Methods and apparatus to apply household-level weights to household-member level audience measurement data
View Patent ↗Methods and apparatus to apply household-level weights to audience measurement data at a household-member level are disclosed. An example method to determine demographics of populations to measure media audiences of populations includes determining demographics for members of a first household of a sub-population. First demographics of a first member of the first household are different than second demographics of a second member of the first household. The example method includes calculating a first household-level weight for the first household based on a demographics distribution of the sub-population and aggregate demographics of a population. The example method includes applying the first household-level weight to the first demographics of the first member, applying the first household-level weight to the second demographics of the second member, and estimating a demographics distribution of the population to measure a media audience of the population based on the weighted first demographics and the weighted second demographics.
1. method to determine demographics of populations to measure media audiences of the populations, the method comprising:
determining, by executing instructions with a processor, first demographics and second demographics for members of a first household of a sub-population of the populations, the first demographics of a first member of the first household being different than the second demographics of a second member of the first household;
retrieving, by executing instructions with the processor, an initial weight for the first household;
calculating, by executing instructions with the processor, a first household-level weight for the first household based on (a) a demographics distribution of the sub-population and (b) a target value total constructed from aggregate demographics of a population, the population including the sub-population, the calculating including performing an iterative proportional fitting based on the initial weight, the target value total, and a constraint matrix;
applying, by executing instructions with the processor, the first household-level weight to the first demographics of the first member of the first household;
applying, by executing instructions with the processor, the first household-level weight to the second demographics of the second member of the first household; and
reducing an amount of computations executed by the processor without calculating individual-level weights by estimating, by executing instructions with the processor, a demographics distribution of the population to measure a media audience of the population based on the weighted first demographics and the weighted second demographics.
2. The method as defined in claim 1 , further including:
determining, by executing instructions with the processor, third demographics and fourth demographics for members of a second household of the sub-population, the third demographics of a first member of the second household being different than the fourth demographics of a second member of the second household;
calculating, by executing instructions with the processor, a second household-level weight for the second household based on the demographics distribution of the sub-population and the aggregate demographics of the population;
applying, by executing instructions with the processor, the second household-level weight to the third demographics of the first household member of the second household;
applying, by executing instructions with the processor, the second household-level weight to the fourth demographics of the second household member of the second household; and
estimating, by executing instructions with the processor, the demographics distribution of the population based on the weighted third demographics and the weighted fourth demographics.
3. The method as defined in claim 1 , wherein the first demographics of the first member of the first household include a first demographic constraint associated with the first member, and the second demographics of the second member of the first household include a second demographic constraint associated with the second member.
4. The method as defined in claim 1 , further including:
associating a tuning event of the first household with the first household-level weight;
associating the tuning event with the first demographics of the first member and the second demographics of the second member; and
scaling up the first demographics and the second demographics by the first household-level weight to measure the media audience of the population for the tuning event.
5. The method as defined in claim 1 , wherein estimating the demographics distribution includes constructing the constraint matrix based on the demographics of the members of the sub-population.
6. The method as defined in claim 5 , wherein constructing the constraint matrix includes:
assigning the first household as a column of the constraint matrix;
assigning a demographic constraint as a row of the constraint matrix; and
inserting a value indicative of a quantity of members of the first household associated with the demographic constraint as an element in the column and the row of the constraint matrix.
7. The method as defined in claim 1 , wherein reducing the amount of computations executed by the processor includes measuring an audience of the population by utilizing tuning data collected from the first household of the sub-population without collecting consumption data from the members of the sub-population.
8. The method as defined in claim 1 , wherein applying the first household-level weight to the first demographics of the first member includes multiplying the first demographics by the first household-level weight and applying the first household-level weight to the second demographics of the second member includes multiplying the second demographics by the first household-level weight, the weighted first demographics representing a first quantity of members of the population having the first demographics, the weighted second demographics representing a second quantity of members of the population having the second demographics.
9. The method as defined in claim 1 , wherein the processor includes at least a first processor of a first hardware computer system and a second processor of a second hardware computer system.
10. An apparatus to determine demographics of populations to measure media audiences of the populations, the apparatus comprising:
a sub-population determiner to determine first demographics and second demographics for members of a first household of a sub-population of the populations, the first demographics of a first member of the first household being different than the second demographics of a second member of the first household;
a weight calculator to:
retrieve an initial weight for the first household; and
calculate a first household-level weight for the first household based on (a) a demographics distribution of the sub-population and (b) a target value total constructed from aggregate demographics of a population, the population including the sub-population, the weight calculator to calculate the first household-level weight by performing an iterative proportional fitting based on the initial weight, the target value total, and a constraint matrix; and
a population predictor to:
apply the first household-level weight to the first demographics of the first member of the first household;
apply the first household-level weight to the second demographics of the second member of the first household; and
reduce an amount of computations executed by a processor without calculating individual-level weights by estimating a demographics distribution of the population to measure a media audience of the population based on the weighted first demographics and the weighted second demographics, wherein at least one of the sub-population determiner, the weight calculator, or the population predictor include a logic circuit.
11. The apparatus as defined in claim 10 , wherein:
the sub-population determiner is to determine third demographics and fourth demographics for members of a second household of the sub-population, the third demographics of a first member of the second household being different than the fourth demographics of a second member of the second household;
the weight calculator is to calculate a second household-level weight for the second household based on the demographics distribution of the sub-population and aggregate demographics of a population; and
the population predictor is to:
apply the second household-level weight to the third demographics of the first member of the second household;
apply the second household-level weight to the fourth demographics of the second member of the second household; and
estimate the demographics distribution of the population based on the weighted third demographics and the weighted fourth demographics.
12. The apparatus as defined in claim 10 , wherein the first demographics of the first member of the first household includes a first demographic constraint associated with the first member, and the second demographics of the second member of the first household includes a second demographic constraint associated with the second member.
13. The apparatus as defined in claim 10 , wherein the population predictor is to:
associate a tuning event of the first household with the first household-level weight;
associate the tuning event with the first demographics of the first member and the second demographics of the second member; and
scale up the first demographics and the second demographics by the first household-level weight to measure the media audience of the population for the tuning event.
14. The apparatus as defined in claim 10 , wherein the population predictor is to estimate the constraint matrix based on the demographics of the members of the sub-population.
15. The apparatus as defined in claim 14 , wherein, to construct the constraint matrix, the sub-population determiner is to:
assign map the first household as a column of the constraint matrix;
assign a demographic constraint as a row of the constraint matrix; and
insert a value indicative of a quantity of members of the first household associated with the demographic constraint as an element in the column and the row of the constraint matrix.
16. The apparatus as defined in claim 10 , wherein the weight calculator is to reduce an amount of computations executed by the processor by
measuring an audience of the population by utilizing tuning data collected from the first household of the sub-population without collecting consumption data from the members of the sub-population.
17. The apparatus as defined in claim 10 , wherein the population predictor is to:
apply the first household-level weight to the first demographics of the first member by multiplying the first demographics by the first household-level weight, the weighted first demographics representing a first quantity of members of the population having the first demographics; and
apply the first household-level weight to the second demographics of the second member by multiplying the second demographics by the first household-level weight, the weighted second demographics representing a second quantity of members of the population having the second demographics.
18. A tangible computer readable storage medium to determine demographics of populations to measure media audiences of the populations, the tangible computer readable storage medium comprising instructions which, when executed, cause a machine to at least:
determine first demographics and second demographics for members of a first household of a sub-population of the populations, the first demographics of a first member of the first household being different than the second demographics of a second member of the first household;
retrieve an initial weight for the first household;
calculate a first household-level weight for the first household based on (a) a demographics distribution of the sub-population and (b) a target value total constructed from aggregate demographics of a population, the population including the sub-population, the calculating including performing an iterative proportional fitting based on the initial weight, the target value total, and a constraint matrix;
apply the first household-level weight to the first demographics of the first member of the first household;
apply the first household-level weight to the second demographics of the second member of the first household; and
reduce an amount of computations executed by a processor without calculating individual level weights by estimating a demographics distribution of the population to measure a media audience of the population based on the weighted first demographics and the weighted second demographics.
19. The tangible computer readable storage medium as defined in claim 18 , wherein the instructions further cause the machine to:
determine third demographics and fourth demographics for members of a second household of the sub-population, the third demographics of a first member of the second household being different than the fourth demographics of a second member of the second household;
calculate a second household-level weight for the second household based on the demographics distribution of the sub-population and the aggregate demographics of the population;
apply the second household-level weight to the third demographics of the first household member of the second household;
apply the second household-level weight to the fourth demographics of the second household member of the second household; and
estimate the demographics distribution of the population based on the weighted third demographics and the weighted fourth demographics.
20. The tangible computer readable storage medium as defined in claim 18 , wherein the instructions further cause the machine to:
associate a tuning event of the first household with the first household-level weight;
associate the tuning event with the first demographics of the first member and the second demographics; and
scale up the first demographics and the second demographics by the first household-level weight to measure the media audience of the population for the tuning event.
21. The tangible computer readable storage medium as defined in claim 18 , wherein, to estimate the demographics distribution, the instructions further cause the machine to construct the constraint matrix based on the demographics of the members of the sub-population.