Methods and apparatus to impute media consumption behavior
Methods, apparatus, systems and articles of manufacture to impute media consumption behavior are disclosed. An example method includes calculating, by executing an instruction with a processor, first viewing probabilities for tuning panelists in a tuning household. The example method also includes calculating, by executing an instruction with the processor, second viewing probabilities for viewing panelists in a plurality of learning households having a same household size as the tuning household. The example method also includes identifying, by executing an instruction with the processor, a matching one of the plurality of learning households based on a difference between the first viewing probabilities and respective ones of the second viewing probabilities.
1. An apparatus to impute media consumption behavior, the apparatus comprising:
a localized event engine implemented with hardware or at least one processor to:
obtain first tuning data, which is associated with first panelists in first households having tuned to first media in a first area, from media meters that are unable to identify respective ones of the first panelists;
identify a subset of the first tuning data as heavy tuning data when (1) a number of the first households tuned to the first media in the first area satisfies a first threshold and (2) a first percentage of exposure minutes tuned to the first media in the first area satisfies a second threshold; and
determine that the heavy tuning data represents a local bias when a difference between a second percentage of exposure minutes associated with second panelists having viewed second media in a second area and a third percentage of exposure minutes associated with third panelists having tuned to the second media in the first area satisfies a third threshold, the second media being comparable media relative to the first media;
a probability engine implemented with hardware or at least one processor to:
calculate first viewing probabilities for the first panelists in a first one of the first households when the heavy tuning data is determined to represent the local bias; and
calculate second viewing probabilities for fourth panelists in a plurality of second households having household sizes corresponding to the first one of the first households, the fourth panelists associated with heavy viewing data, the heavy viewing data obtained from people meters that are able to identify respective ones of the fourth panelists; and
a most likely viewer engine implemented with hardware or at least one processor to:
identify a matching one of the plurality of second households that matches the first one of the first households based on a difference between the first viewing probabilities and respective ones of the second viewing probabilities; and
impute the heavy viewing data associated with the fourth panelists for tuning data included in the heavy tuning data when the matching one is identified.
2. The apparatus as defined in claim 1 , further including:
a collection engine to determine if a first household member of the first one of the first households satisfies an age threshold, wherein
the probability engine is to calculate third viewing probabilities for fifth panelists in a subset of the plurality of second households when the first household member of the first one of the first households satisfies the age threshold, the subset of the plurality of second households corresponding to households that include at least one household member who satisfies the age threshold.
3. The apparatus as defined in claim 1 , further including:
a comparable media identifier to identify the second media, the second media being comparable media to the first media; and
a comparable media percentage calculator to:
calculate the second percentage of exposure minutes, the second percentage being a first comparable media percentage associated with exposure to the comparable media in the second area, the second percentage corresponding to a first total number of exposure minutes associated with viewing the second media with respect to a plurality of exposure minutes associated with viewing a plurality of media associated with the second area;
calculate the third percentage of exposure minutes, the third percentage being a second comparable media percentage associated with exposure to the comparable media in the first area, the third percentage corresponding to a second total number of exposure minutes tuned to the second media with respect to a plurality of exposure minutes tuned to a plurality of media associated with the first area; and
calculate the difference between the first comparable media percentage and the second comparable media percentage.
4. A method to impute media consumption behavior, the method comprising:
obtaining, by executing an instruction with a processor, first tuning data, which is associated with first panelists in first households having tuned to first media in a first area, from media meters that are unable to identify respective ones of the first panelists;
identifying, by executing an instruction with the processor, a subset of the first tuning data as heavy tuning data when (1) a number of the first households tuned to the first media in the first area satisfies a first threshold and (2) a first percentage of exposure minutes tuned to the first media in the first area satisfies a second threshold;
determining, by executing an instruction with the processor, that the heavy tuning data represents a local bias when a second percentage of exposure minutes associated with second panelists having viewed second media in a second area and a third percentage of exposure minutes associated with third panelists having tuned to the second media in the first area satisfies a third threshold, the second media being comparable media relative to the first media;
calculating, by executing an instruction with the processor, first viewing probabilities for the first panelists in a first one of the first households when the heavy tuning data is determined to represent the local bias;
calculating, by executing an instruction with the processor, second viewing probabilities for fourth panelists in a plurality of second households having household sizes corresponding to the first one of the first households, the fourth panelists associated with heavy viewing data, the heavy viewing data obtained from people meters that are able to identify respective ones of the second panelists;
identifying, by executing an instruction with the processor, a matching one of the plurality of second households that matches the first one of the first households based on a difference between the first viewing probabilities and respective ones of the second viewing probabilities; and
imputing, by executing an instruction with the processor, the heavy viewing data associated with the fourth panelists for tuning data included in the heavy tuning data when the matching one is identified.
5. The method as defined in claim 4 , wherein the first households and the plurality of second households each have a respective household size of two members.
6. The method as defined in claim 4 , further including:
determining if a first household member of the first one of the first households satisfies an age threshold; and
in response to determining that the first household member of the first one of the first households satisfies the age threshold, calculating third viewing probabilities for fifth panelists in a subset of the plurality of second households, the subset of the plurality of second households corresponding to households that include at least one household member who satisfies the age threshold.
7. A non-transitory computer readable storage medium comprising instructions that, when executed, cause a machine to at least:
obtain first tuning data, which is associated with first panelists in first households having tuned to first media in a first area, from media meters that are unable to identify respective ones of the first panelists;
identify a subset of the first tuning data as heavy tuning data when (1) a number of the first households tuned to the first media in the first area satisfies a first threshold and (2) a first percentage of exposure minutes tuned to the first media in the first area satisfies a second threshold;
determine that the heavy tuning data represents a local bias when a second percentage of exposure minutes associated with second panelists having viewed second media in a second area and a third percentage of exposure minutes associated with third panelists having tuned to the second media in the first area satisfies a third threshold, the second media being comparable media relative to the first media;
calculate first viewing probabilities for the first panelists in a first one of the first households when the heavy tuning data is determined to represent the local bias;
calculate second viewing probabilities for fourth panelists in a plurality of second households having household sizes corresponding to the first one of the first households, the fourth panelists associated with heavy viewing data, the heavy viewing data obtained from people meters that are able to identify respective ones of the second panelists;
identify a matching one of the plurality of second households that matches the first one of the first households based on a difference between the first viewing probabilities and respective ones of the second viewing probabilities; and
impute the heavy viewing data associated with the fourth panelists for tuning data included in the heavy tuning data when the matching one is identified.
8. The non-transitory computer readable storage medium as defined in claim 7 , wherein the first households and the plurality of second households each have a respective household size of two members.
9. The non-transitory computer readable storage medium as defined in claim 8 , wherein the instructions, when executed, cause the machine to:
determine if a first household member of the first one of the first households satisfies an age threshold; and
calculate third viewing probabilities for fifth panelists in a subset of the plurality of second households when the first household member of the first one of the first households satisfies the age threshold, the subset of the plurality of second households corresponding to households that include at least one household member who satisfies the age threshold.
10. The apparatus as defined in claim 1 , further including a heavy exposure classifier to:
determine that the number of the first households tuned to the first media in the first area satisfies the first threshold, the first threshold being a household number threshold, wherein respective ones of the first households have respective household sizes of two members; and
determine that the first percentage of exposure minutes tuned to the first media in the first area satisfies the second threshold, the first percentage being a total 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, the second threshold being an exposure percentage threshold.
11. The apparatus as defined in claim 1 , further including a localized event donor data identifier to identify viewing data, which is associated with the fourth panelists in the plurality of second households, as the heavy viewing data when (1) a number of the plurality of second households viewing third media in a third area satisfies the first threshold and (2) a percentage of viewing minutes viewing the third media in the third area satisfies the second threshold.
12. The apparatus as defined in claim 1 , further including a collection engine to collect data associated with panelists in the first area and the second area from the media meters and the people meters.
13. The apparatus as defined in claim 1 , wherein ones of the media meters are arranged to monitor media presentation devices associated with the first panelists.
14. The method as defined in claim 4 , further including:
determining that the number of the first households tuned to the first media in the first area satisfies the first threshold, the first threshold being a household number threshold, wherein respective ones of the first households have respective household sizes of two members; and
determining that the first percentage of exposure minutes tuned to the first media in the first area satisfies the second threshold, the first percentage being a total 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, the second threshold being an exposure percentage threshold.
15. The method of claim 4 , further including identifying viewing data, which is associated with the fourth panelists in the plurality of second households, as the heavy viewing data when (1) a number of the plurality of second households viewing third media in a third area satisfies the first threshold and (2) a percentage of viewing minutes viewing the third media in the third area satisfies the second threshold.
16. The method as defined in claim 4 , wherein the first tuning data includes a number of minutes credited to a media presentation device tuned to the first media in the first households.
17. The method as defined in claim 4 , wherein ones of the media meters are arranged to monitor media presentation devices associated with the first panelists.
18. The non-transitory computer readable storage medium as defined in claim 7 , further including instructions that, when executed, cause the machine to:
determine that the number of the first households tuned to the first media in the first area satisfies the first threshold, the first threshold being a household number threshold, wherein respective ones of the first households have respective household sizes of two members; and
determine that the first percentage of exposure minutes tuned to the first media in the first area satisfies the second threshold, the first percentage being a total 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, the second threshold being an exposure percentage threshold.
19. The non-transitory computer readable storage medium as defined in claim 7 , further including instructions that, when executed, cause the machine to identify viewing data, which is associated with the fourth panelists in the plurality of second households, as the heavy viewing data when (1) a number of the plurality of second households viewing third media in a third area satisfies the first threshold and (2) a percentage of viewing minutes viewing the third media in the third area satisfies the second threshold.
20. The non-transitory computer readable storage medium as defined in claim 7 , wherein the first tuning data includes a number of minutes credited to a media presentation device tuned to the first media in the first households.