Methods and apparatus to impute media consumption behavior
Methods, apparatus, systems and articles of manufacture to impute media consumption behavior are disclosed. An example apparatus includes a localized event engine implemented with hardware or at least one processor to, in response to a difference satisfying a first threshold, determine that a first subset of tuning data exhibits local bias, the difference to be between (i) a first percentage of exposure minutes determined from the tuning data, and (ii) a second percentage of exposure minutes determined from viewing data, determine that a second subset of the viewing data represents heavy viewing in response to (1) a first number of households satisfying a second threshold and (2) a third percentage of exposure minutes satisfying a third threshold, and a localized event selector implemented with hardware or the at least one processor to impute the second subset of the viewing data for the first subset of the tuning data.
1. An apparatus to perform viewership assignment, the apparatus comprising:
a localized event engine implemented with hardware or at least one processor to:
in response to a difference satisfying a first threshold, determine that a first subset of tuning data associated with first panelist households having tuned to first media in a first area exhibits local bias, the difference to be between (i) a first percentage of exposure minutes determined from a second subset of the tuning data associated with exposure of the first panelist households to second media in the first area, the second media different from the first media, and (ii) a second percentage of exposure minutes determined from a first subset of viewing data associated with exposure of second panelist households to the second media in a second area, the second media related to the first media;
determine that a second subset of the viewing data associated with the second panelist households having viewed the first media in the second area represents heavy viewing in response to (1) a first number of households represented in the second subset of the viewing data satisfying a second threshold and (2) a third percentage of exposure minutes determined from the second subset of the viewing data satisfying a third threshold; and
a localized event selector implemented with hardware or the at least one processor to impute the second subset of the viewing data for the first subset of the tuning data when the second subset of the viewing data represents heavy viewing.
2. The apparatus of claim 1 , wherein the localized event engine is to obtain the tuning data from media meters that are unable to identify respective ones of first panelists of the first panelist households, and the localized event engine is to obtain the viewing data from people meters that are able to identify respective ones of second panelists of the second panelist households.
3. The apparatus of claim 1 , wherein the localized event engine is to:
determine a second number of households tuned to the first media in the first area;
determine a fourth percentage of exposure minutes tuned to the first media in the first area, the fourth percentage being a total of number of exposure minutes tuned to the first media with respect to a plurality of exposure minutes tuned to a plurality of media in the first area;
determine that the second number of households satisfies a household number threshold;
determine that the fourth percentage of exposure minutes satisfies an exposure percentage threshold; and
determine that the first subset of the tuning data is heavy tuning data based on the household number threshold and the exposure percentage threshold being satisfied.
4. The apparatus of claim 1 , wherein the viewing data is first viewing data and the second subset of the first viewing data is first heavy viewing data, and the localized event engine is to identify a first subset of second viewing data as second heavy viewing data in response to (1) a second number of households viewing the first media in the first area satisfying a household number threshold and (2) a fourth percentage of exposure minutes viewing the first media in the first area satisfying an exposure percentage threshold; and
the localized event selector is to impute the second heavy viewing data for the first subset of the tuning data when the first subset of the second viewing data is identified as the second heavy viewing data.
5. The apparatus of claim 4 , wherein the localized event selector is to impute at least one of the second subset of the first viewing data or the second subset of the second viewing data for the first subset of the tuning data when the second subset of the first viewing data is identified as the first heavy viewing data and the second subset of the second viewing data is identified as the second heavy viewing data.
6. The apparatus of claim 1 , wherein the viewing data is first viewing data and the second subset of the first viewing data is first heaving viewing data, and the localized event engine is to identify a first subset of second viewing data as second heavy viewing data when (1) a second number of households viewing the first media in a third area satisfies a household number threshold and (2) a fourth percentage of exposure minutes viewing the first media in the third area satisfies an exposure percentage threshold; and
the localized event selector is to impute the second heavy viewing data for the first subset of the tuning data when the first subset of the second viewing data is identified as the second heavy viewing data.
7. The apparatus of claim 1 , wherein the first subset of the tuning data is heavy tuning data, the second subset of the viewing data is heavy viewing data, and further including:
a localized event probability calculator implemented with hardware or the at least one processor to:
calculate first probability values associated with a subset of the first panelist households associated with the heavy tuning data, the first probability values indicative of a likelihood that the subset of the first panelist households include first panelists that are likely viewers of the first media;
calculate second probability values associated with a subset of the second panelist households associated with the heavy viewing data, the second probability values indicative of a likelihood that the subset of the second panelist households include second panelists that are likely viewers of the first media; and
the localized event selector is to associate a respective one of the subset of the first panelists with a respective one of the subset of the second panelists based on the first probability values and the second probability values.
8. A non-transitory computer readable storage medium comprising instructions that, when executed, cause a machine to at least:
in response to a difference satisfying a first threshold, determine that a first subset of tuning data associated with first panelist households having tuned to first media in a first area exhibits local bias, the difference to be between (i) a first percentage of exposure minutes determined from a second subset of the tuning data associated with exposure of the first panelist households to second media in the first area, the second media different from the first media, and (ii) a second percentage of exposure minutes determined from a first subset of viewing data associated with exposure of second panelist households to the second media in a second area, the second media related to the first media;
determine that a second subset of the viewing data associated with the second panelist households having viewed the first media in the second area represents heavy viewing in response to (1) a first number of households represented in the second subset of the viewing data satisfying a second threshold and (2) a third percentage of exposure minutes determined from the second subset of the viewing data satisfying a third threshold; and
impute the second subset of the viewing data for the first subset of the tuning data when the second subset of the viewing data represents heavy viewing.
9. The non-transitory computer readable storage medium comprising of claim 8 , wherein the instructions, when executed, cause the machine to obtain the tuning data from media meters that are unable to identify respective ones of first panelists of the first panelist households, and obtain the viewing data from people meters that are able to identify respective ones of second panelists of the second panelist households.
10. The non-transitory computer readable storage medium comprising of claim 8 , wherein the instructions, when executed, cause the machine to:
determine a second number of households tuned to the first media in the first area;
determine a fourth percentage of exposure minutes tuned to the first media in the first area, the fourth percentage being a total of number of exposure minutes tuned to the first media with respect to a plurality of exposure minutes tuned to a plurality of media in the first area;
determine that the second number of households satisfies a household number threshold;
determine that the fourth percentage of exposure minutes satisfies an exposure percentage threshold; and
determine that the first subset of the tuning data is heavy tuning data based on the household number threshold and the exposure percentage threshold being satisfied.
11. The non-transitory computer readable storage medium comprising of claim 8 , wherein the viewing data is first viewing data and the second subset of the first viewing data is first heavy viewing data, and instructions, when executed, cause the machine to:
identify a first subset of second viewing data as second heavy viewing data in response to (1) a second number of households viewing the first media in the first area satisfying a household number threshold and (2) a fourth percentage of exposure minutes viewing the first media in the first area satisfying an exposure percentage threshold; and
impute the second heavy viewing data for the first subset of the tuning data when the first subset of the second viewing data is identified as the second heavy viewing data.
12. The non-transitory computer readable storage medium comprising of claim 11 , wherein the instructions, when executed, cause the machine to impute at least one of the second subset of the first viewing data or the second subset of the second viewing data for the first subset of the tuning data when the second subset of the first viewing data is identified as the first heavy viewing data and the second subset of the second viewing data is identified as the second heavy viewing data.
13. The non-transitory computer readable storage medium comprising of claim 8 , wherein the viewing data is first viewing data and the second subset of the first viewing data is first heaving viewing data, and the instructions, when executed, cause the machine to:
identify a first subset of second viewing data as second heavy viewing data when (1) a second number of households viewing the first media in a third area satisfies a household number threshold and (2) a fourth percentage of exposure minutes viewing the first media in the third area satisfies an exposure percentage threshold; and
impute the second heavy viewing data for the first subset of the tuning data when the first subset of the second viewing data is identified as the second heavy viewing data.
14. The non-transitory computer readable storage medium comprising of claim 8 , wherein the first subset of the tuning data is heavy tuning data, the second subset of the viewing data is heavy viewing data, and the instructions, when executed, cause the machine to:
calculate first probability values associated with a subset of the first panelist households associated with the heavy tuning data, the first probability values indicative of a likelihood that the subset of the first panelist households include first panelists that are likely viewers of the first media;
calculate second probability values associated with a subset of the second panelist households associated with the heavy viewing data, the second probability values indicative of a likelihood that the subset of the second panelist households include second panelists that are likely viewers of the first media; and
associate a respective one of the subset of the first panelists with a respective one of the subset of the second panelists based on the first probability values and the second probability values.
15. A method to perform viewership assignment, the method comprising:
in response to a difference satisfying a first threshold, determining, by executing an instruction with a processor, that a first subset of tuning data associated with first panelist households having tuned to first media in a first area exhibits local bias, the difference being between (i) a first percentage of exposure minutes determined from a second subset of the tuning data associated with exposure of the first panelist households to second media in the first area, the second media different from the first media, and (ii) a second percentage of exposure minutes determined from a first subset of viewing data associated with exposure of second panelist households to the second media in a second area, the second media related to the first media;
determining, by executing an instruction with the processor, that a second subset of the viewing data associated with the second panelist households having viewed the first media in the second area represents heavy viewing in response to (1) a first number of households represented in the second subset of the viewing data satisfying a second threshold and (2) a third percentage of exposure minutes determined from the second subset of the viewing data satisfying a third threshold; and
imputing, by executing an instruction with the processor, the second subset of the viewing data for the first subset of the tuning data when the second subset of the viewing data represents heavy viewing.
16. The method of claim 15 , further including obtaining the tuning data from media meters that are unable to identify respective ones of first panelists of the first panelist households, and obtaining the viewing data from people meters that are able to identify respective ones of second panelists of the second panelist households.
17. The method of claim 15 , further including:
determining a second number of households tuned to the first media in the first area;
determining a fourth percentage of exposure minutes tuned to the first media in the first area, the fourth percentage being a total of number of exposure minutes tuned to the first media with respect to a plurality of exposure minutes tuned to a plurality of media in the first area;
determining that the second number of households satisfies a household number threshold;
determining that the fourth percentage of exposure minutes satisfies an exposure percentage threshold; and
determining that the first subset of the tuning data is heavy tuning data based on the household number threshold and the exposure percentage threshold being satisfied.
18. The method of claim 15 , wherein the viewing data is first viewing data and the second subset of the first viewing data is first heavy viewing data, and further including:
identifying a first subset of second viewing data as second heavy viewing data in response to (1) a second number of households viewing the first media in the first area satisfying a household number threshold and (2) a fourth percentage of exposure minutes viewing the first media in the first area satisfying an exposure percentage threshold; and
imputing the second heavy viewing data for the first subset of the tuning data when the first subset of the second viewing data is identified as the second heavy viewing data.
19. The method of claim 18 , further including imputing at least one of the second subset of the first viewing data or the second subset of the second viewing data for the first subset of the tuning data when the second subset of the first viewing data is identified as the first heavy viewing data and the second subset of the second viewing data is identified as the second heavy viewing data.
20. The method of claim 15 , wherein the viewing data is first viewing data and the second subset of the first viewing data is first heaving viewing data, and further including:
identifying a first subset of second viewing data as second heavy viewing data when (1) a second number of households viewing the first media in a third area satisfies a household number threshold and (2) a fourth percentage of exposure minutes viewing the first media in the third area satisfies an exposure percentage threshold; and
imputing the second heavy viewing data for the first subset of the tuning data when the first subset of the second viewing data is identified as the second heavy viewing data.
21. The method of claim 15 , wherein the first subset of the tuning data is heavy tuning data, the second subset of the viewing data is heavy viewing data, and further including:
calculating first probability values associated with a subset of the first panelist households associated with the heavy tuning data, the first probability values indicative of a likelihood that the subset of the first panelist households include first panelists that are likely viewers of the first media;
calculating second probability values associated with a subset of the second panelist households associated with the heavy viewing data, the second probability values indicative of a likelihood that the subset of the second panelist households include second panelists that are likely viewers of the first media; and
associating a respective one of the subset of the first panelists with a respective one of the subset of the second panelists based on the first probability values and the second probability values.