Method, medium, and apparatus to generate electronic mobile measurement census data

1. An apparatus to correct media measurement data generated by a server, comprising:

an impression server, at an audience measurement entity, to generate electronic mobile measurement data based on network communications received from client devices, the network communications corresponding to media accessed at the client devices;

an assignment modeler to:

select attributes associated with the electronic mobile measurement data to include in a model;

generate the model based on the attributes and a first portion of the electronic mobile measurement data; and

determine a percentage of a second portion of the electronic mobile measurement data that the model correctly associates with corresponding users of the client devices; and

an impression corrector to, when the percentage satisfies a threshold:

determine: (a) when a user operating a client device is a primary user based on the model, and (b) when the user operating the client device is one of a plurality of secondary users based on the model; and

associate demographic information of the identified user with the electronic mobile measurement data to reduce a misattribution error generated by the impression server when generating the electronic mobile measurement data.

2. The apparatus of claim 1 , wherein the associating of the demographic information of the identified user operating the client device with the electronic mobile measurement data is further to reduce processing resources on the client device by identifying the user operating the client device without requiring the user operating the client device to self-identify.

3. The apparatus of claim 1 further including:

a processor to implement the assignment modeler and the impression corrector; and

a memory in circuit with the processor, the memory to store the electronic mobile measurement data and the attributes.

4. The apparatus of claim 1 , wherein the attributes are first attributes and the model is a first model, and when the percentage does not satisfy the threshold:

the assignment modeler to:

select second attributes associated with the electronic mobile measurement data; and

generate a second model based on the second attributes and the first portion of the electronic mobile measurement data; and

the impression corrector to:

identify the user operating the client device as at least one of the primary user or one of the plurality of secondary users based on the second model; and

associate demographic information of the identified user with the electronic mobile measurement data to reduce the misattribution error.

5. The apparatus of claim 1 , wherein the attributes include at least one of demographic information, household information, media information, or information related to the client devices.

6. The apparatus of claim 1 , wherein the threshold is based on at least one of an error tolerance of a customer of the audience measurement entity or a known amount of error in the first portion of the electronic mobile measurement data.

7. The apparatus of claim 1 , wherein when a size of a household of the user operating the client device equals one, the impression corrector is to identify the primary user as the user operating the client device.

8. A tangible computer readable storage medium comprising instructions which, when executed, cause a processor to at least:

generate electronic mobile measurement data based on network communications received from client devices, the network communications corresponding to media accessed at the client devices;

select attributes associated with the electronic mobile measurement data to include in a model;

generate the model based on the attributes and a first portion of the electronic mobile measurement data;

determine a percentage of a second portion of the electronic mobile measurement data that the model correctly associates with corresponding users of the client devices; and

when the percentage satisfies a threshold:

determine: (a) when a user operating a client device is a primary user based on the model, and (b) when the user operating the client device is one of a plurality of secondary users based on the model; and

associate demographic information of the identified user with the electronic mobile measurement data to reduce a misattribution error generated by an impression server when generating the electronic mobile measurement data.

9. The tangible computer readable storage medium as defined in claim 8 , wherein the instructions cause the processor to associate the demographic information of the identified user operating the client device with the electronic mobile measurement data to reduce processing resources on the client device by identifying the user operating the client device without requiring the user operating the client device to self-identify.

10. The tangible computer readable storage medium as defined in claim 8 , wherein the attributes are first attributes, the model is a first model, and when the percentage does not satisfy the threshold, the instructions are to further cause the processor to:

select second attributes associated with the electronic mobile measurement data;

generate a second model based on the second attributes and the first portion of the electronic mobile measurement data;

identify the user operating the client device as at least one of the primary user or one of the plurality of secondary users based on the second model; and

associate the demographic information of the identified user with the electronic mobile measurement data to reduce the misattribution error.

11. The tangible computer readable storage medium as defined in claim 8 , wherein the threshold is based on at least one of an error tolerance of the user identification or a known amount of error in the first portion of the electronic mobile measurement data.

12. The tangible computer readable storage medium as defined in claim 8 , wherein the instructions are to further cause the processor to, when a size of a household of the user operating the client device equals one, identify the primary user as the user operating the client device.

13. The tangible computer readable storage medium as defined in claim 8 , wherein the instructions are to further cause the processor to, when the primary user indicates that the primary user does not share the client device, identify the primary user as the user operating the client device.

14. A method to correct media measurement data generated by a server, comprising:

generating electronic mobile measurement data based on network communications received from client devices, the network communications corresponding to media accessed at the client devices;

selecting, by executing an instruction with a processor, attributes associated with the electronic mobile measurement data to include in a model;

generating, by executing an instruction with the processor, the model based on the attributes and a first portion of the electronic mobile measurement data;

determining, by executing an instruction with the processor, a percentage of a second portion of the electronic mobile measurement data that the model correctly associates with corresponding users of the client devices; and

in response to the percentage satisfying a threshold, executing an instruction with the processor to:

determine: (a) when a user operating a client device is a primary user based on the model, and (b) when the user operating the client device is one of a plurality of secondary users based on the model; and

associate demographic information of the identified user with the electronic mobile measurement data to reduce a misattribution error generated by an impression server when generating the electronic mobile measurement data.

15. The method of claim 14 , wherein executing an instruction with the processor to associate the demographic information of the identified user operating the client device with the electronic mobile measurement data is further to reduce processing resources on the client by identifying the user operating the client device without requiring the user operating the client device to self-identify.

16. The method of claim 14 , wherein the attributes are first attributes, the model is a first model, and further including, in response to the percentage not satisfying the threshold:

selecting second attributes associated with the electronic mobile measurement data;

generating a second model based on the second attributes and the first portion of the electronic mobile measurement data;

identifying the user operating the client device as at least one of the primary user or one of the plurality of secondary users based on the second model; and

associating the demographic information of the identified user with the electronic mobile measurement data to reduce the misattribution error.

17. The method of claim 14 , wherein the attributes include at least one of demographic information, household information, media information, or information related to the client devices.

18. The method of claim 14 , wherein the threshold is based on at least one of an error tolerance of the user identification or a known amount of error in the first portion of the electronic mobile measurement data.

19. The method of claim 14 , further including, in response to a size of a household of the user operating the client device equaling one, identifying the primary user as the user operating the client device.

20. The method of claim 14 , further including, in response to the primary user indicating that the primary user does not share the client device, identifying the primary user as the user operating the client device.