeasured person based on the individual panelist and the individual unmeasured person having corresponding demographic characteristics, and
adding a copy of a portion of the panelist data corresponding to the individual panelist to the virtual panel to include the individual unmeasured person in the virtual panel; and
classify page view data associated with the unmeasured locations based on the virtual panel including the individual unmeasured person.
15. The storage medium of claim 14 , wherein the virtual panel includes a subset of the plurality of panelists.
16. The storage medium of claim 14 , wherein the instructions are to cause the processor to generate the virtual panel by estimating a total audience.
17. The storage medium of claim 14 , wherein the instructions are to cause the processor to generate the virtual panel based on at least one of a demographic target or an activity target.
18. The storage medium of claim 14 , wherein the instructions are to cause the processor to classify the panelist data by estimating a first portion of an online audience based on the panelist data.
19. The storage medium of claim 18 , wherein the instructions are to cause the processor to classify the page view data by estimating a second portion of the online audience, the estimates of the first portion and the second portion identfying numbers of unique audience members.
20. The storage medium of claim 19 , wherein the instructions are further to cause the processor to:
determine a difference between the census-based count of page views and a number of page views associated with the first and second portions of the page view data; and
estimate additional audience activity not represented by the plurality of panelists and the virtual panel based on the difference.
21. A tangible computer readable storage medium comprising machine readable instructions which, when executed, cause a processor to at least:
obtain panelist data corresponding to a plurality of panelists accessing web pages at measured locations;
classify the panelist data according to demographic information of the panelists;
determine demographics of an individual unmeasured person that accesses a first one of the web pages from a first one of unmeasured locations, the panelist data not including media monitoring data for the individual unmeasured person;
identify an individual panelist that accesses the first one of the web pages at a first one of the measured locations, the individual panelist selected to be representative of the individual unmeasured person based on the individual panelist and the individual unmeasured person having a corresponding demographic characteristic;
add a copy of a portion of the panelist data corresponding to the individual panelist to a virtual panel to include the individual unmeasured person in the virtual panel; and
classify page view data associated with the unmeasured locations based on data corresponding to individual unmeasured persons of the virtual panel.