Methods and apparatus to de-duplicate impression information

1. A method to de-duplicate media impression information, comprising:

receiving a plurality of first network communications at a first server from client computers, the first network communications indicative of accesses to media at the client computers;

in response to the first network communications, instructing the client computers, via redirect network communications, to send third network communications to a second server of a first database proprietor and fourth network communications to a third server of a second database proprietor, the third and fourth network communications to be indicative of the accesses to the media at the client computers, a portion of the client computers including first cookies for use by the first database proprietor to identify users that are subscribers registered with the first database proprietor, and the portion of the client computers including second cookies for use by the second database proprietor to identify the same users as subscribers registered with the second database proprietor;

determining an overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor on the portions of the client computers based on first impressions including the first cookies and logged by the second server of the first database proprietor in response to the third network communications, and based on second impressions including the second cookies and logged by the third server of the second database proprietor in response to the fourth network communications;

generate, by performing an instruction-based process using a processor, a total impressions count, by summing the first impressions logged by the first database proprietor and the second impressions logged by the second database proprietor, the total impressions count having a duplication error;

determining, using the processor, duplicate reach based on the overlap, the duplicate reach corresponding to the media for which the first and second impressions are logged in association with the first and second cookies of the first and second database proprietors;

determining, using the processor, duplicate impression information based on: (1) the first impressions logged by the first database proprietor, (2) the second impressions logged by the second database proprietor, (3) the duplicate reach of the first and second database proprietors and (4) the overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor; and

applying the duplicate impression information to the total impressions count to correct the instruction-based process of the processor by determining a corrected total impressions count to replace the total impressions count generated by the instruction-based process thereby reducing overcounting of impressions for the media.

2. The method as defined in claim 1 , wherein the determining of the overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor includes determining a number of panelist computers having the first cookie associated with the first database proprietor and the second cookie associated with the second database proprietor.

3. The method as defined in claim 1 , wherein the duplicated impression information includes a number of duplicated audience members.

4. The method as defined in claim 3 , wherein the determining of the corrected total impressions count includes subtracting the number of duplicated audience members from a total of first and second audience members, the first audience members being associated with the first impression information and the second audience members being associated with the second impression information.

5. The method as defined in claim 1 , wherein the overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor on the client computers includes a proportion of overlap.

6. The method as defined in claim 1 , wherein the duplicate impression information includes a number of duplicated impressions, and the determining of the corrected total impressions count includes subtracting the number of duplicated impressions from a sum of the first impressions of the media collected by the first database proprietor and the second impressions of the media collected by the second database proprietor.

7. The method as defined in claim 1 , wherein the duplicated impression information includes a number of duplicated impressions, the method further including determining the number of duplicated impressions by multiplying a number of duplicated audience members by a measured frequency.

8. An apparatus to de-duplicate media impression information, comprising:

a data collector to:

receive a plurality of first network communications at a first server from client computers, the first network communications indicative of accesses to media at the client computers;

in response to the first network communications, instructing the client computers, via redirect network communications, to send third network communications to a second server of a first database proprietor and fourth network communications to a third server of a second database proprietor, the third and fourth network communications to be indicative of the accesses to the media at the client computers, a portion of the client computers including first cookies for use by the first database proprietor to identify users that are subscribers registered with the first database proprietor, and the portion of the client computers including second cookies for use by the second database proprietor to identify the same users as subscribers registered with the second database proprietor;

a duplication estimator to:

determine an overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor on the portions of the client computers based on first impressions including the first cookies and logged by the second server of the first database proprietor in response to the third network communications, and based on second impressions including the second cookies and logged by the third server of the second database proprietor in response to the fourth network communications;

generate, by performing an instruction-based process using a processor, a total impressions count, by summing the first impressions logged by the first database proprietor and the second impressions logged by the second database proprietor, the total impressions count having a duplication error;

determine duplicate impression information based on the first impressions logged by the first database proprietor, the second impressions logged by the second database proprietor, duplicate reach of the first and second database proprietors based on the overlap, the duplicate reach corresponding to the media associated with the first and second impression information; and

a de-duplicator to apply the duplicate impression information to the total impressions count to correct the instruction-based process of the processor by determining a corrected total impressions count to replace the total impressions count generated by the instruction-based process thereby reducing overcounting of impressions for the media, at least one of the data collector, the duplication estimator and the de-duplicator implemented with memory coupled to a processor.

9. The apparatus as defined in claim 8 , further including a panel data processor to determine a number of panelist computers having both the first cookie associated with the first database proprietor and the second cookie associated with the second database proprietor, the duplication estimator to determine the overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor for the client computers based on the number of the panelist computers determined by the panel data processor to have both the first cookie and the second cookie.

10. The apparatus as defined in claim 8 , wherein the duplicated impression information includes a number of duplicated audience members.

11. The apparatus as defined in claim 10 , wherein the de-duplicator is to determine the corrected total impressions count by subtracting the number of duplicated audience members from a total of first and second audience members, the first audience members being associated with the first impression information and the second audience members being associated with the second impression information.

12. The apparatus as defined in claim 8 , wherein the duplicated impression information includes a number of duplicated impressions.

13. The apparatus as defined in claim 8 , wherein the de-duplicator is to determine the corrected total impressions count by subtracting a number of duplicated impressions from a sum of the first impressions collected by the first database proprietor and the second impressions collected by the second database proprietor.

14. The apparatus as defined in claim 8 , wherein the duplication estimator is to determine a number of duplicated impressions by multiplying a number of duplicated audience members by a measured frequency.

15. A non-transitory machine readable storage medium comprising machine readable instructions which, when executed, cause a processor of a first server to at least:

receive a plurality of first network communications from client computers, the first network communications indicative of accesses to media at the client computers;

in response to the first network communications, instructing the client computers, via redirect network communications, to send third network communications to a second server of a first database proprietor and fourth network communications to a third server of a second database proprietor, the third and fourth network communications to be indicative of the accesses to the media at the client computers, a portion of the client computers including first cookies for use by the first database proprietor to identify users that are subscribers registered with the first database proprietor, and the portion of the client computers including second cookies for use by the second database proprietor to identify the same users as subscribers registered with the second database proprietor;

determine an overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor on the portions of the client computers based on first impressions including the first cookies and logged by the second server of the first database proprietor in response to the third network communications, and based on second impressions including the second cookies and logged by the third server of the second database proprietor in response to the fourth network communications;

generate, by performing an instruction-based process using the processor, a total impressions count, by summing the first impressions logged by the first database proprietor and the second impressions logged by the second database proprietor, the total impressions count having a duplication error;

determine duplicate reach based on the overlap, the duplicate reach corresponding to the media for which the first and second impressions are logged in association with the first and second cookies of the first and second database proprietors;

determine duplicate impression information based on the first impressions logged by the first database proprietor, the second impressions logged by the second database proprietor, the duplicate reach of the first and second database proprietors and the overlap; and

applying the duplicate impression information to the total impressions count to correct the instruction-based process of the processor by determining a corrected total impressions count to replace the total impressions count generated by the instruction-based process thereby reducing overcounting of impressions for the media.

16. The storage device as defined in claim 15 , wherein the instructions are to cause the processor to determine the overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor by determining a proportion of panelist computers having the first cookie associated with the first database proprietor and the second cookie associated with the second database proprietor.

17. The storage device as defined in claim 15 , wherein the duplicated impression information includes at least one of a number of duplicated audience members.

18. The storage device as defined in claim 15 , wherein the overlap between the first cookies of the first database proprietor and the second cookies of the second database proprietor on the client computers includes a proportion of overlap.

19. The storage device as defined in claim 15 , wherein the duplicated impression information includes a number of duplicated impressions, and the instructions are to cause the processor to determine the number of duplicated impressions by multiplying a number of duplicated audience members by a measured impression frequency.

20. The storage device as defined in claim 15 , wherein the duplicated impression information includes a number of duplicated impressions, and the instructions are to cause the processor to determine a measured impression frequency from an audience measurement panel.