Methods and apparatus to characterize households with media meter data
First Claim
1. A method to calculate a panelist behavior probability, comprising:
- identifying, with a processor, a target set of household categories associated with a target research geography;
when a quantity of panelist households within the target research geography representing the target set of household categories does not satisfy a threshold value required to support statistical significance,increasing an available sample size of households by generating a first subset of categories and a second subset of categories from the target set of household categories;
identifying a first set of households representing the first subset of categories from the target set of household categories and identifying an associated total number of household tuning minutes and a total number of household exposure minutes associated with the first set of households;
for each category in the second subset of categories from the target set of household categories, calculating a household tuning proportion and an exposure proportion, the household tuning proportion for each category in the second subset being based on a ratio of respective category tuning minutes and the total number of household tuning minutes associated with the first set of households, and the exposure proportion for each category in the second subset based on a ratio of respective category exposure minutes and the total number of exposure minutes associated with the first set of households; and
calculating the panelist behavior probability for the combined first and second subset of categories based on the respective exposure proportions and the household tuning proportions.
10 Assignments
0 Petitions
Accused Products
Abstract
Methods, apparatus, systems and articles of manufacture are disclosed to characterize households with media meter data. An example method includes identifying, with a processor, a target set of household categories associated with a target research geography, when a quantity of households within the target research geography representing the target set of household categories does not satisfy a threshold value, generating a first subset of categories and a second subset of categories from the target set of household categories, identifying a first set of households representing the first subset of categories from the target set of household categories and identifying an associated total number of household tuning minutes and a total number of household exposure minutes associated therewith, for each category in the second subset of categories from the target set of household categories, calculating a household tuning proportion and an exposure proportion, the household tuning proportion and exposure proportion based on the total number of household tuning minutes and exposure minutes, respectively, and calculating the panelist behavior probability based on the exposure proportion and the household tuning proportion.
373 Citations
23 Claims
-
1. A method to calculate a panelist behavior probability, comprising:
-
identifying, with a processor, a target set of household categories associated with a target research geography; when a quantity of panelist households within the target research geography representing the target set of household categories does not satisfy a threshold value required to support statistical significance, increasing an available sample size of households by generating a first subset of categories and a second subset of categories from the target set of household categories; identifying a first set of households representing the first subset of categories from the target set of household categories and identifying an associated total number of household tuning minutes and a total number of household exposure minutes associated with the first set of households; for each category in the second subset of categories from the target set of household categories, calculating a household tuning proportion and an exposure proportion, the household tuning proportion for each category in the second subset being based on a ratio of respective category tuning minutes and the total number of household tuning minutes associated with the first set of households, and the exposure proportion for each category in the second subset based on a ratio of respective category exposure minutes and the total number of exposure minutes associated with the first set of households; and calculating the panelist behavior probability for the combined first and second subset of categories based on the respective exposure proportions and the household tuning proportions. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. An apparatus to calculate panelist behavior probability, comprising:
-
a categorizer to identify a target set of household categories associated with a target research geography; a category manager to, when a quantity of panelist households within the target research geography representing the target set of household categories does not satisfy a threshold value sufficient to support statistical significance, increase an available sample size of households by generating a first subset of categories and a second subset of categories from the target set of household categories; a proportion manager to; identify a first set of households representing the first subset of categories from the target set of household categories; identify a total number of household tuning minutes and a total number of household exposure minutes associated associated with the first set of households; calculate a household tuning proportion and an exposure proportion for each category in the second subset of categories from the target set of household categories, the household tuning proportion based on a ratio of category tuning minutes and the total number of household tuning minutes associated with the first set of households, and the exposure proportion based on a ratio of category exposure minutes and the total number of exposure minutes associated with the first set of households; and a distribution engine to calculate the panelist behavior probability for the combined first and second subset of categories based on the respective exposure proportions and the household tuning proportions. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
-
17. A tangible machine readable storage medium comprising instructions that, when executed, cause a machine to at least:
-
identify a target set of household categories associated with a target research geography; when a quantity of panelist households within the target research geography representing the target set of household categories does not satisfy a threshold value sufficient to support statistical significance; increase an available sample size of households by generating a first subset of categories and a second subset of categories from the target set of household categories; identify a first set of households representing the first subset of categories from the target set of household categories and identify an associated total number of household tuning minutes and a total number of household exposure minutes associated with the first set of households; for each category in the second subset of categories from the target set of household categories, calculate a household tuning proportion and an exposure proportion, the household tuning proportion for each category in the second subset based on a ratio of respective category tuning minutes and the total number of household tuning minutes associated with the first set of households, and the exposure proportion for each category in the second subset being based on a ratio of respective category exposure minutes and the total number of household exposure minutes associated with the first set of households; and calculate a panelist behavior probability for the combined first and second subset of categories based on the respective exposure proportions and the household tuning proportions. - View Dependent Claims (18, 19, 20, 21, 22, 23)
-
Specification